Questions and Answers on IDE drives from the Ontrack BBS ------------------------------------------------------------------------ QUESTION: I NEED TO KNOW IF A PERSON CAN GET THE DMPARMS DRIVERS FOR THE CONNERS CP-3184 80MEG HARD DRIVE OR IF THE MOST CURRENT DISK MANAGER WILL HAVE THEM ANSWER: The most current version of Disk Manager (v4.20) directly supports the Conner CP-3184. Versions of Disk Manager prior to v4.20 didn't know how to handle the translation of a IDE drive, so while you COULD install one, you had to jump though a lot of hoops along the way. As of v4.20, this has become an automatic process. ------------------------------------------------------------------------- QUESTION: I am having trouble installing a minscribe 8051a with IDE controller. The system seems not to see the drive - I get "drive fails recal or test drive ready" error when running DM. I have a floppy controller in the AT also, and am wondering if there is cointention between the two controllers (Floppy controller has C: drive port also.) Any help appreciated. ANSWER: You are right. There will be a conflict between the two HD controllers if they are set to the same address. If you have a way to set the HD portion of the floppy controller to a different address, you may be able to get rid of the conflict. ------------------------------------------------------------------------- QUESTION: I looked at bulletin number 6 and in this bulletin it made reference to the file DMPARMS.OCS ! I am not sure as to where this file lives or how to create it. Any help would be greatly appricated. DM-ver-4.02 s/n IMP04952742 drive cp-3204 !! Thanks ANSWER: From the "IMP" in your serial number, I see that this is an "Imprimis only" version of Disk Manager. The Imprimis version of Disk Manager is for Imprimis drives only, and therefore does not have the file DMPARMS.OCS. You will not be able to install a Conner drive with this Imprimis only version of Disk Manager. You should contact our sales department at (800)752-1333 and purchase the "generic" version of Disk Manager that can install any drive. ------------------------------------------------------------------------- QUESTION: I have recently purchased an IMPRIMIS 143MB drive model# 94354-160 3.5" hard drive and Miniscribe AT Disk Controller (WD37C65B-PL controller chip) for my Fujikama 20Mhz 386. The drive came with a version of Disk Manager for IMPRIMIS drives. I can format the drive under DOS to have a 33MB and 109MB partition and it works just great however OS/2 seems to see the drive a a 76MB drive not 143MB. Is there either a version of SWBIOS, device driver, or BIOS patch that I can use to get OS/2 to see all 143MB of this drive. I would even settle for for 4 33MB partitions and 1 11 MB partition if this would work. Any help would be appreciated. ANSWER: At this time there is not a version of Disk Manager that supports OS/2. Under OS/2, you will be limited to the standard portion of this drive (whatever CMOS says the parameters are). If you have a user definable CMOS drive type, you could try using that, but Disk Manager can't help you in this situation. ------------------------------------------------------------------------- QUESTION: I have Disk Manager 4.03 and have gotten rather dependent on it. Recently we purchase 9 machines with Quauntum DE drives and found that OnTrack didn't support these. Is there a newer version or plans for a newer version which will include Quauntum support? ANSWER: I don't have any information on the Quantun ID42. Your message said it was a DE drive, did you mean IDE? If so, you should take a look at bulletin number 6 concerning installation of IDE drives. If you know the parameters of the drive, the same procedure should work with this drive. Remember that you don't want to initialize an IDE drive, just configure and partition it. If you could leave me more info on this drive, I would stand a better chance of helping you with it. The drive information that goes into Disk Manager is directly from the drive manufacturers. So far Quantum has not provided us with any information on this drive. ------------------------------------------------------------------------- QUESTION: Hi there... I'm with a VAR in the Washington, DC area. I'm the software guy, but sometimes get pressed into service installing systems. A few days ago was having a LOT of trouble installing a Conner 3204 HD in an AST/P286 to its full capacity (205 megs), which probably comes as no great surprise. What was a PLEASANT surprise though, was how well things went with DM after reading bulletin number 6 from this BBS, and the DM help screens. SWBIOS is quite a gem... . We are seeing a lot of interest, suddenly, in the Conner/IDE drives. Please continue to post messages and files with details on DM and IDE drives. Specifically, I would appreciate more information on SWBIOS.COM. . So we will certainly be bundling DM with our systems. . In the meantime, one thing concerns me... I read a msg you sent to someone in which you stated one "should never attempt to low level format (initialize) a IDE drive through any current version of DM".... . Unfortunately, I read this AFTER doing just that. YIKES. What danger lurks? Seriously, what are the consequences of initializing IDE drive in DM? And what if anything should I do now? . Thanks for the info. Keep it coming. ANSWER: My statement about not ever low-level initializing an IDE drive was probably a little on the conservative side. It's true that all IDE drives are low-level formatted at the factory and do not require low-leveling by Disk Manager. If you DO low-level format an IDE drive for some reason, the ramifications of that are dependant on whose drive it is. The Conner drives protect themselves from this well since they are embedded servo drives. There are really no known ill effects of performing a low-level format on a Conner drive. On the other hand, there are some drives out there that will lose their internal defect list if you low-level format them. There are also some drives that the jury is still out on what will happen in this situation. The IDE style drive is new enough that we are all still learning (even the manufacturers sometimes) about what really goes on inside the little black box. As I find out more infomation about specific IDE drives, I will attempt to get that information in some form onto the BBS and try to share the knowledge. ------------------------------------------------------------------------- QUESTION: I am trying to install a Conner CP3204 drive in a real IBM AT. The ROM dayes in the AT are 1981 1985. I am using a floppy hard controller that comes with the drive Model NCL 530. Using IBM DOS 4.01. I am using Disk Manger Ver. 4.03. This drive is not listed under the Conner drives on the disk. I try to format the drive in manual mode one paration 2 MEGS. The next paration 209 MEGS. The drive is 1366 cylinders, 8 Heads, by 38 sectors per track. Everything seems to go fine until I have to re-boot the computer. Then the drive says missing operating system. I try to format/s from the A: drive but I get write errors on drive C:. If I take the DMDRVR.Bin from the config.sys the the drive boots fine. But I lose my partions. And I will only have a drive of 2 MEGS. Can you give me any info about what I should do? Thanks........ ANSWER: I suggest you either read or download bulletin number 6, for information on IDE type drives in general. The most common way to setup this drive is selecting a 15 hd x 17 spt CMOS drive type entry, then install it with non-standard parameters of 1607 cyls x 15 hds x 17 spt. Whatever drive type you end up selecting, it is imperative that the heads and sectors per track remain the same between the CMOS drive type you select, and the non-standard parameters that you install with Disk Manager. You can't change them on the fly like you used to be able to with a standard ST506 style drive, (you'll get a "Missing Operating System" or similar error). ------------------------------------------------------------------------- QUESTION: Two problems using Disk Manager - N version 3.10 occurred while trying to install a Conner Peripherals CP-3104 today 5-01-90 on an 12MHz AT class computer to replace the current 40 MB drive. . The computer: Achieve Microsystems mother board using an AMI Bios and VLSI chip set; 1664K RAM; SMC PC-130e arcnet card; EGA; 360K and 1.44M floppy drives. This computer has been running ELS I fine for 9 months with a Seagate ST-251-1. . The procedure: Configure CMOS. Boot with MS-DOS 3.3: no config.sys, autoexec.bat, etc. DMN executed in manual mode. F3 option to install drive. F6 option to netware partition. F7 option to prepare drive (compsurf equiv.) Novell prepare (never got this far) . Problem One: The DMN prepare function always gives time out errors with the CP-3104 configured non-standard type 1 or DMN's choice type 37. No difference if configured standard with Bios user-definable type 47 at 776 cyl., 8 heads, and 33 sectors per track. . Problem Two: Upon exiting DMN EVERY TIME (4 times) got the error "Memory allocation error." "Cannot load Command.com" . So after rebooting tried Novell's ELS I prepare.exe command and always got "Abend: Improper ROM parameter table for AT hard disk controller. Couldn't be that easy, could it? ANSWER: In reading your message, I noticed a couple of things that are pointing to hardware related problems. If DMN is giving you a time-out error during preparation of the NetWare partition, you probably actually have a hard disk problem. Also, getting a "Memory allocation error" is pointing to an actual memory problem. You will need to get these hardware issues straightened out before we can proceed with DMN. When you got the "Improper ROM parameter table" message, that was because you had evidently attempted to set this drive up as a non-standard drive in DMN, but failed to run MODUTILS to patch your ELS level 1 to accept a non-standard drive prior to running PREPARE. The basic procedure for this drive should be as follows: 1. Get rid of your user defined entry in CMOS by entering 0's for number of cyls,hds, and spt. 2. Run DMN and select the Conner 3104 3. Let DMN partition & prepare the drive as non-standard, 775 x 8 x 33 4. Run MODUTILS to patch your NetWare for a non-standard drive 5. Run DMN/I just for good measure 6. Run PREPARE to setup your hotfix area 7. Run DMN/I again (VERY IMPORTANT) 8. Run Install and select menu option #1 to initially set up a disk for NetWare, and when asked if you wish to initialize the drive answer Y. 9. Complete the NetWare installation as normal. 10.Create a boot diskette that has an autoexec file that calls DMN/I, then NET$OS. If you still have problems after following this procedure, then they are evidently hardware related, and will have to be rectified before going any further. ------------------------------------------------------------------------- QUESTION: I HAVE PURCHASED DISK MANAGER-N V 3.04 TO USE A CONNER PERIPHERAL 3104 100MB DRIVE WITH A COMPAQ 386 DESKPRO. THE COMPUTERS BIOS TABLE DOES NOT HAVE THIS PARTICULAR TYPE LISTED, AND I WAS TOLD THAT D-M COULD RUN NOVELL BY UPDATING THE BIOS. APPARENTLY THAT IS NOT QUITE THE WAY IT WORKS. THERE IS NOT ANY REFERENCE IN THE MANUAL TO V2.12 FOR LEVEL 1, SO I RAN DMN TO SET UP THE NETWARE PARTITION AND THEN RAN NOVELLS INSTALL PROGRAM, WHICH WENT ALONG FINE UNTIL IT WAS WRITING TO THE DISK AND I GOT AN "ABEND: IMPROPER ROM PARAMETER TABLE FOR DISK CONTROLLER". I DID NOT TURN OFF OR RESET MY MACHINE AFTER RUNNING DMN. ANSWER: Someone evidently sold you an old version of DMN that they had lying on the shelf getting dusty. DMN v3.04 did not support ELS level 1 v2.12. Since DMN v3.11 was the first version to support ELS level 1 v2.12, you will have to contact our sales department and upgrade to DMN v3.11 for this install, (or go discuss the "old copy" situation with your software distributor). ------------------------------------------------------------------------- QUESTION: I RECENTLY PURCHASED A 'AST PREMIUM 386/16' COMPUTER WITH AN 'CONNER' 80 MEG HARD DRIVE. THE DRIVE USES AN "IDE" INTERFACE I BELIEVE. THE DRIVE PART NUMBER IS 'CP-3184'. ON MY PREVIOUS XT CLONES; I HAVE USED BOTH 'SPINRITE' & 'DISK TECHNICIAN ADVANCED' TO MAINTAIN AND CORRECT MEDIA ERRORS. SPINRITE DOCUMENTATION STATES THAT CONNER DRIVES CANNOT BE "LOW LEVEL FORMATTED" AND THEREFORE SPINRITE WILL NOT WORK WITH THEM! DISK TECHNICIAN ADVANCED WILL NOT RECOGNIZE MY CONTROLLER AND NOT OPERATE AT ALL. MY QUESTION IS: DO YOU HAVE A PRODUCT THAT WILL SUIT MY NEEDS ? A PROGRAM THAT WILL TEST AND REPAIR MEDIA FLAWS (WRITE TO AND READ FROM MEDIA, NOT READ ONLY AS IN PC TOOLS SURFACE ANALYSIS). IT MUST WORK WITH MY H/W CONFIGURATION AND DOS 3.3; OR DO YOU HAVE ANY RECOMMENDATIONS ? ANSWER: Our product DOSUTILS has the ability to perform a write/read scan on a IDE drive. If it finds any bad spots, it will ask if you want to map them out to prevent any further use. If there is a file sitting on the bad spot, it will attempt to move it (very often successfully). What do you mean by "repairing" media defects? Any program that does not simply map out the defect to prevent further use, or purports to "repair" a known defective area and make it available for data storage is extremely dangerous to your data. When the manufacturer of the drive tells you that a certain area of the disk is defective, you'd better listen to him. They have multi-thousand dollar pieces of equipment that perform an analog based test on the surface of the drive. These machines can spot a marginal track; a track that stands a very good chance of going bad in the future. You should make absolutly sure that no program tries to re-allocate one of these known defective areas. No piece of software can do anywhere near as good of job of testing the disk surface than the manufacturers testing equipment. ------------------------------------------------------------------------- QUESTION: We maintain our own PCs here and have purchased Disk Manager 4.02 for all drives. We like it and use it for low level formatting a lot. However, we have been getting in some Miniscibe 7080 AT IDE drives lately. They do not appear on the menu for configuration in DM. What we want to know is how many cylinders, heads, sectors, WPcomp, and Lzone, for this drive. We also know that the IDE will allow us to configure the cmos as any drive which has the equivalent amount of space or less. But we would like to get as much as possible out of our drives. We are using AMI BIOS which allows us to custom configure a drive, and we have done that for a Conner CP-3184 and found out that we get about 7 megs more space doing that. If you can help we would appreciate it. ANSWER: The Miniscribe 7080 IDE drive is not a universal translation drive. This is to say that it can't translate to just any combination of parameters like some other IDE drives can. It has 4 basic modes of translation that it can do. These are as follows: 10 heads by 17 sectors per track, with cyls in BIOS less than or = 981 6 heads by 33 sectors per track, with cyls in BIOS less than or = 832 9 heads by 17 sectors per track, with cyls in BIOS less than or = 1023 8 heads by 39 sectors per track, with cyls in BIOS less than or = 528 You will either have to have a drive type with one of these geometries, or create one as a user defined drive type. The selection of 981x10x17 will be the one that will give you the largest capacity. ------------------------------------------------------------------------- QUESTION: I will have literally thousands of small files (ie, less than 1k) on my drive. What would be the best way to format/partition the drive so that there is as little wasted space as possible? I am willing to trade some performance for some capacity. ANSWER: A partition greater than 32 Meg will have a larger cluster size. A cluster is your minimum allocation unit of space on the drive. So for example, if you have a 30 Meg partition on the drive, your cluster size is 2k. Therefore, any file on the drive takes up at least 2k. If it's a 3k file, it would take up 4k of space. If, on the other hand, you have a 150 Meg partition on the drive, you would end up with a 16k cluster size. This could leave you with alot of slack space on the drive. (Each file would take up at least 16k of space.) If you wish to keep your cluster size down to 2k (that's the minimum), then keep your partition sizes between 17 and 32 Meg. (If you go below 17 Meg in size, you'll get a 4k partition.) The break points are as follows (under DOS 3.x): 1 - 16 Meg = 4k cluster size 16 - 32 Meg = 2k cluster size 32 - 64 Meg = 4k cluster size 64 -128 Meg = 8k cluster size 128-256 Meg = 16k cluster size 256-512 Meg = 32k cluster size Disk Manager also has a /V switch available that gives you some manual controller over your cluster sizes. (You can make them smaller than the defaults shown above.) But if you do, beware that there are some programs that won't deal with a nonstandard cluster size; starting with CHKDSK. ------------------------------------------------------------------------- QUESTION: I need to configure CMOS to recognize my Conner CP3184 Ontracks Version # 4.02 is the on S/N like printing on my disk label. I have a Compaq Deskpro 386 - 16, I am using the original controller. DM manual mode tells me that it is defaulting to the WRONG drive configuration, but I can not reach the menu which allows me to select the CP3184 directly. Diagnostics lets me select this model, but it does not write to the CMOS configuration. Note: CP3184 is an IDE drive. I have good data on this disk, the CMOS config. got trashed somehow. My Compaq "setup" program does not provide a "type" entry for the CP3184 since they are 3 years appart in vintage. If I tell DM that I have two drives (although I phusically have on only one, DM gives me the menu to cursor select the CP3184, but it will not do this for drive 1. I even ran my Compaq "setup", told it that I have 0 hard drives. DM asks me if I want to change the CMOS config due to test failure, I answer yes, DM responds by writting the default drive's config, NOT the CP3184. DM fails to allow me to access the menu to select CP3184 regardless of what I try. ANSWER: No generic version of Disk Manager (prior to v4.20) had the Conner 3184 in the drive selection menu. Are you using the generic/universal version of DM, or an OEM version? Does it have various manufacturers to choose from in the configuration menu, or just several model numbers? Setting up an IDE drive correctly with a version of Disk Manager prior to v4.20 was not a automatic procedure, and you would be better off calling us on the tech support voice line so we could work thru this together in real time. QUESTION: At this time, after a day of trying to track this one down, I hopefully have this situation under control. It turns out that a drive type 31 is correct for the CP3184. I took my disk to a 386SX we have in house that has an IDE interface and a Phoenix BIOS. Using a ROM-based setup, I configured to drive type 31 and it booted without a hitch. So at this point, I knew the disk was fine. I backed up all three logical drives at this point since I had access to the disk. Now I have gone back and repartitioned and reformatted my disk and it boots fine. Is it possible that you have a problem with the IDE version that modifies the boot block such that the Compaq BIOS does not recognize the Operating System on the hard drive??? By the way, Version 4.20 (it is not generic nor an OEM version... Restated, I do not know the difference, we do not OEM Ontrack Disk Manager, it was provided with my Conner CP3184. This version does provide a manufacture's menu for a number of Conner drives, although, I talked to Hard Drives International, Tempe, AZ where I purchased the drive, and all I had to do was write (option W) back to CMOS config. HD Internation was a lot of help, I will recommend them in the future! Please provide me with an answer to my question regarding the Compaq Boot Disk problem. I prefer to use the Ontrack DM as I can use 20 Mbytes more of my disk than with FDISK, but I can not have this problem occuring every month! Additionally, previous to the purchase of my Conner disk, I was using SUPER PCK with my CDC 40 Meg (type 17) with great success. The DM driver anversion 2.03 of PCK do not cooperate! I read that you do not recommed disk Caches. I do, but only if they cooperate! You also said that the developers of PCK had contacted you. Do you know if the latest and greatest version of PCK will cooperate with the IDE version of the DM driver? Thank you for your time. ANSWER: I don't think you understand how an IDE drive operates. An IDE drive will remap it's self to appear as though it has whatever geometry that you tell it that it has, either thru your CMOS drive type selection, or thru the nonstandard parameters that you write to the drive when using Disk Manager, (read bulletin 6 concerning IDE/AT interface drives for more info). The problem that you were experiencing was due to a mismatch between the number of heads and/or sectors per track of the CMOS drive type you had selected and the nonstandard parameters that you selected thru DM. When you have a mismatch on either of these parameters, you will experience a "missing operating system" or similar message because you are telling the IDE drive to remap it's self to a different number of heads and/or sectors per track after it has already mounted the DOS partition, therefore it can't read that partition any more. If you keep the heads and spt consistent between what CMOS says and the nonstandard parms that DM uses, you won't have this problem. The only parameter that can change between what CMOS says the drive is and what Disk Manager says the drive is, is the number of cylinders. It seems that you are also a little confused about what the W option in the config menu does. It in no way writes to CMOS. The W option writes the nonstandard parameters that you have selected with DM to sector 8 of track 0 on the hard disk. This information is then read by DMDRVR.BIN at boot time and the drive parameters in low memory are replaced with the new ones. The two machines that you were using evidently had different entries in their drive tables, therefore if you couldn't find a selection that was the same between the two machines, you would have to reconfigure and repartition the drive when you moved it from one machine to another. What I meant by an OEM version of Disk Manager was this: If you purchase DM directly thru our sales department, or from a computer dealer as a stand alone package, and it has ALL manufacturers drives listed (over 250 different drives in total) and has a file on the diskette that is named DMPARMS.OCS (the drive parameter file), it is the GENERIC version of DM. Any other version only supports a limited number of drives and is called an OEM version. With an OEM version, the parameters of the drives that are supported by that version are hard coded into the program, (not in the outboard file DMPARMS.OCS), and you are not allowed to manually change the number of cylinders or heads of any drive that you select. Also, remember that if you are using Compaq DOS v3.31 you must use the /4 parameter when entering DM to tell it that this DOS is going to try and pass it's self off as a 3.x DOS, but not to believe it, it's actually using a DOS 4.x type format structure. All kinds of problems can result from preparing a write/read partition under DM and this DOS if you don't use the /4 switch. As for Super PC Qwik, I don't think they were DM compatible prior to their version 3.21. I know they are now up to at least version 3.5. You should call them directly for current version information. ------------------------------------------------------------------------- QUESTION: Hi, I am having major problems...i am trying to install a imprimis 94244 -383 in a GCH 286 useing Awardware Bios v 3.03. I am useing the disk manager that came with the drive [v4.02 i think]. The problem i am having is this, i set the type of drive in the beginning, and it tells me i will lose 64meg, which i accept because of the type of bios, i then run the autoconfig, and i get a syntax file control error when it tries to partition. i tried manually, and it accepts the partitioning, til i try to write it, then it says there is no partition after i try to write it. not sure what the problem is...can you help? ANSWER: The errors that you are receiving are due to the fact that you don't have a CMOS drive type in your ROM BIOS that will support this drive without going over 2047 cyls. It looks like the only version of Disk Manager that will install this drive correctly would be Disk Manager generic v4.20. It was just released on 5-31-90 and has the new XBIOS overlay, (the new replacement for SWBIOS). XBIOS allows you for the first time to go over 2048 cyls. With XBIOS, you would be able to configure this drive as 2587 cyls x 15 hds x 17 spt and would be able to use the full capacity of the drive. Otherwise, you'll have to set this drive up as a standard drive. To do this, you would pick the CMOS drive type in your drive table in ROM that is closest to 310 Meg, without going over, and either use Disk Manager or FDISK to partition the drive. You would only get the capacity of the drive type that you selected. ------------------------------------------------------------------------- QUESTION: I am installing a Microscience 7100-00 drive, which is 110.4 MB (formatted), with 855 cylinders and 7 heads. My machine is a HIMS '386 with an AMI BIOS (dated 12/14/89). The closest I can get to a drive type is 31 (98MB). There are also types 9 (112MB) and 45 (114MB) that are "close," but a little bit over the 110MB the manufacturer states. The controller (if that's what you want to call this thing with almost nothing on it) is a JZ-I-001 (???) made by God-Knows-Who! Anyway, the little controller manual states that they "utilize the industry standard controller chip, guaranteeing command level compatibility with the IBM PC/AT standard and provides floppy disk and 'AT bus' hard disk interface." Two questions: 1. Should I worry about the Drive Type in the CMOS, or will Disk Manager's device driver handle that? 2. One of the drive types in the BIOS is Type 47 ("roll your own"). Should I choose that one? 3. (OK, three questions ...) Can I purchase a printed manual from you nice folks? The on-line help is fabulous, but takes a little doing to print it out, and it has no index. Have been using Storage Dimensions' SpeedStor for years, but attended a Seagate seminar in Silicon Valley the other night and they alluded to the fact that Disk Manager Is the Future! Congratulations! I've just received your brand-new version, and it looks real good! ANSWER: The 7100 presents a problem. You would be best off entering the parameters yourself as a user defined drive type 47 and partition it as a standard drive. The problem is, Microscience originally gave us the wrong sector per track value for the drive, and said, "Just put it in. It'll work.", without giving us the chance to test it. It didn't work. Besides the wrong spt value, it turned out that it used a nonstandard defect handling technique and if you attempted to initialize the drive thru DM, you would wipe out the defect list and would have to send it back to the factory for re-initialization. As of version 4.05 of Disk Manager for Microscience, the 7100 has been removed from the drive selection menu completely. The 7100 also does not appear in the current generic version of DM for the same reason. You can still partition the drive thru DM as a standard drive, just don't attempt to initialize it. As for your question about a printed manual, there is none for Disk Manager. The HELP.EXE program on the diskette is the manual. Our future versions of DM will print page numbers and will have an index. ------------------------------------------------------------------------- QUESTION: Does Disk Manager currently, or will it in the near future, have the ability to do a low level format on IDE hard drives? ANSWER: Disk Manager (generic/universal) v4.20 which was released on 5/31/90 knows more about IDE drives than any earlier version. It will allow you to initialize (low-level format) some IDE drives (for whatever reason). But not all IDE drives are the same. Some IDE drives are "embedded servo" drives that you can re-initialize with no problems because they protect their sector headers from being zeroed out, thus protecting their defect list. Others don't have this protection and allow a real initialization, and use a standard method of defect handling but don't come with a defect list for entering the bad tracks. Some others don't use a standard method of defect handling, so you don't have any way to enter the bad tracks, even if you had the list. There are even some drives that will be logically destroyed if you initialize them, and will need to be sent back to the factory for re-initialization. The safest solution is to NOT initialize an IDE drive. All IDE drives are initialized at the factory, and shouldn't need to be re-initialized in the field. ------------------------------------------------------------------------- QUESTION: First thanks for the BBS for info..... I would like to know if the Fujitsu drive M-2612T Info is in your BBS system.... Also if you have had success running this drive, IDE with your software and Netware ELS 2.15, if so maybe some hints that might help in install.... I am familiar with Network O/S and am now going to try Netware for a client... ANSWER: You will need to add the parameters for the M2612T to your DMPARMS.OCS file on the DMN diskette before it will support this drive. The line that you will need to add is as follows: M2612T 1334,4,0,65535,11,5,12,180,40,0,33 You can edit the DMPARMS.OCS file with any straight ASCII text editor, (EDLIN if nothing else). Once you start editing the file, the structure of that file should be obvious. Add the above line to the Fujitsu section of the file. After you do that, your DMN will offer this drive as a selection from within the Fujitsu drives menu. Since this will be a non-standard drive installation, you will have to run MODELSII to patch your ELS level 2 v2.15 to accept a non-standard drive. (You run MODELSII prior to running ELSGEN). Follow the section in the manual about installing non-standard drives with greater than 1024 cylinders under ELS level 2. If you have any other questions, feel free to ask. ------------------------------------------------------------------------- QUESTION: I am installing a Connor CP-3184 drive in a clone 12MHZ AT. The Company I purchased the drive from (Microtex) said to use the same settings as the Seagate ST-4096 (1024cyl, 9 heads). However, the recording at Connor Corporation says this drive has 832 cyl, 6 heads and 33 sectors. When I set it up as an ST-4096, it formats and boots under DOS 3.3. I haven't used DiskManager-N yet until I find out if I have set it up right. Please tell me how to set up this drive since I must have the system back up prior to Monday morning. ANSWER: If you are willing to take a 4 Meg reduction in capacity, then setting up the drive as 1024x9x17 will work just fine under both DOS and NetWare. If you use DMN, by default it will set the drive up with it's true geometry of 832x6x33, and you would get the full capacity. The considerations for installing an IDE drive under DOS are different than those for a NetWare install. Under DOS, you must make sure that the number of heads and sectors per track for the CMOS drive type that you are set to match the nonstandard parameters that Disk manager is using. (See bulletin number 6 for more info on how IDE drives operate). Under NetWare and DMN, these considerations go away and you can set the drive up as 832x6x33 with no problems. ------------------------------------------------------------------------- QUESTION: I have a Compaq 42Mb drive (CDC 94208-51) which was inadvertantly subjected to a low level initialization attempt: the drive appears to be unusable at this point. Does Disk Manager support initialization of this drive? Controller is a CMS IDE controller, running in a clone AT. ANSWER: I am not familiar with the 94208-51. If this is actually the model number, then it may be a Compaq only model, in which case I would have no information about it. In general, if you low-level an Imprimis IDE drive, you'll lose the defect list, and probably will flatten the skewing of the drive. (Skewing is a special formatting technique used to speed up the disk). What the net effect will be is you will have a slow drive with defects that have not been mapped out, but the drive will not be completely dead. (Other brands and types of IDE drives could be effected differently than an Imprimis drive). Disk Manager v4.20 has a /K switch that would allow you to enter the skewing information at format time, but that information would have to come from the drive manufacturer, not Ontrack. The Imprimis drive uses a traditional defect handling method, but no defect list is sent with the drive, so you don't know where the defects are. Overall, low-level formatting an Imprimis drive is not a good idea, but it won't absolutely kill the drive. ------------------------------------------------------------------------- QUESTION: The imprimis has max capacity of 279,270 sectors per drive and a native mode of 9 heads and 29 sectors per track. Since it is an IDE trnslatable drive I chose a drive type in the ALR bios that gave me 1023 cyl, 15 heads and 17 spt to maximize the capacity. I want to prep the drive for Novell with exactly the same paramaters but while the above mentioned parameters show up as the "standard" I can't get the initialization menus to register this. Even choosing the imprimis 94354-160 drive from the menu and modifying the parameters to 1023x15x17 , and then seeingly write them, DMN shows 1150x15x17. The verification part of DMN shows no problem but I can't take a chance since the 1150x15x17 multiplied out exceed the limit of 279,270 that seagate warns not to exceed. Any ideas as to what is going on? Also snce this is an IDE drive and should never be low levelled, is it fatal/not reccomended/required in choosing the hard drive initialization part of the proceedure? Lastly, is the reason for never field low levelling the ide drives never knowing what the physical drive looks like due to the translation mode. If this is so and you knew what the physical mode and formatted to that would it work, in theory? Incidentally western digital drives (40 mb ide type) can be low levelled in the field with a utility from WD called ISPFMT. ANSWER: In general, unlike DOS, under NetWare you can forget about the translation questions on an IDE drive. Install it as a nonstandard drive with it's physical parameters. What drive type you are set to makes no difference, as long as it's not 0 (zero), and the capacity for that selection is not above the total capacity of the disk. Version 3.04 of DMN directly supports the 94354- 160, just pick it and go. There are more concerns about low-level formatting an IDE drive than just the parameters. Even the "physical" parameters listed by the factory are seldom just that. They usually represent a default translation mode. An IDE drive usually has only one or two physical platters and often has a greater number of sectors per track on the outer cylinders than on the inner ones. Some IDE drives also have a nontraditional method of defect handling, and the ones that don't often don't come with a defect list. In general, don't mess around attempting to low-level format one. 1. It's not necessary. 2. You could mess it up. It's understandable for WD to have a program to low-level format their own drives. But on the other hand you wouldn't want to use that program on an Imprimis drive. ------------------------------------------------------------------------- QUESTION: Help! I have a Fujitsu 120MB IDE Drive that I am trying to install with ELS Netware V2.15 (Level II) and cannot get it to initialize using DMN/I I use the automatic routine to format it and I put it in the DMPARMS.OCS file. When I go to use DMN/I I get an error message that reads: . INVALID MBR DATA, PARAMETER TABLES NOT INSTALLED. ANSWER: You didn't tell me the model number of your hard disk, so I'm only going to be able to be of limited help. In general, if you are getting the message "INVALID MBR DATA" when you run DMN/I, it means that you didn't get the parameters of the drive written to the disk at configuration time for some reason. You should choose F3 from the main menu of DMN and select the drive that you added to DMPARMS.OCS. At that point DMN will ask you if you wish to write that configuration to the disk. You must answer Yes to this question. If you have already done this, and it still doesn't work, then evidently we aren't able to talk to the disk for some reason. ------------------------------------------------------------------------- QUESTION: I have a Toshiba T5200 Laptop. This is a 386 machine with 6 Megs of RAM. I recently installed a Connor CP3204 212 Meg. drive and tried to use Disk Manager to setup the new drive. I ran the dm/c option and everything seemed to go ok although when I finished the process the computer would not boot from the hard disk. If I booted from a floppy with the appropriate driver in the config.sys file I could access the hard drive and use it--but I could not boot from it! This is problem #1. The second problem is that I transfered about 35Megabytes of data onto the drive and it used nearly 50 Megabytes of space on the drive. What have I done wrong, and how can I correct it? Note: The Toshiba T5200 has no setup information for this drive, the only drive options it has are for a 40 Meg, 100 Meg, and No Drive. ANSWER: Regarding your first question, what version of Disk Manager are you using? No version of Disk Manager prior to v4.20 was meant for IDE drives. Some versions had the default drive geometry for several IDE drives, but if you didn't have these specific drive type entries in your computer's drive table, Disk Manager wouldn't be able to handle the translation of the drive correctly. As of v4.20 of DM, it now handles IDE drive translations correctly. If you don't have version 4.20 of DM, getting it should solve your problem. If you DO have v4.20 of DM, then try a SYS C: to manually transfer the system files to the hard disk, or a FORMAT C:/S if the SYS C: doesn't fix it. As for your second question, it sounds like you have a large partition (>32Meg) under DOS 3.x. When you have a partition greater than 32 Megabytes under DOS 3.x, your cluster size increases along with the partition size. A cluster is your minimum allocation unit on the drive. For example, if you have a 200 Meg partition, you would have a 16k cluster size. This means that even a 1 byte file would take up 16k of disk space. A 17k file would take up 32k of disk space. The increased cluster size is a tradeoff for having a large partition under a version of DOS that didn't directly support it. If you make your partitions smaller, that would decrease the cluster size. ------------------------------------------------------------------------- QUESTION: I am not a techie, but have heard horror stories re Ontrack Diskmanager and Windows 3.0. I am running an IBM clone 386 16/25 Turbo CPU with DOS 3.3 currently loaded. I have a 70 meg IMPRIMUS Model 9424 half height hard drive with a newfangled controller with a name I cannot recall. The controller is not an RLL or an MFM. I am using DiskManager v. 3.2. The hard drive is split in 2 partitions. I have read Bul #12, but am still uncertain of how the info actually impacts on my beast. ANSWER: I see that you have read bulletin number 12. This should explain the current situation between Windows and Disk Manager. There are two cases of not being able to do something that are explained in that bulletin, neither of which has anything to do with data corruption. The horror stories that you have heard are being spread by hysterical people who don't know what they are talking about. There have been absolutely no confirmed cases of data corruption being caused by Disk Manager. On the other hand, SMARTDRV.SYS, (the "caching" program that comes with Windows), may cause problems if you have a nonstandard drive, or a drive with greater than 1024 cylinders. If this is the case, (I don't know for sure since you only gave me part of your Imprimis model number), then you should remove SMARTDRV.SYS from your system. This has always been the case with SMARTDRV.SYS, even before Windows 3 came out. QUESTION: I sent msg #407 Thurs 6/29 re Ontrack Disk Manager vs Windows 3.0. I mistated my hard drive model number as #9424. The correct model number is #94204. To restate: I am not techie enough to be sure of what Bul 12 is telling me. I use an IBM clone 386 16/25 turbo CPU, use DOS 3.3, have an IMPRIMUS 70 meg 1/2 height hard drive-model 94204. The controller is not an RLL or FMF but I can't recall the name - it's a new type. My concern was the stories I've heard from supposedly "smart" folks that if Ontrack (v 3.2 is mine) and Windows 3.0 are used together, the disk WILL eventually crash and require reformatting - a hassle I wish to avoid. I run DESQVIEW/QEMM386 a lot but wish to run Windows in all of its glory to see if it really has any virtue. Now that you have my disk's model number, am I in trouble running Disk Manager and Windows 3.0 (including smartdrive). I have 8 meg RAM and might as well use it. ANSWER: There are at least three different drives with model numbers that start with the number 94204. The 94204-65 is 941 cylinders by 8 heads by 17 sectors per track. The 94204-71 is 1024 cylinders by 8 heads by 17 sectors per track. The 94204-74 is 941 cylinders by 8 heads by 17 sectors per track. None of these drives have more than 1024 physical cylinders. But if you don't have a drive type in your drive table in ROM with 8 heads, (IBM standard drive type 4 is 940x8x17, but not all computers use the IBM standard drive table), then because it's an IDE drive, it may be translating to a fewer number of heads but more logical cylinders, (effectively putting you over 1024 cylinders apparent to the system). One way to find out for sure is to perform the following procedure: 1. Run Disk Manager in manual mode; (DM/M). 2. Once in the main menu, press C and hit return for the configuration menu. 3. Write down what it says between the two horizontal lines on the screen. For example: Disk parameters (cylinders by heads by sectors) are now REPLACED. Drive 1 is setup as 939 x 8 x 17 and is actually 1023 x 8 x 17 or, Disk parameters (cylinders by heads by sectors) are now STANDARD. Drive 1 is actually 1023 x 8 x 17 4. Press ESC twice. When asked if you wish to return to DOS, answer Y. If any of the number of cylinders shown on the screen is over 1024, then you will need to place the VIRTUALHDIRQ=OFF line in your SYSTEM.INI file, and you must stay away from SMARTDRV.SYS. If none of the numbers shown are over 1024, but the numbers show up as in the first example, (parameters are REPLACED), and the number of heads or sectors are ACTUALLY different than they are SETUP as, then you don't need the VIRTUALHDIRQ=OFF line in your SYSTEM.INI file, but you must stay away from SMARTDRV.SYS if you are using DOS 3.3 or higher. In any event, if you are using DMDRVR.BIN, Windows will not allow you to set up a permanent swap file, but temporary swap files will work fine. In general, SMARTDRV.SYS is a rather rudimentary cache, as caches go nowadays. If you are interested in using a cache that actually makes your system run faster, and would not have the possibility of malfunctioning in the above scenarios, I would suggest Super PC-Kwik from Multisoft as one to try. QUESTION: One more go around on my windows 3.0/Disk Manager situation. My hard drive is an Imprimus 94204-71. DM/M reports setup as 1023 x 8 x 17 and actual config as the same. It also reports "parameters replaced." Where does this leave me with Windows 3.0. I gather that Smartdv.sys is a loser and Qemm386 can't coexist with Windows - or can it? Is Smartdv the problem using QEMM with Windows? ANSWER: Once again... If your drive does not exceed 1024 cylinders, (yours doesn't), and does not have a head or sector per track mismatch between what it is "setup as" and what it "actually is", (yours doesn't), then there should be no problem with SMARTDRV.SYS. The drive you are using does not have more than 1024 cylinders, so you don't need the VIRTUALHDIRQ=OFF line in your SYSTEM.INI file. You WILL have DMDRVR.BIN in your CONFIG.SYS file though, and because of this Windows won't allow you to create a permanent swap file. Temporary swap files work fine. We are currently working with MicroSoft and hope that they will modify their swapfile program to allow a permanent swapfile on a nonstandard hard disk soon. I am not able to answer your questions about QEMM. MicroSoft or the makers of QEMM would be a better source of information on that matter. ------------------------------------------------------------------------- QUESTION: I am trying to attach a Conner 3104, 102 MB H/D to an AT compatible PC that does not have the correct drive type in it's CMOS drive table. I am looking for a BIOS that has "user defineable" hard drive parameters so that I can use this drive. I have not yet looked at the files that I downloaded, but do you have a BIOS that is compatible with Award S/W Inc's ver 3.03 BIOS? I would greatly appreciate your reply. Thanks. ANSWER: Version 4.20 of Disk Manager will install the 3104 at full capacity in any AT style machine, without the use of Superproms or a user definable drive type. Superproms are not a BIOS as such, but a set of add-on ROMS that will give you about 150 more drive types to choose from. They do not at this point have an entry for the Conner 3104, but have several that will come close to the full capacity of the drive. If you are using DOS, I would recommend Disk Manager v4.20 over Superproms. ------------------------------------------------------------------------- QUESTION: I've got an Imprimis IDE drive, 94354-160. It's 160M unformatted, 143 formatted. Using Disk Manager it partitions just fine, with my setup being 136M on C: and 6M on D:. That all seems to work fine. The DMDRVR.BIN file is out in the config.sys and invokes without any apparent problems. So I should, and can, go to drive D: and there is in fact 6 M out there. Patching the CHKDSK program in DOS 4.01 gives me the disk information as it should. I can copy a file to the disk and copy it back -- intact. Now for the rub; If I copy a group of files to the D: drive using a wildcard copy it takes off as if to copy and then aborts with an error message. I don't remember now what the message was, because it has been several months since I did it. The effect was a complete trashing of the partition table. The solution was to repartition, reformat and restore. Not a process I enjoy doing on a regular basis!!!!! I managed to trash the thing 3 times in one day on my machine and gave up. Then trying it on another machine, same specifications as the first it produced the same results. So, it's not the hardware! The net result is that the 6M drive at D: is useless. In fact it's dangerous because of what can happen if a user were to copy files to the thing. ANSWER: I need a little more information about your install before I can attempt to figure out what is happening on your drive D:. How is it set up? The information that I would need can be found in the configuration menu of DM. If you run DM/M and then press C for the configuration menu, the screen will say something to the effect of: Drive 1 is set up as xxxx by xx by xx and is actually xxxx by xx by xx I need to know what the x's are. I also need to know about your partition structure (starting and ending cylinders for each partition), as well as the version of Disk Manager you are using and what other things are in your CONFIG.SYS and AUTOEXEC.BAT files. It sound like you are seeing a 1024 cylinder "wrap" situation occur. This is where a program will attempt to bypass DOS function calls and go directly to an interrupt 13 for disk access (for speed reasons) and doesn't know how to get along with SWBIOS. Are you using a software cache program of any kind? A cache that doesn't know how to get along with SWBIOS is the most common reason for a 1024 cylinder "wrap" like this. If you are using a cache program, try taking it out and try your test again. If this turns out to be the cause of the problem, I can recommend a cache that does know how to get along with SWBIOS. The makers of Super PC Kwik have been in contact with us and have learned how to function correctly on a drive with greater than 1024 cylinders. (Current versions of Super PC Kwik only). QUESTION: I'll check the info from the configuration. My disk is 1072 cylinders, by 9 heads by 29 sectors. I am partitioned for 136 & 6 Megs partitions. I am running a cache, which might be the problem. I'll have to try it without the cache. Have you got any idea of how much memory the PC Kwik program takes from base memory? The one I am using is EVCACHE.SYS and uses only 11K of base memory. Running under 4.01 DOS I don't have much system memory to spare right now. I am considering trying 386 to the Max, do you have any experiences with that? The SWBIOS doesn't need to be in the config.sys or autoexec.bat does it? Ok, here's the specifics you requested: Drive is actually 1071 by 9 by 29 Imprimis version 4.02 Start End Type Bootable System 0 1022 DOS Y MS-DOS 4.0 1023 1070 Write/Read N Nosystem I understand that SWBIOS has to be present to use the DM/M properly with this IDE interfaced drive, otherwise I get a SCSI on the screen and can't do anything. But, do I have to have the SWBIOS active all the time? I am using DMDRVR.BIN in the config.sys to allow me to get to D:, this was all I thought that I needed. Please clarify this for me. As I said in my earlier message, I AM running EVCACHE.SYS in the config.sys file -- which could be the root of the problem. I am more than a little reluctant to test the thing since if it doesn't work it will trash the partition table and I will have to reformat and restore the system. Also, I was under the impression that I can't make one large partition with DM because of some sort of ROM BIOS limitation that a drive be no more than 1024 cylinders per partition. Is this correct thinking? ANSWER: If, as you said, the parameters of this drive in the configuration menu show as being STANDARD and ACTUALLY 1071x9x29, then you must be using a user-defined drive type. You should change the parameters of your user defined drive type to show no more than 1024 cylinders. This is because if DMDRVR sees the parameters as being STANDARD and over 1024 cylinders, it assumes that the BIOS and operating system must know how to actually access the area beyond cylinder 1024 (since it's appearing as a standard drive). This, unfortunately, is very rarely the case. Once you change the user defined drive type to show only 1024 cylinders, you will need to run SWBIOS, go into Disk Manager in manual mode (DM/M), go to the configuration menu, press N for nonstandard parameters, select the 94354-160, answer N when asked if you want to change the parameters of the drive, then press W to write the parameters to the disk. At this point the parameters should be appearing as shown below: Disk parameters (cylinders by heads by sectors) are now REPLACED Drive 1 is setup as 1023 x 9 x 29 and is actually 1072 x 9 x 29 This will ensure that the SWBIOS portion of DMDRVR.BIN is invoked at boot time. You do not have to have SWBIOS.COM in your CONFIG.SYS or AUTOEXEC.BAT, (or even on your hard drive at all). DMDRVR.BIN has the SWBIOS code within itself, and when it sees the parameters as being "replaced" and over 1024 cylinders, it kicks in it's own internal version of the SWBIOS code to support that portion of the drive. You can double check that SWBIOS is being loaded by watching the Disk Manager sign-on banner at boot time. There should be a line that says "Software BIOS enhancement installed" right above the banner. As far as the cache goes, my recommendation still stands. Multisoft's Super PC-Kwik is the ONLY cache that I know of that works properly with a drive that has greater than 1024 cylinders. I don't know it's memory requirements. As for your last question; All DOS partitions are limited to the first 1024 cylinders. No exceptions. No version of DOS produced to date, (that's right, not even 4.01), can see anything above 1024 cylinders. What SWBIOS and DMDRVR.BIN are allowing you to do is to place a "write/read" DOS compatible partition above the 1024th cylinder so this area of your drive is not wasted. As long as you allow SWBIOS to load at boot time by not trying to fool it with a user-defined drive type, and don't use caches that aren't designed to function on a drive with greater than 1024 cylinders, it is a safe partition. ------------------------------------------------------------------------- QUESTION: I have been trying to install a WD 95044-a 41 MB IDE drive with WDAT240 controller in a clone 286 and keep recieving the following error message from the Novell Installation Utility when I try to execute the Installation utility ; "ABEND: Invalid process ID passed by interrupt procedure to Kernel." This error occurrs right after install is selected. The screen gives the usual NetWare "Analyzing your system to determine number of drives and types..." message and then displays the above error. I have tried the DMN autoinstall with drive type 2 (recommended by DMN after install) and with drive type 47 (User defined with 781 cyls, 4 heads, and 27 sectors per track) all after running MODELSII on the 3 Netware disks requested. Any ideas? P.S. The OEM I recieved the clone from CompTech (Toronto, Ontario) says that this configuration is viable and has been done by them on the same hardware. The system has a Hercules monochrome adapter and another WD 95044-a hard drive installed as a DOS drive and will have an AST RamPage installed with 2.0 Meg on board ( I have tried the install with and without the RamPage and do not believe it to be a problem same error results.) Thank-you in advance. ANSWER: The most common causes of the message "Invalid process ID passed to interrupt procedure to kernal" are: 1. Selecting ISADISK twice during the configure portion of NETGEN. Once as channel 0, and again as channel 1, when you actually only have one controller (therefore only one channel) in the machine. You may have two disks attached to the controller, but that is still only one channel. 2. Not running DMN/I prior to running NETGEN to install NetWare. If neither of these seem to be the case, then one other thing you could try is to run DMN/M and then exit immediately, (instead of DMN/I), then run NETGEN and see if it makes a difference. This was a known problem with version 3.04 of DMN manufactured for Calabco, (DMN/I didn't do all it was supposed to do in that version), but v3.10 didn't have the same problem. If you find that this does make a difference, (I don't think it will), please let me know! ------------------------------------------------------------------------- QUESTION: Installed Conner 104 meg drive with Disk Manager. Couldn't tell from doc what brand of controller (IDE 16-bit, from Hard Disks International). Runs fine. When DMDRVR.BIN is Installed in my CONFIG.SYS I can't get a cursor in Lotus 1-2-3. Could you please let me know what the problem is? Thanks. ANSWER: There are no known problems between Lotus 1-2-3 and DMDRVR.BIN. Please fill me in a little more on what you mean by "can't get a cursor". Do you mean that the system "hangs" when you try to run Lotus, or do you mean that everything works fine in Lotus but there is no cursor displayed on the screen? For a test, try making a clean system by renaming both your CONFIG.SYS and AUTOEXEC.BAT files to something else, then create a CONFIG.SYS that only has the line DEVICE=DMDRVR.BIN in it. Then reboot and try running Lotus. Let me know what you find out. QUESTION: To follow up: I follow you suggestion, and the problem is definitely related to the presence of DMDRVR.BIN. When it is defined in CONFIG.SYS Lotus operates functions correctly except for the loss of the cursor. Normally, the cursor appears as a dark underline against the reverse video that designates the current cell. In edit mode, the cursor appears as a white underline on the line which displays the cell contents. When DMDRVR.BIN is loaded, Lotus boots with a white underline in the far upper left corner of the screen (directly under the current cell indicator (example:A1). You can move around the speadsheet normally, but the cursor remains fixed. In edit mode, there is no change, you can edit the cell but have to guess about cursor position. When DMDRVR.BIN is removed, everything goes back to normal. Help!! Lotus use is critical to my company, but I also need access to all that space (104 megs) that we just bought. Thanks! ANSWER: I'm at a loss to come up with a solution (or reason) as to why the cursor fouls up when you run lotus... The areas of the system that DM works in have nothing to do with screen functions.. most of our admin and accounting department are running LOTUS on hard drives that were of course prepared with DM and are loading the DMDRVR.BIN with no side effects as you describe.....we were unable to reproduce the problem in our lab.. suggest you call our tech support staff by voice for more assistance (be at the computer if possible) ------------------------------------------------------------------------- QUESTION: HI... I HOPE YOU CAN HELP... I'VE JUST COMPLETED A 386-25MHz CACHE INSTALLATION AND HAD A PROBLEM INSTALLING WINDOWS 3.0... THE HARD DRIVE IS A SEAGATE ST-1162A 143MB/15ms IDE AND THE ADAPTER IS A MAGITRONIC B237 SUPER IDE ADAPTER w/FDC & I/O... THE DRIVE WAS FORMATTED USING SWBIOS & DM 4.0 [IMPRIMIS] AND THE BIOS IS AMI... I USED DRIVE TYPE 47 [USER DEFINED: 1073 CYL & 9 HD]... I HAD TO DISABLE SWBIOS IN ORDER FOR WINDOWS 3.0 TO INSTALL PROPERLY... WITH SWBIOS I GOT AN ERROR MESSAGE ABOUT BEING UNABLE TO FIND A FILE CALLED KRNL386.EXE... AFTER DISABLING SWBIOS, WINDOWS 3.0 INSTALLED AND FUNCTIONED FINE... BUT, NOW WE CAN ONLY ACCESS 136MB/1024 CYL... IS THERE A SOLUTION FOR THIS NOW OR ONE ON THE WAY SOON?... THANK YOU FOR ANY ANSWERS TO THIS PROBLEM... Also, after downloading and reading bulletin number 6, (IDE drives), I realized I had another problem besides the original one with SWBIOS & Windows 3.0: I re-initialized a WD95044-A0MB IDE drive... YIPES!!!... Can you tell me if this drive is the "embedded servo" type that protects it's defect list when it is re-initialized?... If it is not, what should I do if it seems to be working properly at the customer site?... Also, I had the "missing command interpreter" problem after setting up a Toshiba MK234-FC IDE drive... I must have changed my setup drive type?... Anyway, to get it to work I may have used SYS C: and then copied command.com to the boot partition... Is there a potential problem with this; and, should I re-partition and prepare the drive over again... This one seems to be working fine at the customer site also... I worry when I've possibly done something funky... ANSWER: Bulletin number 12 on this BBS explains the conflict between Windows and SWBIOS. If you are using SWBIOS on a 386 based machine, then you must place the line VIRTUALHDIRQ=OFF in your SYSTEM.INI file, or the system will hang when you try to run Windows. But... here's a better solution: Since you've got a user defined drive type available, and are using an IDE universal translation drive, why not take full advantage of the capabilities of both? Let me explain... This particular IDE drive is of the universal translation type. This means it has the ability to "reconfigure" itself to appear as though it is whatever you tell it that it is. Using this capability of the drive in conjunction with the user defined drive type, you could set this drive up to have full capacity in a configuration having less than 1024 cylinders so you wouldn't need to use SWBIOS. This drive is 1072x9x29 for a total of 279792 sectors on the drive. Now all you need to do is work backwards and arrive at a geometry that equates to 279792 sectors, but is less than 1024 cylinders. For example: 277x16x63, or 717x15x26, or 643x15x29, etc... Set your user defined drive type up with one of these sets of values, repartition the drive with either DM or FDISK (if you're using DOS 4.01 you can have the whole thing as drive C:), and you've then got a drive that appears to the system as a standard drive with less than 1024 cylinders. In the process, you've gotten rid of the need for SWBIOS, thus no more conflict with Windows. As for your questions about the WD95044-A, Western Digital would really be a better source of information about the internal workings of their drives. My limited information on WD drives says they are stepper motor drives, and therefore would lose their defect list when inited. But, my information also says that they use a traditional defect handling method and come with a published defect list, so you can re- enter the defects with Disk Manager and init the drive again to re-map them out. A better solution may be to use WD's FMT program since it is designed to support their drives directly. Your problems with the Toshiba MK234-FC may indeed have been caused by changing the CMOS drive type after partitioning the drive. This would cause the "missing command interpreter" situation that you spoke of. Simply running SYS C: and copying over COMMAND.COM would not solve this problem, since you have actually changed the geometry of the drive after partitioning it. I would recommend re-running Disk Manager (v4.20) on the drive to make sure it's setup correctly. By the way, if you don't already have Disk Manager v4.20 generic, I would recommend getting it. DM v4.20 is the first generic (universal) version that directly supports IDE style translations. ------------------------------------------------------------------------- QUESTION: Can Disk Manager for Novell be used to install Novell ELS II on a Conner 3204 200 MB IDE drive? The drives physical parameters are 1366 cylinders, 8 heads, 38 sectors per track. These drives have a translation mode which allows a user to substitute a standard MFM drive type. However the largest MFM drive in the computer is 1024 cylinders, 15 heads, 17 sectors per track. This is equivalent to 133 MB. I would like to get at least a 150MB or more. I don't need the full 200 MB if I can't get it all. ANSWER: Yes. Disk Manager-N v3.10 directly supports both the Conner CP3204 and ELS II v2.15. With v3.10 of DMN, you will be able to install this drive under ELS II v2.15 at it's full capacity. ------------------------------------------------------------------------- QUESTION: After modifying ELS files, I run dmn. I have set cylinders and heads using user definable drive type 49 in set up. Disk manager insists I have 902 cylinders and 8 heads. Even when I go in and select the conner drive from the non standard drive types, It insists on 902 cylinders. When I am selecting the conner 3104, it lists the standard values I input in drive type 49. The values are ok for the conner, too. Yet when I go back to the menu after pressing enter, the damn thing still insists I have 902 cylinders and the partitioning routine craps out after 776 cylinders (the correct number). ANSWER: The version of Phoenix BIOS that ALR has been using as of late, produces a phantom SCSI drive for some bizarre reason. More than likely this is what you are experiencing. If you were using Disk Manager for a DOS installation (DM), you could use the /o switch to strip out SCSI support and this would solve the problem. Unfortunately, DM-N doesn't have this same switch, so you have no way of telling DM-N to ignore the screwy parameters that the ALR Phoenix BIOS is responding with when we do a "getparms" command. Since ALR/Phoenix don't seem to be overly interested in solving this problem, (we have attempted to discuss the problem with them), the only solution that we know of at this time is to use a different BIOS. ------------------------------------------------------------------------- QUESTION: Hello! I am having some difficulty with Disk Manager, V3.40. I am using disk manager because the application I am installing requires at least one LARGE (over 33 meg.) partition on the hard disk for its data directory - all of the data files must remain in the same directory ( and therefore, in the same partition.) Further, the folks who designed the application are recommending I NOT use DOS 4.0. Enter disk manager. The machine I am working with is the Packard-Bell Force 386sxh-10, the hardrive installed is the ST-1162. I'm using MS-DOS 3.30. I wanted to use the automated mode of DM, because I'm pretty much illiterate when it comes to Hard Drives. I tried this, and the process appeared to work fine, with the following exceptions: A:) DM reported the drive to have 1023 cylinders, the drive says it has 1024; (We're using a version of DM configured for SEAGATE drives.) B:) There was no defect list to be found. We looked both on and around the computer, and on and around the drive itself. C.) The folks the computer was purchased from said that the drive was 100 meg. DM reported (and initialized) a little over 140 (!) D.) Finally, I noticed when I was asked to to enter return if it was a standard drive that my specific model number (ST-1162) was NOT listed in the table. As I mentioned, the process appeared to work correctly, the machine did boot correctly when finished, and I did end up with a D: partition that was, lo and behold, 140 meg. I even wrote out a file that was 140 meg large, with no errors. So I was happy. For a time. Shortly after we began using the application which I had installed, I began receiving DATA ERROR READING (and WRITING) DRIVE D:. It didn't happen constantly, but enough to indicate a definate problem (Say, a few times a day). Also, the application is running quite slowly (two to three times longer) when ever it is reading from or writing to the D drive, as if it's getting quite a few retries. Is it because I didn't enter a defect list? If so, how do I determine where the defects are, so I can enter them? Am I using the right version of DM? HELP!!! ANSWER: Version 3.40 of Disk Manager for Seagate was not designed to operate with IDE/AT interface drives, (v3.40 was released way back in 1988, before the advent of IDE drives). This being the case, it was not aware that this drive was anything different than a normal ST506 type drive and allowed you to low-level initialize the ST1162A. Unfortunately, this process wiped out the defect list on the ST1162A. Even worse, the ST1162A doesn't come with a defect list, so you have no way of entering them back in. The process of initializing the ST1162A also may have flattened out the "skewing" of the drive, making it slow in performance. If you want to keep using this particular drive, you should run a write/read/compare type bad sector scanning program on it such as Dosutils, in order to get the actual hard errors mapped out. Even at that, you still may have intermittent errors on sectors that were originally mapped out as borderline sectors by the manufacturer crop up every so often because even the best disk scanning utilities can't find these in all cases. This particular drive was not supported until v4.20 of Disk Manager generic, and version 4.03 of Disk Manager for Seagate. Even with these newer version of Disk Manager, you still never should low-level initialize an IDE drive. For more information on IDE drives, take a look at bulletin number 6 on this BBS. ------------------------------------------------------------------------- QUESTION: HI. WE ARE INSTALLING AN ST1239A USING A MANUFACTURER PROVIDED COPY OF DM(4.0). THE FORMATTED CAPACITY IS RUMORED TO BEE 211MB. I HAVE TRIED USING THE SWBIOS UTILITY PROVIDED BY ONTRACK, YET THE LARGEST PARTITION SIZE I SEEM TO BE ABLE TO GET IS 169MB. WE WANT ONE SINGLE PARTITION WITH MAX SIZE AVAILABLE - 211M.....IS DMDRVR.BIN CAPABLE OF SUPPORTING A 211MB PARTITION ? IF SO, DO YOU SEE ANY FLAWS IN THE INSTALLATION PROCEDURE WHICH WOULD CAUSE A 169MB CEILING ON PARTITION SIZE? ANSWER: First of all, v4.0 of Disk Manager manufactured for Seagate does not have the ST1239A as a selection from within the configuration menu. You would need either v4.03 of Disk Manager manufactured for Seagate drives only, or v4.20 of Disk Manager generic (handles any drive, available thru our sales department). Our sales department can be reached at (800)752-1333. Even with a version of Disk Manager that directly handles your drive, you still probably won't be able to get the whole drive as one big partition. This is because the DOS partition, (versus a Disk Manager write/read partition), has to reside within the "standard portion" of the drive. The "standard portion" is whatever the CMOS drive type that you are set to says the parameters of the drive are. Any cylinders over the "standard portion", (or over 1024 cylinders, whichever comes first), is known as the "nonstandard portion" of the drive. The only thing that can reside in the "nonstandard portion" of the drive is a Disk Manager write/read partition. If you create a small drive C: DOS boot partition, (about 1 meg), you can then have the entire rest of the drive as drive D:, all in one big chunk, as a Disk Manager write/read partition. ------------------------------------------------------------------------- QUESTION: I am using an IDE drive (Conner) with Netware v2.12 ELS Level I. I am unable to load Netware because there are no instructions for Level I, v.2.12 (Level I, v.2.0 and Level II, v.2.1x are in documentation, but not Level I, v.2.1x). When I try to modify Netware files with Modutils, Modelsii, etc., I do not have the proper files for modification. When I continue with the Netware installation, I get the ABEND: Improper ROM parameters table message. Also, just out of curiosity, what happens if you do initialize an IDE drive with the DM-N program? ANSWER: There is no documentation on ELS level 1 v2.12 included in the Disk Manager-N v3.10 manual because ELS level 1 v2.12 was released after Disk Manager-N v3.10 was released. You will need to contact sales and upgrade to DMN v3.11 which supports this version of NetWare. As for your question about initializing an IDE drive, there are many ifs ands and buts involved. In general, the Conner drives protect themselves well from this. In light of the fact that you have a Conner drive, you should be safe. On the other hand, if you haven't already done so, (inited the drive), then don't. More information on IDE drives can be found in bulletin #6 and in the file IDE_Q&A.EXE both on this BBS. ------------------------------------------------------------------------- QUESTION: IDE drives again..... Have aConnor 3104 (104 MB - 776-8-33) ru7nning on a WD IDE interface board. My AST 286 AMI Bios supports this drive so I used Fdisk for partitioning and DOS formatting. Worked beautifully and the drive was seden as 776 cylinders before and after. My problem is too much curiosity. I decided to then take a look at the ONTRACK V 4.2 program that came with the disk. I ran DM/m mainly to look at the drive selection in the Non Standard section. I made no selection, but in leaving the screen the program changed my HD specs. to show 775 cylinders (As if DM had partitioned it)> Went back to Fdisk and repartitioned but now even Fdish could not see the drive as 776 cyl. Everything works fine, even with the "lost" cylinder that DM reserves, but is there any safe way to return the drive to show it's 776 cyl. virgin status under Fdisk? ANSWER: If the drive was showing up in FDISK as having 776 cylinders, then it should still be showing up as 776 cylinders, since FDISK does not pay any attention to Disk Manager's "nonstandard" parameters. I don't understand how you could have lost a cylinder within FDISK. Are you sure you not just seeing your partition going from cyl 0 to cyl 775, (a total of 776 cylinders)? If not, try this: Go into Disk Manager in manual mode, (DM/M), go to the config menu, press S for standard parameters, press W to write those parameters to the disk, exit Disk Manager, reboot the machine, go back into FDISK and see if you can allocate a partition from cyl 0 out to cyl 775. QUESTION: I still am not getting anywhere. To reiterate: AST 286, Connor IDE drive 3104, DM V 4.21 (CAL ABCO) MY AMI bios supports this drive, and I used FDISK to secment and Dos format. Worked fine!!! Wrote to cly range 0 - 775. Specs in Bios and on Fdisk show 776 cylinders. At this point I wanted to look at the DM disk that is referenced above especially to see Non stndard disk information. On DM/m command, it came on screen listing my standard drive and 776 cylinders. I went to configure, then non standard, and looked at screen. Did not do a write, but exited (however on exit i did see a writing file?? msg). From this point on I show 775 cylinders. Went back to FDISK to change my partitions, and fdisk came up with a 775 cylinder message. Repartitioning still left me one cylinder short. Your suggestion did not help. I know that DM does reserve one cylinder for a park/data area, doesn't it? Somehow this is visible to FDISK and I lost one cylinder. I can partition from 0 to 774 cylinders under FDISK, where before I could partition from 0 to 775. All works OK in the Fdisk environment, but I will be partitioning more drives, and wonder if there is any way (SAFE) to restore the drive to it's original FDISK virgin status with 776 cylinders showing. Thanks for your prompt answer and any more information that you will be able to give. ANSWER: What you are telling me doesn't make sense. If your drive type says 776 cylinders, the BIOS of your computer will cut off one cylinder all by itself for a CE cylinder/landing zone. This is not a function of Disk Manager, but your computer's BIOS. This being the case, I don't think it ever really offered you access to all 776 cylinders, since this would not be in keeping with IBM compatible BIOS standards. If your BIOS was in fact nonstandard, and was actually giving you access to all 776 cylinders, then Disk Manager would give you access to 776 cylinders as well. When Disk Manager polls the system for the "standard" parameters, it does a standard DOS "getparms" function call. It is simply asking DOS what drive parameters were put into memory at powerup time. Disk Manager does not add or subtract from the number of cylinders that DOS responds with. Even if it did, then booting from a DOS floppy, or from the hard disk without having the line "DEVICE=DMDRVR.BIN" in your CONFIG.SYS file, would circumvent all of Disk Manager's activity. Disk Manager does not have the capability to write new parameters to your ROM BIOS drive table. (If you can figure out how to make a program write to ROM, let us know! We could make a billion.) With the exception of a BIOS with a user defined drive type, the only thing stored in CMOS is a pointer to which drive type in ROM you are using. There is no room in CMOS to store the actual parameters of the drive, therefore Disk Manager can't store them there either. If you do have a user defined drive type, then there is room for the actual parameters of the drive to be stored in CMOS, but Disk Manager could not have changed those parameters. That functionality has not been written into Disk Manager. Disk Manager makes no distinction between a "real" drive type and a user defined drive type, nor does it ever write anything else to CMOS except the drive type number. The only thing that could have possibly caused this scenario would be if you happened to have two drive types that were very close to being the same. One with the parameters of 776x8x33, and one with 777x8x33. (Either one of these, or both, could have been user defined drive types, possibly set up by the in-store technician of the computer dealer that you bought the computer from.) If you were set to the entry with 777 cylinders, your computer's BIOS would slice off one cylinder, and FDISK would indeed give you access to 776 cylinders, (0 thru 775). (Although in this situation, you should have received an error from the power up self test of your machine, since you would be setup for a drive type with one more cylinder than your drive physically has. Also, FORMAT would end up mapping out the entire 776th cylinder as being bad.) If you then ran Disk Manager, went into the CMOS config menu and selected the CP3104, it would have changed your CMOS drive type number to the entry with 776 cylinders, since it knows that the CP3104 only has 776 cylinders. Subsequent runnings of FDISK would then only give you access to 775 cylinders, (0 thru 774, which would be correct in the first place). CLOSING REMARKS: Thanks for the detailed and excellent explanation. Sure appreciate your patience with a slow learner. Thanks again. Love your product and the excellent back-up! ------------------------------------------------------------------------- QUESTION: Hi!... You helped me last month with a Seagate ST-1162A IDE hard drive installation... We used the AMI bios type 47/user defined type with: 717 cyl/15 hd/26 sec [I believe] and the ST-1162A's translation mode... Everything works fine, but I'm curious if setting the write pre-comp and landing zone at cyl 717 [I guessed] would cause any problems?... If so, would simply changing these values in the CMOS setup be okay? or will I have to re-format the drive with the new values [blank?] ?... ANSWER: IDE drives handle write precompensation all by themselves internally, so your write precomp value in your user defined drive type should be set to OFF or NO precomp, or the equivalent. Setting the landing zone to 717 should be fine, as it actually means nothing in this scenario either. You might want to ask Seagate to make sure, but I believe the ST1162A is an autoparking drive anyway. ------------------------------------------------------------------------- QUESTION: We are dealers. We just received a Seagate/Imprimis Model ST1239A IDE hard disk drive from Tech Data. It came with an OEM copy of DM (Imprimis Ver 4.0). There are no drive types that match this drive on the DM disk. The drive came with no further documentation. Please respond. This drive is to be for our own internal use as our Technical Information Data Base for Novell NetWare & Microsoft Lan Manager. ANSWER: You are correct. Seagate did not have us add the ST1239A (old Imprimis model number 94354-230) to their version of Disk Manager until version 4.03, (the most current version). It IS supported by the current generic/universal versions of both Disk Manager (v4.20) and Disk Manager-N (v3.11). If I understood you correctly though, you stated that you will be installing NetWare on this drive. If this is the case, then you will need Disk Manager-N (for NetWare installs) anyway, not just regular Disk Manager (for DOS installations). For further information about Disk Manager-N, you can contact our sales department at (800)752-1333. ------------------------------------------------------------------------- QUESTION: I recently purchased the generic version of DM 4.20 to use with my new Connor CP3204 IDE drive. I've had numerous problems this past week after setting it up in my ALR Powerflex 386SX machine, but after reading your messages about IDE drives, I think I may have solved some of my problems. I run a multi-line BBS system on the computer under DesqView with QEMM and I needed more disk space than the the 40 meg unit that came with the computer. My parameters all seem to be correct in my user configurable bios type setting #49 and the first time around, I set up a couple of regular 32 meg (even though DM said the default was 33 meg) drives and one small 7 meg drive with 512 byte clusters for my 4000+ echo messages and one large 72 meg drive for file storage. Well, as soon as I tried to transfer files over to the 72 meg partition using Norton Commander, my C drive became totatly scrambled and I had to start from scratch again. This happened a couple of times and drove me crazy! I think it's what you refer to as the 1024 wrap problem??? Anyways, I finally ended up partitioning everything at 32 meg or less. Then I still had some problems with some directories getting scrambled after doing some file transfers or even running Norton Disk Doctor. I finally read your warnings about cache programs! I was using one called Cache86 by The Aldridge Co v3.0 and I took it out. Things seem ok now but I'm nervous about doing file maintenance!! I was going to ask you about a recommended cache program but have already read your response to earlier messages so I will be looking for PC-Kwik right away. Also, I was loading Dmdrvr.bin after Qemm and loading it into high memory but have now listed it first in my config.sys file and load it low. Do you have any knowledge of whether it makes a difference or which way is safest? One last question, are there any problems that you know of with a setup like mine with the following software: Norton Utilities, Norton Backup, Norton Commander, and Optune (optimizer by Gazelle)???? ANSWER: It sounds like the cache program that you were using was causing most of the problems, but I do have another question. What version of the Norton Utilities are you using? Versions with a file date prior to 1/3/89 did NOT work correctly on a nonstandard disk, or a disk with partitions greater than 32 Meg under DOS 3.x, or a disk with more than 1024 cylinders. If your version of the Norton utilities has file dates prior to 1/3/89, you will need to contact Mr. Norton and upgrade to the current version. Optune, according to my information is not compatible with a disk with more than 1024 cylinders either. (With an IDE drive, interleave optimization becomes a moot point anyway. You should never low-level format an IDE drive under any circumstance.) As far as QEMM and DMDRVR go, I don't think it will make any difference if you use QEMM to load DMDRVR high. QUESTION: Thanks for your feedback. The cache was a problem but now I'm using Super PC_Kwik and things seem ok. My Norton Utilities is version 4.5 dated 1-3-89 so I guess I'm ok but I have just recently ordered my upgrade to version 5. As for Optune, you're right it won't let me use the Tune-Disk option to check the interleave but I have always just used it for defragmentation and so far it seems to be doing ok on that; would you advise me against it? I have just ordered your DosUtils program to check and set my interleave....sound good to you? Also, before I knew better and when I was having all my troubles, I did re-initialize the Connor drive and then paniced when I started reading your messages to others until I read that it didn't destroy the Connor drive (CP 3204), although I'm suspicious that I lost the bad sectors map because there is nothing blocked out on the whole 200 megs which surprises me. I believe your DosUtils program can check the disk and block out any bad sectors, right?? I hope so....I'm counting on DosUtils to fix me up to a normal running status. Thanks again for your response! I've really learned a lot from reading the messages here.... wished I'd done it sooner <>. ANSWER: The Norton Utilities v4.5 with a file date of 1/3/89 or later is fine, but we don't know about v5.0 yet. There will probably be added functionality in v5.0, and sometimes new functions don't work quite correctly right out of the chute. I would be careful at first with it, (keep a good backup). I will repeat what I said about Optune. Do not use it with a disk with greater than 1024 cylinders until Gazelle states that it is compatible with such a disk. We have had reports to the contrary in the past, and don't know if the current version of Optune has fixed this situation. You may have just been lucky so far, and it has never had the need to write to any area above cylinder 1024 yet. You stated that you were planning to use Dosutils to "check and set your interleave". I must repeat myself again. While interleave optimization does have a bearing on the performance of standard MFM and RLL drives; when speaking about IDE drives, it becomes a moot point, (it has no relevance). IDE drives are low-level formatted at the factory with the most efficient interleave, and cylinder & head skewing possible. Since the actual controller is built into the drive itself, (the drive/controller unit is then simply attached to the system bus with a simple "paddle board" with only a couple of timing chips on it), this optimization that is obtained at the factory will never change. I would not suggest using ANY program to attempt to change this, including Dosutils. Dosutils will allow you to run an intensive write/read/compare surface defect scan on an IDE drive with more than 1024 cylinders though, and will effectively map out any hard errors on the disk. QUESTION: Thank you for all the information, you have been most helpful. Sorry that you are repeating yourself, I didn't remember reading about the interleave situation on IDE drives in the messages that I downloaded, must have missed it.....this is all new info to me since I'm used to MFM type drives. I will stop using Optune for now but since you feel that Norton Utilities v 4.5 is safe, can I safely use Norton's Speed Disk for file defragmentation? With all the activity from the BBS, I get a lot of fragmentation fairly quickly. Glad to hear that DosUtils will handle the blocking out of bad sectors, just received it today. I have one last question concerning Dmdrvr.bin. You mentioned that it would be ok to load it after QEMM, but how about loading it into high memory with QEMM?? Several of my sysop friends are divided on this point and I'm not sure if it really matters. What do you think? Thanks again for all your help!! ANSWER: I cannot guarantee the functionality of other companies products, but to the best of my knowledge the following two statements are true: All aspects, (including Speed Disk), of the Norton Utilities v4.5 dated 1/3/89 or later, work fine on a drive with greater than 1024 cylinders that has been installed with Disk Manager. QEMM is able to relocate DMDRVR.BIN into high memory in general. However, when using a drive with greater than 1024 cylinders, we have seen this to cause some problems with the SWBIOS portion of the driver, causing a "general failure" error message when attempting to access areas of the drive beyond the 1023rd cylinder. Therefore, we do not recommend loading DMDRVR.BIN into high memory with any memory manager when using a disk with greater than 1024 cylinders. ------------------------------------------------------------------------- QUESTION: I'm trying to use diskmanger-n to comp-surf a Maxtor LXT200A drive its not in the table do you have all the parameters for this drive? its a 3.5" ide type drive but thats about all I can tell except there appears to be a translate mode in it -- please help ANSWER: In order to use v3.10 of DMN to install the Maxtor LXT200A, you will have to add the following line to the DMPARMS.OCS file on your DMN diskette. (Add it under the rest of the Maxtor drives). LXT200A 816,15,0,65535,11,13,12,180,40,0,32,1 After adding this line to the DMPARMS.OCS file with any straight ASCII text editor, the LXT200A will appear as one of the selections within your DMN and will allow you to install this drive automatically. ------------------------------------------------------------------------- QUESTION: I SPOKE THIS AFTERNOON WITH A TECH AT YOUR OFFICES CONCERNING FORMATTING A CONNORS 200 MB HARD DISK (CP 3204) WITH A IDE CONTROLLER IN A USIT SYSTEM (THIS WAS ON AUGUST, 21ST). BASED ON THE INFO HE GAVE ME, THE FORMATTING OF THE HARD DISK FOR NOVELL WENT OK. I USED A TYPE 01 IN THE BIOS AND FORMATTED IT SELECTING THE CONNORS OPTION IN DISK MANAGER (1347 CYL, 8 HEADS, 38 SECTORS/TRACK). THE INSTALLATION OF NOVELL ON THE HARD DISK WENT FINE UNTIL I TRIED TO BOOT THE FILE SERVER WITH THE BOOK DISKETTE. ON THE BOOT DISKETTE, I EXECUTE FROM THE AUTOEXEC.BAT FILE DMN/I AND THEN NET$OS. AFTER THE DMN/I COMMAND IS EXECUTED, IT DISPLAYS THIS MESSAGE: DRIVE 0: 1346 CYLS by 8 HEADS by 38 SECTORS THEN, THE NET$OS COMMAND IS EXECUTED. IT THEN GIVES ME THE FOLLOWING ERROR MESSAGE: Abend: Improper ROM parameter table for DISK Controller. Power off and back on or CTRL-ALT-DEL to restart. COULD YOU GIVE ME AND INDICATION OF WHY I AM GETTING THIS ERROR WHILE TRYING TO BOOT THE SERVER IN NON-DEDICATED MODE. ANSWER: The disk you are installing in your netware setup is going to be non- standard so you must run modgen to patch your novell files for use with a non standard drive... the files that will be patched are elsgen.exe , volumes.dat , instovl.exe , els2_2.obj , ndels2_2.obj , vrepair.exe , and disked.exe.... then you must re-gen your netware and re-install it to the server drive. This will stop that improper rom parameter table abend from occurring... ------------------------------------------------------------------------- QUESTION: AM GETTING A NEW 386SX SYSTEM WITH A ST1162A IDE DRIVE WILL BE USING ZENITH 3.3 PLUS DOS. DO I NEED TO USE DISK MANAGER AND CAN I FORMAT THE ENTIRE DRIVE WITH 3.3 PLUS AS DRIVE C: USING THE "PART" COMMAND? IF NOT WHAT ARE MY OPTIONS AND YOUR RECOMMENDED PROCEDURE? ANSWER: The ST1162A has 1072 cylinders, 9 heads, and 29 sectors per track. Since the drive has more than 1024 cylinders, you will need Disk Manager in order to have access to the full capacity of the drive. Also, it is doubtful if the computer you are using will have a CMOS entry with 29 sectors per track. For this reason as well, you will probably need to use Disk Manager. The best case scenario appears to be to use Disk Manager, (with Zenith DOS 3.3+ you must use the /4 switch when running DM), and set up a small drive C: (about 1 or 2 meg, just enough to boot up on), and then you can have the entire rest of the drive as one big "write/read" partition, accessed as drive D:. QUESTION: THANK YOU FOR YOUR # 947 MESSAGE BACK TO ME ON HOW TO FORMAT A ST1162A WITH ZENITH 3.3+. I THOUGHT THAT I WAS ALL SQUARED AWAY UNTIL I STARTED TO ORDER THE NEWEST VERSION OF DISK MANAGER (VER. 4.30 FOR SEAGATE) FROM ON-TRACK. IT HAS BEEN ORDERED BUT I ALSO TALKED TO ON-TRACK TECH SUPPORT AND PULLED DOWN SEAGATE'S TECH BULLETIN ST1162A.TXT WHICH RECOMMMENDED A SPEC OF 603 CYL, 16 HEADS, 29 SECTORS. I THEN TALKED TO SEAGATE TECH SUPPORT AND GOT ANOTHER RECOMMENDATION OF 1024 CYL, 16 HEADS, 17 SECTORS AND THE TECHIE SAID THAT I SHOULD BE ABLE TO USE THIS SPEC WITH Z-DOS3.3+ AND IT'S FDISK/FORMAT TO GET ONE LARGE C; DRIVE. I HATE TO BOTHER YOU AGAIN BUT ONLY DESIRE IS: OPTIMUM DRIVE USAGE, 1 DRIVE CODE (C:) IF POSSIBLE. I HAVE TOTALLY PREPARED FROM THE BEGINNING TO HAVE A VERY SMALL C: DRIVE AND A LARGE D: DRIVE BUT DO YOU SEE ANYTHING IN THE POSSIBLE CHOICES WITH THE TOOLS THAT I HAVE TO THE BEST OPTIMIZATION WITH ONE DRIVE? MY CMOS WILL SUPPORT "USER DEFINABLE DRIVE SPECS". MY OTHER REASON OF CONCERN WAS THAT THE CLUSTER/SECTOR ? SIZE ON C: IS 512 BYTES BUT ON ALL OTHER DRIVES IS 1024 BYTES. THANK YOU FOR YOUR TIME. ANSWER: You should have told me that you had a user defined drive type a long time ago... Since you do, then setting that drive type up as either 603x16x29, or 1024x16x17 should work fine. In fact, you could set it up as virtually anything you wanted to, as long as the total number of sectors, (multiply cyls * hds * spt), that you set it up as does not exceed 279792, (1072*9*29). If you set it up with a user defined drive type, then you could indeed simply use ZDOS 3.3+ PART & FORMAT commands to have the whole thing as drive C:. ------------------------------------------------------------------------- QUESTION: Hi I have the Imprimis 94244-274 but DManager 4.02 only supports The 94244-383. Thanks. ANSWER: In order to install the Imprimis 94244-274 (also known as the ST2274A), you will need version 4.03 of Disk Manager manufactured for Seagate. You should be able to upgrade to this version of Disk Manager by contacting either Seagate, or our own sales department. ------------------------------------------------------------------------- QUESTION: Hi I am calling from Australia. I have the generic version 4.20 of Disk Manager and I am trying to install a Microscience drive model number 7100-20 in a Fujitech 386-25 with an AMI bios. I cannot initialise the drive with the OCS params that are included with this version. Is it possible to d/l the latest version of the OCS file if it supports this drive? ANSWER: It is not possible to download the DMPARMS.OCS file from the BBS, but if you add the following line to your DMPARMS.OCS file, (under the rest of the Microscience drives), you should be able to install this drive. 7100 855,7,855,65535,8,0,12,180,40,0,35,4 I don't know what the -20 at the end of your model number means. If it means that the parameters of the 7100-20 are different than the above parameters, then obviously it won't work. Also, your question mentioned "initializing", (low-level formatting), the drive. Under absolutely NO circumstance should you ever initialize this drive. If you do, you'll probably have to send it back to the factory. Simply configure it, partition it, and then prepare it. Make absolutely sure that you skip the initialization of the drive. For more information about IDE drives, see bulletin number 6 on this BBS, and take a look at the IDE_Q&A file. ------------------------------------------------------------------------- QUESTION: I have encountered a possible interaction problem. I have just installed a CONNER 200mb IDE, using DM 4.20. If I use QEMM.SYS for an EMS driver, I start getting 'stalls' reading from the HDD. If I use the OFF or AUTO parameter in QEMM, all is Ok, providing I don't use EMS modes. I am using a DTK 386sx-16 motherboard, and do not have the problem with their driver, but lose QEMM advantages. Any ideas on this one? PS. I use MS-DOS 3.30A, with partitions of 20/20/20/20/30/30/30/30. All other progs seem to be OK. Average delay is around 20-30 secs, then resumes in most cases. ANSWER: Other than QEMM.SYS having to come first in your CONFIG.SYS file, we know of no interaction problems between it an DMDRVR.BIN. Are you trying to relocate DMDRVR.BIN into high memory? When using a drive with greater than 1024 cylinders, moving DMDRVR.BIN into high memory can cause problems when accessing the area of the disk above 1024. In this situation, you would get a "General Failure" message. If you take DMDRVR.BIN out of your CONFIG.SYS file, does the problem go away? Give it a try and let me know what happens. CONCLUSION: The trouble with DM vs QEMM vs IDE drives turned out to be using the 4.25 BIOS from DTK. The newer 4.26 BIOS is working fine. Thanks for the quick reply, and keep up the good work. ------------------------------------------------------------------------- QUESTION: I now know that IDE drives are things which are best left alone but hindsight is always 20/20. My question is this. I low level formatted my Maxtor LXT200A drive using the user defined parameters entry of the most recent AMI bios in my computer. I used 816 cyln, 15 heads, 32 sectors and a landing zone of 816. After I discovered DM 4.20 I decided that I had better redo the whole thing since it has some provisions for IDE drives. What I want to know is if my fumbling around has made the drive unreliable or is this one of the embedded servo type drives which can stand a low level format without major problems. If I did cause the drive to lose bad track information, skewung? or anything else is it possible to have the manufacturer re-initialize the drive properly or do I now have an expensive doorstop? ANSWER: According to my information, the Maxtor drives are "embedded servo" drives, which means that you won't lose the defect list or mess up the skewing if you try to low-level format them. Low-level formatting an "embedded servo" IDE drive really does nothing but write 0's to every sector of the drive. It doesn't actually do what we normally call a "low-level format", which is actually defining and laying out the sectors on the drive. ------------------------------------------------------------------------- QUESTION: I HAVE V3.11 OF DMN. ANY INSIGHT FOR INSTALLING CONNER 3204 HARD DRIVE IN NOVELL 2.15 REV A ENVIRONMENT WILL BE HELPFUL. WHEN RUNNING MODGEN, I AM UNABLE TO FIND ALL OF THE FILES THAT REQUIRE MODIFYING. PERHAPS THIS IS THE PROBLEM. THANKS FOR YOUR ASSISTANCE. ANSWER: With DMN v3.11's MODGEN program, if you have an older version of NetWare that uses ATDISK.OBJ, rather than the current ISADISK.OBJ, you will have to select the "SFT NetWare/286 V2.15" entry within MODGEN, instead of the "Advanced NetWare/286 V2.15" entry. The "SFT" entry specifies ATDISK.OBJ, while the "Advanced" entry specifies ISADISK.OBJ. This method will work even though you don't actually have SFT NetWare, because the only difference within MODGEN for these two versions is the name of the driver file. Alternatively, you could select "Edit the MODGEN database", select "Advanced NetWare", change the name of the file that it is looking for from ISADISK.OBJ to ATDISK.OBJ, and then select "Search the NetWare files" from the main menu of MODGEN. One hint about the Conner CP3204: The parameters of this drive are 1366 cylinders x 8 heads x 38 sectors per track. (I know, DMN says 1348 cylinders... Conner seems to have changed their firmware after showing us a pre-release unit which we based the parameters on). Anyway, with a drive that has more than 1024 cylinders, you normally have to boot the server from a floppy disk. But, since this is an IDE drive, it has the ability to translate itself to appear as though it is anything that you tell it is. Taking advantage of this ability would allow you to boot the server directly from the hard drive. You could set the drive up as having half the number of cylinders, but twice the number of heads, (683 cylinders x 16 heads x 38 sectors per track). Setting the CP3204 up like this with DMN still gives you the full capacity of the drive, and also allows you to boot the server directly from the hard disk instead of worrying about booting from a floppy as you normally would. ------------------------------------------------------------------------- QUESTION: IN LOOKING AT THE AVAILABLE DRIVE LIST ON BOTH 3.10 DMN AND 4.2 DM WHICH WE HAVE RECEIVED FROM TMG, I CAN'T SEEM TO FIND THE CONNER 3104. HAVE I GOT A 'OEM' VERSION OF YOUR SOFTWARE WHICH ONLY HAS THE DRIVE LIST THAT TMG SELLS. IF SO CAN I SUBSTITUTE ANOTHER FILE WHICH WOULD HOPEFULLY BE DOWNLOADABLE WHICH WOULD HAVE A WIDER SELECTION OF DRIVES. ALTERNATIVELY CAN I ADD A DRIVE TO THE TABLE OF NON-STANDARD DRIVES AVAILABLE IF I KNOW THE PARAMETERS IN THIS CASE THE CONNER 3104. ANSWER: TMG does in fact sell special versions of DM & DMN designed to support only the drives that they sell. Unfortunately, it has also been designed so that you cannot change the drives that it supports. The drive parameters are hardcoded into the program, and it does not use an external DMPARMS.OCS file as the "generic" version does. You are also prevented from changing the head and cylinder counts manually from within the program. In other words, you're not going to install a Conner CP3104 with these versions of DM or DMN. I guess what happens at this point is between you and TMG... The "generic" version of DMN is available though our sales department, as well as through most software distributors. If choosing the second source, make sure it's either the "generic" version, or one that supports the drive you are going to install. ------------------------------------------------------------------------- QUESTION: I am very interested in the Conner 3204 IDE drive. Since it has 1366 cyls, or 1607 cyls as you suggested for configuring drive type parameters, both are greater than 1024. In our previous discussion on microscience HH1120, which has 1314 cyls, there is a performance penalty when accessing cyls beyond 1024 due to bios conflicts. In my case, there is a factor of 2 to 3 penalty for DOS file operations. If I partition the 3204 into multiple 32 Meg logical drives under DOS 3.3, would there be a differece in speed when I access logical drives that's located beyond cyl. 1024? What exactly is the difficulty in supporting OS2? Or more specifically for my concern, can I reserved a non-DOS, or RDWRtable partition for OS2? Is it because of dmvrdr.bin or swbios? Does this means 3204 is not OS2 compatible? If it is, what compromise does it require to run OS2 ON 3204? I consider chopping every cyls beyond 1024 not viable. ANSWER: In the previous discussions that you referred to, (msgs 315 & 319), I thought we had come to the conclusion that your "performance penalty", was being caused by the fact that you had set up your hard drive / controller combination incorrectly. A performance degradation when accessing cylinders beyond 1024 is not a reality in general. This would mean that your assumption that the CP3204 would have the same performance degradations as you had in the past, just because it has more than 1024 cylinders, is unfounded. Your proposed scenario of setting the drive up as having over 1024 cylinders and partitioning the whole thing as logical 32 meg partitions with DOS's FDISK program alone is not sound either. FDISK cannot access any cylinder beyond the 1024th by itself. But, if your computer has a user definable drive type available, you could utilize the translating capabilities of the IDE style drive and set the CP3204 up as having 683 cylinders x 16 heads x 38 sectors per track. If you do this, you would still have access to the full capacity of the drive, and FDISK would be able to see it all by itself. OS/2 would be able to see it as well. The "problem" with OS/2 and DMDRVR.BIN/SWBIOS, is that DMDRVR.BIN is simply a DOS device driver, not an OS/2 device driver. We are considering writing a OS/2 version of DMDRVR at some point if the demand warrants it. ------------------------------------------------------------------------- QUESTION: This is apparently an out-of-date version, (DMN v2.0). Is an upgrade available, or necessary to install a Conner cp3204 under ELS 2 v2.15? How do you like to handle dealer installs? should we buy new software for each one we do, or is there a site license setup? Thanks. ANSWER: If you've really got a copy of DMN v2.0, I think I know an antique dealer that would be interested in it... Yes, this is an out of date version to say the least. In fact, it's so out of date that the only information I have on it is that it was the first version of DMN to ever be released, and it supported NetWare 86 v2.0a only, (not 286, but 86), and MFM drives only. There is no way you are going to install ELS level 2 v2.15 on a Conner 3204 with it. You are going to have to purchase a new copy of DMN in order to accomplish this install, (v3.11 is the current version). In response to your questions about the license agreement: There is no "site license", or "dealer version" of DMN. DMN is licensed for one copy per server, just like NetWare. If you use DMN to set up a server, that copy of DMN must be sold to the end user with the server. Use of DMN as a "toolbox utility" to set up more than one server constitutes a violation of the license agreement. ------------------------------------------------------------------------- QUESTION: I have a small problem with a new drive I am trying to install & I can't seem to figure it out. I have a Compaq Deskpro 286e, and am installing a Plus 80AT for Compaq models s and e. It is being installed as my C: drive. It powers up ok and the system finds it, but I am unable to get it to think of it as a full 85 meg drive. Compaq's setup program has no entry for an 85 meg drive, so I was using the Disk manager program to set the drive up. It is telling Compaq that the drive is a type 19, and this ends up making it think it has less cylinders than it really does. (something like 753 instead of the full 956; heads seem to be ok) The format command tells me that I've got 72 meg of memory on the drive. The Ontrack diagnostic program says the drive has 85 meg, but the disk manager program doesn't seem to be getting the idea. Any idea what I'm doing wrong here? Compaq's diskinit program also sees it as less cylinders. The Compaq setup program does not allow for custom config- urations, so somehow I have to get the Disk Manager program to pass the right information to the Compaq. Any help would certainly be appreciated. Thanks. ANSWER: Disk Manager knows exactly what this drive is. It's DOS that is not getting the idea, and it never will. That is why you have Disk Manager in the first place. The fact that you are attempting to use "DISKINIT", "FDISK", and the Compaq setup program is what is causing you confusion and problems. Get them out of the picture and let Disk Manager do the job that it was designed to do. This is not a "standard" drive to your system. DOS alone can only see the "standard" portion of the drive, (in your case, only 753 cylinders out of 956.). This is why Disk Manager exists. The remaining portion of the drive, (everything beyond cylinder 753), must be partitioned as a Disk Manager "write/read" partition, which you will access as D:. Without Disk Manager, you would have no access to this portion of the drive at all. One other option is to create a very small drive C:, (about 1 or 2 meg, just enough to boot on), and then create a "write/read" partition with Disk Manager that covers the rest of the disk. This way you will have most of the drive accessible in one big piece as drive D:. If you are using Compaq DOS 3.31, remember that you must us the /4 parameter when running Disk Manager, (DM/4). ------------------------------------------------------------------------- QUESTION: I intend to install a Rodime 3259A ide type drive. My problem is that I purchased the drive and I am ready to install it into the system but dmn doesn't support it! What do I do? ANSWER: None of our products directly support the Rodime 3259A IDE drive yet, but you can enter the parameters of the drive into DMN yourself, and it will work fine. According to my information, the Rodime 3259A has 976 cylinders, 15 heads, and 28 sectors per track. You can edit the DMPARMS.OCS file and add the following entry under the rest of the Rodime drives: 3259A 976,15,0,0,11,13,12,180,40,0,28 The forth parameter is for the "write precomp" value. I don't have any information about what this value is for the 3259A, but another Rodime IDE that I know of sets this to 0. You should really give Rodime a call to be sure of these parameters. I have two possible phone numbers for Rodime. They are (714)770-3085 and (714)730-1388. One of these two ought to get you in the front door at least. ------------------------------------------------------------------------- QUESTION: I have just installed the Conner CP-3104 AT 104.9 MD hard drive using your Disk Manager CAL-ABCO release 4.2. My DOS boot partition is set up as 5 MB and my Write-Read partition I set at 99 MB. I put on the CHKDSK patch to get rid of the 'Invalid Media Type' error. I ran the DM surface scan which found errors. I then wrote a defect list to FLAW1.MAP which contained 59 errors detected. I understand the Conner IDE drives were supposed to be error free. Have I done something wrong? My CPU is DTK PTM-1200V2 12Mh Zero-Wait, 1MB memory, with DTK BIOS with drive types 1-47 with no user definable types. Upon copying all files from a directory in logical drive D: to a directory in logical drive C: I received a 'Sector not Found, Reply Abort, Retry, or Cancel' error in the middle of the copy. PCTOOLS DiskFix utility also indicated errors on the drive. Have I installed it wrong or do I have a bad drive. Any comments would be appreciated. ANSWER: It sounds like your machine may be too fast for either the CP3104 or the IDE interface that you are using. According to my information, (not confirmed), the Conner CP3204 will not work correctly with a machine that has a bus speed of greater than 8 Mhz, and I can only assume that the CP3104 may be rated similarly. Try inserting a wait state, or slowing it down to 8 Mhz. Some machines allow you to slow down the bus speed to 8 Mhz, and keep the processor speed up at 12 Mhz. I don't know if your machine supports this function or not, but it may be worth looking into your documentation to find out. This many errors is not normal and if slowing the machine down doesn't solve it, then the drive may be bad. ------------------------------------------------------------------------- QUESTION: I'm using DMN v3.11 to install a Conner 3204 under ELS 1 v2.12. This installation went according to instructions until bootup attempt was made. DMN /I was not run at that time. When I attempted to run DMN /I, the error message MBR not installed by DMN. Standard Parameters Installed came up. ALR has patches for the same files as MODGEN. Did not use those. Should I? Actually, when I decided to go back to square one and use ALR and NOT use OnTrack I could not get ALR to patch. They did ask for disks from an older version of ELS. Everything was very smooth until bootup time. Novell indicated 193M What did I do wrong here? I am an experienced Novell installer. First time using OnTrack. ANSWER: The ALR version of Phoenix BIOS may be the source of your problems. At one time at least, ALR was using a OEM version of Phoenix BIOS that falsely responded to the SCSI command set, causing DMN to think there was a SCSI drive in the system with screwy parameters. Here's how to test for it: 1. Boot the machine with DOS. 2. Run DMN in manual mode, (DMN/M). 3. Press F3 to select the drive brand and model number. 4. Select the Conner 3204, (if that is what drive you have). 5. Note the number of cyls, heads, and spt of the drive you select. If this is the Conner 3204, it should be 1348x8x38. 6. Answer Y when asked if you want to write the parameters to the disk. 7. Exit DMN 8. Turn the machine off. 9. Turn the machine back on and boot with DOS again. 10 Run DMN in automatic mode, and when the main screen comes up, see if the parameters are showing as "nonstandard" and show the correct number of cylinders, heads, and sectors per track. If they are showing as something like 8,000,000 cylinders or something like that, then you've got the weirdo BIOS. This is not going to work unless you put a different BIOS in the machine. If the parameters are correct, then continue on in automatic mode, and let DMN do what it needs to do. If you still have troubles, then leave me another message. If you need to leave another message, please leave a little more explicit information about your procedure, (step by step), description of the exact trouble you are having, and exact error messages you are getting than you did last time. ------------------------------------------------------------------------- QUESTION: I used Disk Manager - N to install a Connor 3104 200MB IDE hard drive in a machine, and everything went fine, I was very satisfied. However, a certain problem crops up which I have never seen before (I support several NetWare installations, but this is the only one so far that I have needed Disk Manager for). The problem is this: Certain applications, most notably R:Base for DOS, V2.11, try to see how much disk space is available prior to performing certain operations. On my NetWare installations which used Compsurf, this presents no problem, but on my installation which used DM, the applications get the result that there is no space avaialable on the disk. I don't know if this is related to the use of DM, but it does seem rather unusual, and I was wondering if you have other reports of this problem. Also, I have heard that running Novell's VREPAIR utility can have disasterous effects when used inadvertantly on a DM'd disk. Do you know if this is true? If so, what do you advise installers like myself to do to avoid the possibility that a 3rd. party, ignorant of this problem, might run VREPAIR? Thanks for your answers to the above! ANSWER: I haven't ever had any reports of this type of thing happening. Once the server is up and running, NetWare looks at the disk as a block oriented device. It doesn't make any difference to NetWare if the drive is standard or nonstandard, or over 1024 cylinders or not, it simply sees the device as a finite number of blocks available for data storage. So, I don't really see how this could be the fault of DMN. Some questions I would have are: Does the problem only crop up with this one application, or are there others? What does the CHKVOL command tell you about space available? What does the DIR command tell you? Can you actually copy files around and completely fill up the disk, and have the sizes of those files add up to the expected space available on the disk? Is the version of RBASE that you are using NetWare compatible? How did you install this disk? You said that this was a 200M IDE Conner disk. This would be a CP3204, not a CP3104 as you stated. When you selected the drive in DMN did you select the 3104 or the 3204? Did you run DMN in automatic mode or in manual mode? Are you currently booting the server from a DOS floppy, running DMN/I and then NET$OS from the floppy in order to bring the server up? Or, did you change the parameters of the drive to 674x16x38 from within the drive configuration section of DMN in order to be able to boot the server directly from the hard drive? These are only questions that come to mind, and answering them may or may not actually solve anything. They are simply things that I would want to check if I were in your shoes. As far as your questions about VREPAIR; VREPAIR has been known to sometimes cause disastrous results in general, whether or not you are using DMN. VREPAIR's only concern is to get the server to boot again. It is not concerned with salvaging any data. For example, if files are crosslinked or have otherwise invalid DIR or FAT entries, VREPAIR may simply delete them instead of trying to fix the cause of the problem. As far as VREPAIR being a problem when using DMN; With a version of NetWare that you have to link a disk driver into, (everything except ELS level 1), the driver is first modified by DMN to accept a nonstandard drive. This same driver then gets linked into VREPAIR. So as long as you remember to run DMN/I prior to running VREPAIR, (or any other Novell utility), VREPAIR will run as well as it is capable of in the first place. With ELS level 1 versions that come pre-linked, you will have to have v3.11 of DMN, in which case VREPAIR.EXE gets modified to accept a nonstandard drive as well as the rest of the O/S. The main danger is not that unpatched versions of VREPAIR won't accept a nonstandard drive, this would just cause VREPAIR to kick out with a "Invalid ROM parameter table" message and nothing else would happen. The dangerous thing is not remembering to run DMN/I to install the nonstandard disk parameters into memory prior to running VREPAIR in the first place. If you don't run DMN/I, then VREPAIR will make invalid assumptions about the geometry of the disk, and will think that all kinds of things are wrong that really aren't. One particularly dangerous scenario is that if the user attempts to use an unpatched version of VREPAIR, and also forgets to run DMN/I prior to VREPAIR. In this situation, the unpatched version of VREPAIR will not only not understand the true parameters of the disk, it will proceed to try and "fix" the disk anyway instead of bailing out with an "Invalid ROM parameter table" message, because the disk parameters weren't changed by running DMN/I. In this case two wrongs not only don't make a right, they can spell disaster. ------------------------------------------------------------------------- QUESTION: I'm looking to see if you can help me with some information about DM and DM-N. I recently had a situation where a customer trashed a Quantum 40AT hard drive to the point where the partitions were messed up. I normally would have had him low level, fdisk, and high level the drive. The Quantum cannot be low leveled under normal circumstances. Can DM or DM-N low level a Quantum 40AT so that a corrupted drive can be used? The drive that inspired this question was on one of our Slimline computers at a user site. If you can let me know about this it would be a big help. Thanks ANSWER: The current versions of our products, (DM v4.20 and DMN v3.11), fully support the Quantum 40AT. Since the 40AT is an "embedded servo" IDE drive, it is OK to initialize (low-level format) it without fear of damage to the drive. With an "embedded servo" drive, initializing it simply writes zeros to every sector, instead of actually defining and laying out the sectors on the drive as it would with a "normal" drive. Therefore, simply reconstructing the partition map and re-preparing the partitions should be adequate in order to straighten this drive out, (short of an actual hardware failure). COMMENT FROM ANOTHER USER: Regardless of the amount of free space for data on a Novell File Server, you are limited to the Number of Files that may exist on a particular volume. This is set up when you run NetGen. It may be modified (increased or decreased) sometimes without loss of data (Key word Sometimes) by running NetGen again on the file server. I would only do it with a known good backup. The file server will generate the eronious message 'out of space' (or something like that) when it runs out of directory entries even if the volume has plenty of space available. Chkvol will reveal the number of used and available directory entries at any time. Hope this is of some help. ------------------------------------------------------------------------- QUESTION: I have a miniscribe 7080A which I would like to format to its full capacity of 80 MB. The problem is that my CMOS does not support this type of drive. Do you sell a product that will allow me to format my miniscribe to 80 MB? I have a 386 with 1989 AMI BIOS. ANSWER: The Miniscribe 7080 IDE drive is not a universal translation drive. This is to say that it can't translate to just any combination of parameters like some other IDE drives can. It has 4 basic modes of translation that it can do. These are as follows: 10 heads by 17 sectors per track, with cyls in BIOS less than or = 981 6 heads by 33 sectors per track, with cyls in BIOS less than or = 832 9 heads by 17 sectors per track, with cyls in BIOS less than or = 1023 8 heads by 39 sectors per track, with cyls in BIOS less than or = 528 You will either have to have a CMOS drive type with one of these geometries, or create one as a user defined drive type. The selection of 981x10x17 will be the one that will give you the largest capacity, (81.43 meg). The selection of 1023x9x17 would give you 76.43 meg. Since this is an IDE drive we are talking about, you should not initialize, (low-level format), it. All you should need to do is set your CMOS to a drive type that corresponds to one of the above geometries, and then run Disk Manager in automatic mode. When asked if you wish to initialize this drive, answer No. It should go smoothly from there. If your system drive table does not have a drive type that matches one of the above geometries, then our Superproms might be able to provide the 1023x9x17 geometry for you, as well as a 965x10x17 geometry that would give you 80.1 meg of storage. You can find more information about Superproms by doing a TS)text search for the word SUPERPROM from the main menu of this BBS. Also, message number 1319 explains how to determine what kind of Superproms you would need. QUESTION: So, now I know that I need to set a "user defined" drive type in my CMOS. So my original question still stands. Do you have software that will help me do this. I don't know how to set this user defined type. ANSWER: Some computers offer you the ability to manually define the parameters of your drive from within the setup program that came with the computer. Computers that have this ability then store the actual drive parameters in their CMOS memory, rather than just the drive type number as they normally do. This function is called a "user defined drive type", and is either supported by your computer, or isn't. You will have to consult the users manual for your computer in order to find out. If your computer does have this ability, then use the setup program that came with it to define one of the previously mentioned geometries. DM will not be able to do this for you. If your computer does not have this ability, then you have two other options. The first is to select a 1024 cylinder by 9 head by 17 sectors per track drive type, (again, you will have to consult your users manual to determine if your computer has such a drive type entry and if so what drive type number it is), and use Disk Manager to create and prepare the partitions on the drive only. The last option is to look into purchasing Superproms which would provide you with such an entry, if your computer can use them. ------------------------------------------------------------------------- QUESTION: Simple question: Has anybody fixed the problem related to cyl>1024 and/or smartdriv.sys and/or smartdriv.sys with DM ? Has anybody come up with another cacheing prog to use in place of windows 3.0 smartdriv.sys?? I tried hyperdsk410 but still had the drive go down . It has been back up and running windows 3.0 in enhanced but without any cacheing and so far no scrambling. Please don't suggest contacting Microsoft. We see their approach: "....you many experience problems when using certain third party partioning software...". Then they procede to list virturally all third party partioning software! ANSWER: I see that prior to asking your question you read bulletin number 12 that has to do with the Windows/DMDRVR/SMARTDRV situation. The reason that bulletin number 12 exists is so I don't have to go through the whole story every time someone asks about Windows, and I won't. There have been a few minor changes in the situation since the bulletin was written. Ontrack has volunteered to make a few modifications to XBIOS.OVL that will make it unnecessary to place the "virtualHDIRQ=OFF" line in your SYSTEM.INI file when using a disk with more than 1024 cylinders in order to prevent Windows from hanging when attempting run it in 386 enhanced mode. This forthcoming version of XBIOS is Ontracks' contribution to the solution of the overall problem. All of the other conflicts are at this time being addressed by MicroSoft, since it is SMARTDRV's ignorance of the disk environment that is causing the other problems. (I was always taught that "ignorance" is not a bad word in and of itself. It simply means a lack of awareness in a specific area and is easily remedied.) Our engineers have been in close contact with MicroSoft and have given them all the necessary information that they need in order to understand the disk environment and make SMARTDRV a little smarter. The same functionality will have to be incorporated into Windows itself as well in order to make it possible to create a permanent swap file when running Windows on a nonstandard disk. It is my understanding that MicroSoft is currently working on these updates. As far as other caches being available, the answer is yes. "Super PC Qwik" by MultiSoft is one of probably several caches that have been shown to function correctly on a disk with greater than 1024 cylinders, and with Windows in 386 enhanced mode. Thanks for your questions. It gave me the opportunity to bring the current status of the situation out into the light for others to see as well. I know this is going to sound strange, but believe it or not this whole Windows/DMDRVR/SMARTDRV situation has had a bright side. The fact that this conflict happened has forced open some doors between Ontrack and MicroSoft that neither company thought would ever be opened, and that can be nothing but beneficial to both companies, (and the public), in the future. ------------------------------------------------------------------------- QUESTION: Make that 13 confirmed cases of problems with DM and Windows 3.0. I am installing a CONNER 3104 210 meg drive and, after using DM 4.2, and pulling all of my files off tape, everything looked great until I started Windows 3.0. Then it locked up. Thanks for Bulletin 12... I'l try that stuff now. ANSWER: Since your Conner 3104 probably ended up getting configured as having more than 1024 cylinders, you will have to place the "virtualHDIRQ=OFF" line in your SYSTEM.INI file, and not use SMARTDRV as stated in bulletin number 12. ------------------------------------------------------------------------- QUESTION: I was trying to install a CONNER CM3104 104 meg IDE drive in a Packard Bell PB286 (built in Nov 88 with PB Bios ver 3.32p) using Microsoft DOS 4.01. I tried a variety of solutions, yet nothing seemed to work. I talked to CONNER, and Packard Bell as well. The net result was that I haven't been able to install this drive! I finally was able to install a Quantum Pro 80-AT IDE drive in this computer using ONTRACK. I used ONTRACK ver 4.2 for the entire process. DM still required me to partition this drive as one 69 MEG partition, and one 11 MEG partition, even though the bios supports drives of 80 MEG. I tried to install it with one 80 MEG partition, but I get errors such as a reading of a 474 MEG partition in DOS FDISK. Of course, the major problem is that Packard Bell quit developing bios for this particular computer, and the bios does not support user defined drives. Otherwise, DM would not even be necessary. I verified that the CONNER drive was OK, in fact I installed it in a 386 later. Do you have any idea what's going on? I can be reached at (412) 223-4572 (office) or (412) 222-7033 (voice mail). I'd appreciate some help. ANSWER: First of all, when you stated that you were using "Disk Manager v4.2", I assume that you meant v4.20 rather than v4.02. If you do not actually have v4.20, you will need to upgrade your DM in order to install this drive. I will assume that you have v4.20. For the answer to your questions about why you can't have the whole CP3104 as one single partition under DOS 4.x, I will refer you to question #2 of Bulletin #13, (or the TOPTEN.EXE file), which I see that you have already read. This is not a standard drive to your system, therefore the DOS boot partition is restricted to reside within the "standard" portion of your drive. Also, remember that Disk Manager is a REPLACEMENT for FDISK. The two are mutually exclusive. If you are using Disk Manager, you should not ever run FDISK on this machine. At best it will be confused, (and will confuse you), and at worst it could destroy your partition table. When you said that "DOS 4.01 only recognizes this drive as being 76 Meg total capacity", did you mean that FDISK only saw 76 Meg, (this would be understandable and expected), or that CHKDSK only saw 76 Meg total between the two partitions? If you can install the CP3104 in this machine with a 2 Meg & a 102 Meg partition, and can boot the machine and have CHKDSK see the second partition as being 102 Meg, then DM is operating properly. (You will have to patch your 4.x CHKDSK program to recognize a Disk Manager write/read partition first. The procedure for doing this is described in Bulletin #8 on this BBS). If the above is not the case and you are having other problems above and beyond that, then please leave another question stating them. ------------------------------------------------------------------------- QUESTION: I'M HAVING TROUBLE WITH ALL CONNORS IDE DRIVES. I HAVE DM 4.20 AND EVREY THING GOES WELL UNTILL I TRY TO BOOT FROM THE HARD DRIVE. EVERY TIME I GET A "DRIVE BOOT FAILURE" MSG. THIS ONLY HAPPENS WITH CONNORS DRIVES ON 80286 MACHINES. I HAVE NO PROBLEM WITH WD ISP DRIVES EVER. THE CONNORS WILL BOOT PROPERLY ON 386 MACHINES IE SILICONE VALLY MACHINES. CAN YOU HELP? ANSWER: There is no problem in general with Conner IDE drives in a 286 machine. We have helped literally thousands of people install these drives in 286 machines. On the other hand, there may be a problem with your specific brand and model of 286 machine, IDE interface, hard drive, or combination thereof that prevents you from using this particular drive in this particular machine with this particular IDE interface. Or, perhaps the only problem is that you might be using an OEM version of DOS that Disk Manager can't get transferred correctly to the HD. This would cause a "Disk Boot Failure". If you are using a version of DOS that has anybody else's name on it other than MicroSoft, then read question #5 of the Top Ten list, (Bulletin #13 or the TOPTEN.EXE file). If the above suggestion doesn't solve your problem, could you leave me more explicit information on exactly what machine, IDE interface, and Conner drive you are using? Also what version of DOS, how fast is the machine's bus speed, did you use DM's automatic mode, and if you didn't why not? ------------------------------------------------------------------------- QUESTION: I am attempting to install a Seagate ST1239A drive as a network server useing Netware V2.15. The Sequence of installation as written in the Disk-Manager-N user manuals is very vague. It states that the first program to be run is MODGEN. I am installing using a hard drive that is NOT the actual drive that netware will be installed on. If I use netgen to upload all the files required to generate the netware installation, then quit netgen before doing any actual configuration. I then run MODGEN to create the specific file modifications. However, When I then restart netgen and finish the configuration and download the required files, When I run the installation on the system actually containing the ST1239A, I always get the message "ABEND - ROM tables do not match". My question is, is it required that the configuration be run on a machine with an unprepped hard drive using ONLY floppy drives? This seems like a burdensome requirement. Maybe an explanation of the actual sequence of events to install using DMN would save users and installers a lot of time. Also, Why does the DMN program not allow me to use the F3 - select drive function when installing after running MODGEN. The selection is not even on the menu. I need to do that in order to specify the proper drive parameters???? ANSWER: You stated that the message you were getting was "Abend: ROM tables do not match". Was this the exact error message, or was it actually "Abend: Improper ROM parameter table for AT hard disk controller" or something to that affect? If it was the "Improper ROM" message, this means that your Novell disk driver didn't get modified to accept nonstandard parameters correctly. When you ran MODGEN to modify your NetWare, did you select NO to the "Is the disk standard" question? Did it complete without any error messages? Are you using the same diskettes that you downloaded the resulting operating system files to from the other machine? Are you attempting to use a "user defined" CMOS drive type? The absence of the F3 option is probably be due to running MODGEN and getting the Novell disk driver linked into DMN, and then attempting to run DMN standalone, (by just typing in DMN at the DOS prompt), afterwards. If you wish to do this, you will have to use the /V switch in DMN to tell it to not use the driver, (DMN/V). Otherwise, run it from within NetGen and you won't have this problem. It is possible to run MODGEN and NetGen on another system, and then download the generated operating system files back to your floppies and take them over to the machine to be installed. You would then run NetGen on that machine and select "Analyse Disk Surface" from the NetGen menu to run DMN and prepare the hard disk, followed by "NetWare Installation" on the drive. The procedure in this section is the same as it always is with Netware. The only difference is that when you select "Analyse Disk Surface" from NetGen, it calls up DMN instead of COMPSURF. The only problem with this procedure is that in v3.11 of DMN, the DMN files will not be uploaded to the HD or downloaded to the floppies correctly by NetGen. The next version of DMN will correct this. Until then, you can follow the following procedure: 1 Use DISKCOPY to make working copies of all of your NetWare diskettes and your DMN diskette as well. 2 Go to the machine that you will be doing the configuring and linking of the NetWare operating system on, and boot it up with DOS. 3 Put in the DMN diskette and run MODGEN. Select the "Modify" option and then the correct version of NetWare and interface. Answer NO to the "Is the disk standard" question. MODGEN will then prompt you to insert the correct diskettes to modify. 4 Run NETGEN -N and upload the NetWare diskettes to the HD and then exit NetGen without doing anything else. Alternately, if you already have NETWARE subdirectories from past NetGens on this HD, go to the \NETWARE\AUXGEN subdirectory and delete all of the .LNK and .CFG files. This insures that you will get a clean config and link of your NetWare. 5 Go to the \NETWARE subdirectory on the hard disk and create a subdirectory called DMN in it. Copy all of the files from your DMN diskette into the \NETWARE\DMN subdirectory. 6 Go to the \NETWARE subdirectory and run NETGEN\NETGEN -CS, configure and link your operating system and utilities and then download the resulting operating system files back to the floppies. Exit NetGen. 7 Manually copy all of the files from the \NETWARE\DMN subdirectory back to the DMN diskette. 8 Bring the diskettes over to the machine that you will be installing NetWare onto and boot it up with DOS. 9 Put in the NETGEN diskette and run NETGEN -CS. 10 Select "Analyse Disk surface" from the NetGen menu and this will call up DMN. 11 After the disk has been prepared by DMN you will be returned to the NetGen menu. Select "NetWare Installation" from the NetGen menu to install your NetWare. Be careful not to modify the partition table during this process. 12 If your disk has been configured as having more than 1024 cylinders, you will need to prepare a NetWare boot diskette to boot the server with. The AUTOEXEC.BAT file on this diskette will have to call up DMN/I followed by NET$OS. Instructions for creating this diskette are in the HELP program on the DMN diskette. If after following this procedure you are still having problems, let me know and we can go from there. If you need to leave another question, please be very specific in explaining the step-by-step procedure leading up to the problem area. ------------------------------------------------------------------------- QUESTION: installing a conner 104.9 meg drive on a AT-type system dos v3.3 conner drive is 104.9 meg ide type. first drive is a seagate 42meg with the partitioned with fdisk second drive is the conner 104.9 meg. I am trying to setup the conner as 32,32,32,8 meg's each. after I setup the partitions and start to prepair the partitions the first partition is a dos partition and the rest are write/read. after formating (using DISK MANAGER) ,and ending DISK MANAGER, the system can't see drives( D,E,F,G,H). WHAT AM I DOING WRONG ???? Yes,I set the drives as master and slave. ANSWER: The only things I can think of that would be causing this are: 1. If you don't have the files DMDRVR.BIN, XBIOS.OVL, and CONFIG.SYS in the root directory of your boot partition of the hard disk, (C:). 2. Your CONFIG.SYS file doesn't contain the line DEVICE=DMDRVR.BIN. 4. Or if you are not booting from the hard disk. If you are booting from a floppy, the above three files must exist on the floppy. More information about the above files can be found in questions number 5 and 9 in the TOPTEN.EXE file, (or bulletin #13), on this BBS. If this doesn't solve the problem, then leave me another message. If you end up leaving another message, please try to be more specific in describing your hardware layout and DM procedure. -------------------------------------------------------------------------