A *BIG* thanks to the 4 sun managers who took the time to reply. Your help
was much appreciated. Take it away Martin Achilli (martin@gea.hsr.it):
--- Try checking permissions on the file /etc/mnttab, a problem I have had is that whenever you use eitherquotacheck or acquota
the file permissions are reset so that group and world cannot read it. I have in fact added a command in my scripts that weekly check quotas to fix the permissions back so that group and world can read it. ----
In fact, I'd been doing some work on the disk in question and had run a quotacheck a day before the users started to notice the problem. I'd not make the connection however until Martin's message.
Other suggestions were:
-------- From: Ake Sandgren <ake@cs.umu.se> > Sounds like your quota daemon isn't running on server A
I'd already checked this but forgot to add it to the original post.
--------- From: rauthenberg@HHI.DE (Stefan Rauthenberg) > I must admit the following is just a shoot in the dark but under pre 2.5 your > problem was caused by a too restrictive protection of the file > `quotas', which lives in the filesystem having quotas enabled. > >Try to check whether these files have the same protection. > > Under 2.3/2.4 the quota system needs 0644 although the AnswerBook or man > pages may suggest a more restrictive mode.
My quota files are mode 0600 and seem to work fine under solaris 2.5. Although I agree this has been a problem before.
------------ From: Todd Michael Kennedy <tkennedy@phoenix.csc.calpoly.edu> > Make sure you have the following line uncommented in /etc/inetd.conf: > > rquotad/1 tli rpc/datagram_v wait root /usr/lib/nfs/rquotad rquotad > > Also, you'll want to make sure that RPC is working between the two machines. > Look in /var/adm/messages for any errors that may pop-up while remove users > try to view their quota.
A good suggestion, I'd not checked it. But all was well.
Original message follows:
Glenn
_______________________________________________________________________ Dr. Glenn Carver, Phone: +44 (1223) 336524 Centre for Atmospheric Science, Fax: +44 (1223) 336473 Cambridge University, Chemistry Dept., glenn@atm.ch.cam.ac.uk Lensfield Road, Cambridge, CB4 4JZ, UK http://www.atm.ch.cam.ac.uk/~glenn/
"Time is a great teacher but unfortunately it kills all its pupils" Hector Berlioz (1803-69) - French composer. _______________________________________________________________________
>From glenn Wed Mar 13 10:53:02 1996 To: sun-managers@ra.mcs.anl.gov Subject: Problem with quotas under solaris2.5
I've got a problem with user quotas that I cannot figure out.
I have 2 servers, let's call them ServerA and ServerB. For some reason, users on ServerA cannot see their quotas on ServerB but users on ServerB CAN see their quotas on ServerA. Yet, both machines have identical setups for their vfstab, sharetab and dfstab files and I cannot figure out what's going on.
ServerA is our NIS (not NIS+) server whilst ServerB is our DNS server. Both machines have recently been upgraded to Solaris 2.5 (from 4.1.3 where things worked fine). Neither machines are running automount; all filesystems are mounted using /etc/vfstab.
I have checked that quotas are on for all filesystems. I have read through the answerbook and man pages and checked the options in vfstab etc. I've tried deleting and recreating the quotas file in the /home filesystems, to no avail. Suspicion falls on ServerA being the NIS server but as the automounter is not running I fail to see why there should be a problem.
I've given more details of the setup and the output from the quota command below.
If anyone has some ideas as to what the problem is I've most grateful as I've been puzzling over this for days now.
Thanks, Glenn
ServerA ------- Here's 2 entries from /etc/vfstab:
/dev/dsk/c0t0d0s2 /dev/rdsk/c0t0d0s2 /home1 ufs 2 yes rq ServerB:/home6 - /home6 nfs - yes bg,rw,quota
and the relevant entries from /etc/dfs/sharetab:
/home1 - nfs rw,root=ServerB
and the entry from /etc/dfs/dfstab:
share -F nfs -o rw,root=ServerB /home1
and the mount command would tell you:
/home1 on /dev/dsk/c0t0d0s2 setuid/read/write/quota on Thu Feb 29 22:57:24 1996 /home6 on ServerB:/home6 bg/read/write/quota/remote on Wed Mar 13 10:27:46 1996
ServerB ------- Likewise, /etc/vfstab
/dev/dsk/c0t1d0s7 /dev/rdsk/c0t1d0s7 /home6 ufs 2 yes rq ServerA:/home1 - /home1 nfs - yes bg,rw,quota
/etc/dfs/sharetab:
/home6 - nfs rw,root=ServerA
/etc/dfs/dfstab:
share -F nfs -o rw,root=ServerA /home6
and mount says:
/home6 on /dev/dsk/c0t1d0s7 quota/setuid/read/write on Sun Feb 25 20:47:13 1996 /home1 on ServerA:/home1 bg/read/write/quota/remote on Wed Mar 13 10:13:45 1996
Users ----- Now for user 'kirk' whose home directory is on ServerA the output from the quota -v command gives the following on ServerA:
Filesystem usage quota limit timeleft files quota limit timeleft /home1 157026 190000 200000 2452 0 0
but the following on ServerB:
Filesystem usage quota limit timeleft files quota limit timeleft
On the other hand, user 'spock' whose home directory is on ServerB gets the following output from quota -v on ServerB:
Filesystem usage quota limit timeleft files quota limit timeleft /home6 10530 26000 27000 1995 0 0
and when on ServerA manages to get the following output from quota -v:
Filesystem usage quota limit timeleft files quota limit timeleft /home6 10531 26000 27000 1995 0 0
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:10:55 CDT