From: Digestifier To: Linux-Admin@senator-bedfellow.mit.edu Reply-To: Linux-Admin@senator-bedfellow.mit.edu Date: Wed, 1 Sep 93 15:30:53 EDT Subject: Linux-Admin Digest #31 Linux-Admin Digest #31, Volume #1 Wed, 1 Sep 93 15:30:53 EDT Contents: ANNEX communications server installation trouble (Chris Thompson) Re: Net-Admin-Guide CAUTION! (David Gaudine) Re: fsck and partitioning. (Gilbert Callaghan) Re: restricting newsgroups (Stephen Williams) Re: Where's Ross Biro ? (was Re: Let's collect KNOWN BUGS) (Arnt Gulbrandsen) Re: Rnews problem in UUCP (Andreas Klemm) Re: "lpd" won't run (Andreas Klemm) Re: booting .99P12 (new kernel) (Andreas Klemm) SLS 1.03 networking (C Wayne Huling) Re: "lpd" won't run (Brian McCauley) [Q]: eliminating hostname before 'login:' with MCC (Liem Bahneman) Re: Rnews problem in UUCP (Oliver Klink) Result from free (memory usage) (Philip Sinn) Re: Net-Admin-Guide CAUTION! (Hansruedi Heeb) ---------------------------------------------------------------------------- From: thompsch@ccmail.orst.edu (Chris Thompson) Subject: ANNEX communications server installation trouble Date: 31 Aug 1993 21:41:48 GMT Has anyone out there tried to install the annex communication software for release 7.0. I am very new to Unix and Linux so I am not really sure what the hell I am doing, but Iknow someone out there can answer some or one of these questions or just offer someadvice or comment on what they found. First of all it asks during installation what typeof OS that you have and the choices are 1) 4.[23]BSD Unix (&UMAX BSD) 2) System VUnix (& UMAX V) 3) XENIX System V 4) MACH 5) SVR4 ? Then depending on whatyou answer to the above question it will ask for additional loader options for networklibraries, and alternate include directory for network code, and where are the networkinclude files located? Then it will ask what kind of network software does your OS have, 1) 4.[234]BSD or compatible 2) Excelan EXOS package 3) CMC package 4) WollongongWINS package 5) TLI. Transport Layer Interface? If anybody has any idea please e-mail me at Thompsoc@kerr.orst.edu ------------------------------ From: david@donald.concordia.ca (David Gaudine) Subject: Re: Net-Admin-Guide CAUTION! Date: Tue, 31 Aug 1993 22:57:06 GMT In article <25vpb7INNs3o@uwm.edu> rick@ee.uwm.edu (Rick Miller) writes: >Don't waste a ton of paper printing the Net Admin Guide, until you've >inserted the proper scaling commands into the PostScript to make it FIT >on LETTER-sized paper (or print it on LEGAL-size). I didn't have any such problem, using the .ps file. My only problem was that the page numbers don't appear on most pages. -- ========================================================================= David Gaudine, programmer, Concordia University, Loyola Campus, Montreal. david@donald.concordia.ca ------------------------------ From: gilbert@inviso.com (Gilbert Callaghan) Subject: Re: fsck and partitioning. Date: Tue, 31 Aug 1993 21:13:34 GMT In article hjstein@sunrise.huji.ac.il (Harvey J. Stein) writes: > >[Re: bootutils] > >Its README says that it mounts the root filesystem readonly, runs >fsck, and then if there are no problems it remounts the root and >mounts the other filesystems. However, what if there is a problem? >If the root file system has been damaged, how can this fsck fix it >if the root was mounted readonly? Does it write directly to the >device file in /dev? Yes. This is why root is mounted readonly at first, so that no other process can modify the partition while fsck does its thing. >What should I do when fsck finds an error? Nothing usually. fsck should fix it and then /etc/rc continues as normal. >So, after a crash, if there's a disk problem, the system will drop >immediately into a shell and tell me to reboot. Wrong. If there's a crash, fsck will fix it and continue. The only time you get dumped into a shell is if *fsck* failes to fix the problem. I've never seen this happen, and I guess only happens if your filesystem is hopelessly ruined. -- Gilbert Callaghan gilbert@inviso.com ------------------------------ From: sdw@meaddata.com (Stephen Williams) Subject: Re: restricting newsgroups Date: 31 Aug 1993 23:55:28 GMT Olaf Titz (s_titz@ira.uka.de) wrote: .. : You could maintain two active files with different access permissions : and two different newsreaders, respectively (difficult); or set the .. : > You could try changing the directory "group ownerships" of the spool : > directories you want to restrict. Then put your users in appropriate : > groups and set the read permissions of those spool directories : This will fail because newsreader programs are often setuid or setgid : to 'news'. A different approach would be to maintain two different : versions of them, or two different front-ends which are setuid, check : the effective uid and ecxecute the real newsreader in setuid mode or : not, accordingly. I've been thinking of one possibility that should be safe and still not cause a headache with keeping access permissions up to date: Run all users in a custom chroot environment where the user is put into one of a number of standard environments and possibly RO nfs-mount certain directory subtrees. Symbolic links could be used as needed to 'put-together' the proper structure. Linux does have chroot, right? I'm pretty sure it has something like it, but I didn't find it the last time I perused quickly. Can anyone email me definitively to put me at ease... Mounting RO would prevent tampering with /etc, etc. In any case, this would be for bbs style logins. sdw -- Stephen D. Williams Local Internet Gateway Co.; SDW Systems 513 496-5223APager LIG dev./sales Internet: sdw@world.std.com CIS 76244.210@compuserve.com OO R&D Source Dist. By Horse: 2464 Rosina Dr., Miamisburg, OH 45342-6430 GNU Support ICBM: 39 34N 85 15W I love it when a plan comes together ------------------------------ From: agulbra@nvg.unit.no (Arnt Gulbrandsen) Subject: Re: Where's Ross Biro ? (was Re: Let's collect KNOWN BUGS) Date: 1 Sep 1993 04:25:00 GMT In article <1993Aug31.203545.738@sol.cs.wmich.edu>, Brian Stempien wrote: >Well, since this is the question I sent in about 2 months ago, I think I >should point out to you that I wasn't asking the typical swap-file question. I cheated to get a better example. >Whether I phrased it correctly or not, I was trying to get a pointer on how >to change the Linux kernel so that it COULD use more swap. I quote from my reply: ``It's not easy to use a >16MB partition since that contains >4096 pages, and the kernel doesn't like >4k arrays.'' That seemed detailed enough for the question: >Summary: I have a 20 meg swap partition and need linux to use it all >Description: I once saw an article explaining how to increase the arbitary > 16Meg swap partition limit. I unfortunatly didn't save it. >The point to my post: You should actually read the text of the questions, >maybe the service would be used more if more of the answers were useful. Not guilty. I am guilty of wasting bandwith with this post, however. If I were less vain I would have replied via email, but I don't want you (the readers of col.admin) to think that I completely missed the point of his question. -- Arnt Gulbrandsen, agulbra@nvg.unit.no The other .sig is the real one - this one is just a decoy. -- Arnt Gulbrandsen, agulbra@nvg.unit.no The other .sig is the real one - this one is just a decoy. ------------------------------ Crossposted-To: comp.os.linux.help From: andreas@knobel.knirsch.de (Andreas Klemm) Subject: Re: Rnews problem in UUCP Date: Wed, 1 Sep 1993 13:31:48 GMT root@hip-hop.suvl.ca.us (Remco Treffkorn) writes: >andreas@knobel.knirsch.de (Andreas Klemm) writes: >: swwynen@undergrad.math.uwaterloo.ca (Steve W. Wynen) writes: >: >: >Hello Everybody; >: >: >Hope you can help. I am trying to set up a UUCP site to receive email >: >and news. i have got the UUCP faq, the recommended reading, and this >: >one still escapes me. >: >: Why don't you say what kind of Linux version you are running ? >: Would be fine, too, if you would mention, which uucp version you >: run ... >: >: >I have the UUCP feed mostly working, can send and receive email, but >: >i cannot get my news to come in properly, it gets sent over , >: >but then seems to dissappear into the bit bucket. >: >In the file /usr/spool/uucp/.Log/joesys I get many many messages like: >: >: Strange, which kind of uucp logginf report do you use ? >: SLS comes with HDB and taylor compiled in, but is setup to use >: HDB kind (Honey Dan Beer) of config/spooling/logging .... >: >: >Executing X.joesysX0ctr (rnews) >: >ERROR: Execution: Exit Status 1 >: >Execution failed (X.joesysX0ctr) >: >: Could it be the case, that your rnews isn't found by uux ? >: Check your local config files in /usr/local/lib/uucp >: (Permissions is a good start) if your news feed is permitted >: to run the rnews program .... >: >: >I have done everything that the Cnews README said to do but, it doesnt work. >: >Any suggestions would be appreciated. >: >: Why can't your newsfeed help ?! He has obviously a running system ?! >: I think you aren't experienced in configuring uucp/news/mail ?! >: Then your newsfeed should help you in this - general - points >: of questions .... >: -- >: /-\ Andreas Klemm +-----------------+ >: |@|########################################################-@ "pay for it !" | >: \-/ 41469 Neuss Germany phone +49/ 2137 12609 +-----------------+ >Lighten up Andreas :-) Well, well ... why not ;-) >You are making many assumptions about what Steve did not do to get help >elsewhere. "Der moralische Zeigefinger" does not help! Uh is that something indecent (iiIi) ? ;-) >In virtually all the cases I have seen in the past the permissions were >indeed the culprit. I have seen the exact symptoms Steve mentioned and it >took me three days to find the problem. Turned out to be bad *permissions* >on ..... taataa : /tmp. After restoring my root partition from tape I did >not have 1777 any more. The only symptom: no news! Well, maybe there would >have been others, but I like to login as root. (THIS IS *MY* 'PUTER!) Oh, that's another story .... >The return status of 1 from rnews suggests that rnews was found an run. >Check the scripts. So it is almost guaranteed the permissions. But there >are so many, ach ... ;-) Ok, I overlooked it. What about compiling taylor uucp and cnews complete new. Then you learn the most ..... When configuring the sources and so on, you get a better idea of what the package is doing ... Otherwise ....make a backup of your already configured files ... and install cnews and uucp completely new ... read the faq ... And everything should run. BTW, I'm using cnews from SLS 1.01 and compiled taylor uucp 1.04 completely knew, because I upgraded gcc to 2.4.5, gas to 2.2.1 and the shared libs to 4.4.1 ... -- /-\ Andreas Klemm +-----------------+ |@|########################################################-@ "pay for it !" | \-/ 41469 Neuss Germany phone +49/ 2137 12609 +-----------------+ ------------------------------ From: andreas@knobel.knirsch.de (Andreas Klemm) Subject: Re: "lpd" won't run Date: Wed, 1 Sep 1993 13:11:12 GMT chris@chrism.demon.co.uk (Chris Marriott) writes: >I've just installed the SLS 1.03 linux implementation on my 486/33. I'm >trying to get printing working, following the instructions in the FAQ. >I have a printer on LPT1: which works - I can copy a file to /dev/lp1 and >it prints fine. >My problem is that the printer daemon "lpd" refuses to run. If I run it, >it terminates with no error messages immediately (exit code 0, if that >helps). When I try to print through "lpr" I get the message "job queued, >but no daemon present". When I run the printer control program lpc and >give the "restart" command I get the message: > lp: > cannot open lock file ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > lp: > lpc: connect: No such file or directory > couldn't start daemon >I've edited the "printcap" file to simply say: > lp:\ > :lp=/dev/lp1:\ > :sd=/usr/spool/lp1:\ > :mx#0: >I've create the directory "/usr/spool/lp1". Since the lpd program doesn't run under a certain "meta" user id lp or daemon, it runs under the user id of the user, who initiated the print job .... Therefore each user needs permission for reading and writing in the spool directory. My spool directory - as predefined in the SLS 1.01 installation - has the permissions rwxrwxrwx like /tmp. Do a chmod 777 /usr/spool/lp1, and it should work ... -- /-\ Andreas Klemm +-----------------+ |@|########################################################-@ "pay for it !" | \-/ 41469 Neuss Germany phone +49/ 2137 12609 +-----------------+ ------------------------------ From: andreas@knobel.knirsch.de (Andreas Klemm) Subject: Re: booting .99P12 (new kernel) Date: Wed, 1 Sep 1993 13:46:02 GMT jhenders@jonh.wimsey.bc.ca (John Henders) writes: >drew@kinglear.cs.colorado.edu (Drew Eckhardt) writes: >>... GCC 2.4.5 did nicely with a 16M limit in place, but memory >>usage was still frightening, not to mention the size of the .s files >>it spewed without using the -pipe option! >>Try -pipe, you should see some improvement since disk access is >>reduced because cpp and assembler intermediate files are never >>committed to disk. > I just tried using pipe on an 8 meg machine, while using nn, with >just the normal sendmail and selection daemons running. Under this >setup, using gcc 2.4.5 and compiling 99pl12, it was a tossup on whether >-pipe did any good at all. Real time was 35 minutes for -pipe and 34 >without. Free showed 1.5 meg swapped out compared to 800-1meg without >-pipe. > As a comparison, compiling 99pl9 with gcc2.3.3 would take 20-25 >minutes on the same amchine and didn't go into swap at all. With -pipe on a 486/33 Eisa 16MB AHA 1542B Maxtor 8760S (16ms) 939.95u 151.54s 19:24.14 93.7% -- /-\ Andreas Klemm +-----------------+ |@|########################################################-@ "pay for it !" | \-/ 41469 Neuss Germany phone +49/ 2137 12609 +-----------------+ ------------------------------ From: wayne@rose.cs.odu.edu (C Wayne Huling) Subject: SLS 1.03 networking Date: 01 Sep 1993 14:45:28 GMT I have upgraded from SLS 1.01 to SLS 1.03 because it seemed to make a lot of changes that I was making manually. Anyhow, the net-2 software doesn't seem to communicate with the rest of the network at any level? Does anyone have any suggestions as to why the network is unreachable from the newly upgraded machine, but the other machines on the network that are still SLS 1.01 can ping the upgraded machine? No other network communication works (telnet, ftp, rsh, rlogin). Wayne ------------------------------ From: bam@wcl-l.bham.ac.uk (Brian McCauley) Subject: Re: "lpd" won't run Date: 01 Sep 1993 17:13:10 GMT Reply-To: B.A.McCauley@bham.ac.uk In article <1993Sep1.131112.1571@knobel.knirsch.de> andreas@knobel.knirsch.de (Andreas Klemm) writes: Since the lpd program doesn't run under a certain "meta" user id lp or daemon, it runs under the user id of the user, who initiated the print job .... Therefore each user needs permission for reading and writing in the spool directory. My spool directory - as predefined in the SLS 1.01 installation - has the permissions rwxrwxrwx like /tmp. Wrong. lpd is always run as root. If you want others to be able to start it make it suid. And the spool directories should not be public since lpr is sgid. The spool directories should have group write permission and should belong to the same group as lpr (usually group=lp). See printing-FAQ, or printing-HOWTO or whatever we're calling it this week :-) BTW /tmp should be rwxrwxrwt not rwxrwxrwx unless you want non-priviledged users to be able to delete each other's files. -- \\ ( ) No Bullshit! | Email: B.A.McCauley@bham.ac.uk . _\\__[oo from | Voice: +44 21 471 3789 (home) .__/ \\ /\@ /~) /~[ /\/[ | +44 21 627 2171 (work) . l___\\ /~~) /~~[ / [ | Fax: +44 21 627 2175 (work) # ll l\\ ~~~~ ~ ~ ~ ~ | Snail: 197 Harborne Lane, B29 6SS, UK ###LL LL\\ (Brian McCauley) | ICBM: 52.5N 1.9W ------------------------------ From: roland@cac.washington.edu (Liem Bahneman) Subject: [Q]: eliminating hostname before 'login:' with MCC Date: 1 Sep 1993 17:30:04 GMT How do I eliminate the 'hostname login:' and restore to just 'login:' on incoming telnet connections? MCC's getty doesn't have an /etc/gettydefs or /etc/gettytab file, so I'm wondering where do I change this and how? I find this layout rather annoying and would like the regular: login: Password: Any ideas on how to set this up? (I've also tried getty_pos with no luck, as it doesn't mention ttyp*) - liem -- =======[roland@cac.washington.edu]=====[The Last Gunslinger]================== Outside of a dog, computers are a man's best | UCS Consulting friend, inside a dog it's too dark to type. | UW Ice Hockey Co-captain =============================================================================== ------------------------------ From: oliver@oktagon.rd.sub.org (Oliver Klink) Crossposted-To: comp.os.linux.help Subject: Re: Rnews problem in UUCP Reply-To: oliver@oktagon.rd.sub.org Date: Tue, 31 Aug 1993 19:52:02 CEST In <1993Aug30.130736.2319@knobel.knirsch.de>, Andreas Klemm writes: >>Executing X.joesysX0ctr (rnews) >>ERROR: Execution: Exit Status 1 >>Execution failed (X.joesysX0ctr) > >Could it be the case, that your rnews isn't found by uux ? >Check your local config files in /usr/local/lib/uucp >(Permissions is a good start) if your news feed is permitted >to run the rnews program .... rnews has to be in the PATH. That was a problem I had with my SLS 1.02, but I think the log file would tell something about "rnews: permission denied" in that case... "Exit Status 1" looks more like wrong file/directory permissions somewhere... Bye, Oliver -- # Oliver Klink, Pommernweg 8 # eMail: oliver@oktagon.rd.sub.org # # D 24782 Buedelsdorf, FRG # Don't like my postings? Dial 555-EAT-SHIT. # ------------------------------ From: sinn@carson.u.washington.edu (Philip Sinn) Subject: Result from free (memory usage) Date: 1 Sep 1993 19:00:53 GMT My system memory has just run out when I running a image processing program. So, I exit the program and check the avaiable memory by using the program free. My memory configuration 16Mb memory, 8Mb swapfile (I don't have a swap partition). The report says: total used free shared buffers Mem: 15268 8604 6664 5568 3340 Swap: 8188 356 7832 I am using X at that time, so I run a ps to check for memory usages. The output are: USER PID %CPU %MEM SIZE RSS TT STAT START TIME COMMAND root 1 0.0 0.7 48 120 ? S 10:29 0:00 init philip 2109 0.0 2.1 272 324 1 S 11:38 0:00 sh /usr/bin/X11/startx philip 35 0.0 2.2 304 344 1 S 10:29 0:04 -bash root 4 0.0 0.1 5 28 ? S 10:29 0:01 /etc/update root 36 0.0 0.6 54 104 2 S 10:29 0:00 /bin/getty 9600 tty2 root 12 0.0 0.1 64 24 ? S 10:29 0:00 /etc/lpd philip 2110 0.0 1.5 52 240 1 S 11:38 0:00 xinit /home/philip/.x root 37 0.0 0.6 54 104 3 S 10:29 0:00 /bin/getty 9600 tty3 root 28 0.0 0.7 53 112 ? S 10:29 0:00 /usr/etc/syslogd root 30 0.0 0.3 56 56 ? S 10:29 0:00 /usr/etc/inetd root 2111 4.0 9.6 2051 1468 ? R 11:38 0:28 X root 38 0.0 0.6 54 104 4 S 10:29 0:00 /bin/getty 9600 tty4 root 39 0.0 0.6 54 104 5 S 10:29 0:00 /bin/getty 9600 tty5 root 40 0.0 0.6 54 104 6 S 10:29 0:00 /bin/getty 9600 tty6 philip 2113 0.0 2.0 272 320 1 S 11:38 0:00 sh /home/philip/.xini root 2115 0.8 6.3 384 968 1 S 11:38 0:05 /usr/bin/X11/xterm -r philip 2116 0.0 4.5 80 688 1 S 11:38 0:00 /usr/bin/X11/xclock - root 2117 0.0 6.2 392 956 1 S 11:38 0:00 /usr/bin/X11/xterm -r root 2118 0.0 6.3 392 972 1 S 11:38 0:00 /usr/bin/X11/xterm -r philip 2119 0.0 4.9 418 756 1 S 11:38 0:00 /usr/bin/X11/twm philip 2120 0.0 2.6 301 404 p0 S 11:38 0:00 bash philip 2121 0.0 2.6 303 404 p2 S 11:38 0:00 bash philip 2122 0.0 2.6 301 404 p1 S 11:38 0:00 bash philip 2135 0.1 1.4 105 220 p1 S 11:38 0:01 xc -l /dev/modem philip 2136 0.5 1.4 105 228 p1 S 11:38 0:03 xc -l /dev/modem philip 2144 0.0 1.4 69 224 p0 R 11:50 0:00 ps aux The total RSS is 9780. If I substract ps aux process, 224, it comes to 9556. So by simple arithmetic, it has exceed the used memory number reported by free. Why? Also, for the swap file, I run ps axmp, the results are as follows: PID TT MAJFLT MINFLT TRS DRS SIZE SWAP RSS SHRD LIB DT COMMAND 1 ? 76 65 2 10 79 7 72 62 63 3 init 2109 1 11 95 32 22 126 0 126 104 76 4 sh /usr/bin/X11/ 35 1 168 3573 36 25 138 8 130 105 73 4 -bash 4 ? 7 16 0 4 56 1 55 52 52 1 /etc/update 36 2 24 61 3 8 75 4 71 63 64 4 /bin/getty 9600 12 ? 7 43 1 1 74 24 50 48 48 0 /etc/lpd 2110 1 28 69 2 24 105 0 105 81 92 13 xinit /home/phil 37 3 15 65 3 8 75 4 71 63 64 4 /bin/getty 9600 28 ? 8 50 2 11 82 12 70 57 63 6 /usr/etc/syslogd 30 ? 7 36 1 6 75 17 58 51 53 2 /usr/etc/inetd 2111 ? 189 488 145 203 430 0 430 244 93 11 X 38 4 15 65 3 8 75 4 71 63 64 4 /bin/getty 9600 39 5 15 65 3 8 75 4 71 63 64 4 /bin/getty 9600 40 6 15 64 3 8 75 4 71 63 64 4 /bin/getty 9600 2113 1 5 166 31 22 125 0 125 103 76 4 sh /home/philip/ 2115 1 115 269 20 112 287 0 287 175 194 39 /usr/bin/X11/xte 2116 1 69 187 1 62 216 0 216 155 194 41 /usr/bin/X11/xcl 2117 1 43 274 16 114 284 0 284 171 193 39 /usr/bin/X11/xte 2118 1 49 279 19 114 288 0 288 174 194 39 /usr/bin/X11/xte 2119 1 63 202 25 102 239 0 239 141 133 21 /usr/bin/X11/twm 2120 p0 11 194 40 32 145 0 145 113 79 6 bash 2121 p2 10 138 40 32 145 0 145 113 79 6 bash 2122 p1 11 145 40 32 145 0 145 113 79 6 bash 2135 p1 27 54 9 25 101 0 101 95 78 11 xc -l /dev/modem 2136 p1 0 5 9 25 101 0 101 93 78 11 xc -l /dev/modem 2148 p0 13 54 2 23 96 0 96 70 80 9 ps axmp The total SWAP is only 89 which is must less than the figure in free for swap: 356. So I am wondering, is there a problem with the free program, or did I miss something? Sorry for the long post. Philip Sinn sinn@carson.u.washington.edu University of Washington ------------------------------ From: heeb@watson.ibm.com (Hansruedi Heeb) Subject: Re: Net-Admin-Guide CAUTION! Date: Wed, 1 Sep 1993 18:43:11 GMT In article <25vpb7INNs3o@uwm.edu> rick@ee.uwm.edu (Rick Miller) writes: >Don't waste a ton of paper printing the Net Admin Guide, until you've >inserted the proper scaling commands into the PostScript to make it FIT >on LETTER-sized paper (or print it on LEGAL-size). > >I don't have a LEGAL-sized tray, so I insterted the line: > > 1.0 0.9 scale > >immediately after each and every one of the 240-odd "%%Page ### ###" lines. > >That scales it down to LETTER size. You'll hardly notice the distortion. > If your printer can print on both sides of the paper you must also add an empty page near the beginning or you will get odd page numbers on the left end even ones on the right. Hansruedi ------------------------------ ** FOR YOUR REFERENCE ** The service address, to which questions about the list itself and requests to be added to or deleted from it should be directed, is: Internet: Linux-Admin-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.admin) via: Internet: Linux-Admin@NEWS-DIGESTS.MIT.EDU Linux may be obtained via one of these FTP sites: nic.funet.fi pub/OS/Linux tsx-11.mit.edu pub/linux sunsite.unc.edu pub/Linux End of Linux-Admin Digest ******************************