|Summary:||Quotas fail for some users on a 3000+ users raid0 volume|
|Product:||[Retired] Red Hat Linux||Reporter:||Christian Tosta <tosta>|
|Component:||quota||Assignee:||Steve Dickson <steved>|
|Status:||CLOSED CURRENTRELEASE||QA Contact:||Brock Organ <borgan>|
|Fixed In Version:||Doc Type:||Bug Fix|
|Doc Text:||Story Points:||---|
|Last Closed:||2004-01-12 12:28:34 UTC||Type:||---|
|oVirt Team:||---||RHEL 7.3 requirements from Atomic Host:|
Description Christian Tosta 2003-02-10 18:35:04 UTC
From Bugzilla Helper: User-Agent: Mozilla/5.0 (compatible; Konqueror/3; Linux) Description of problem: I'm implementating quotas for users in a Samba 2.2.7 PDC for Rwindows XP clients. When I run (for example) 'setquota -u 9911054 20000 20000 0 0 /home' the command don't return errors. But when I do a 'quota username' then return "Disk quotas for user #9911054 (uid 9911054): none". My /home is mounted from /dev/md1 which is a raid0 device spanned into two discs and formatted with ext2 filesystem. I think is problem the name of the user (a numeric code)?? Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1.Create a numeric username for a new user 2.Set the quota for this user in volume /home (raid0) 3.Check the quota for this user Actual Results: Quota is "none" when I check quota for user Expected Results: Quota must to be the value I set to user Additional info:
Comment 1 Christian Tosta 2003-02-12 19:54:17 UTC
Problem occur when you try to setup disc quotas for users with a numeric username (not specially on raid devices). You need alfa-numeric usernames to quota works, otherwise it choose random users to set the quota.
Comment 2 Need Real Name 2003-05-12 22:15:01 UTC
Weellll - None of my usernames are entirely numeric I'm using redhat 8 with quotas on a new 120G ext3 file system and as I "rsync -a" files over I notice some usernames aren't showing up in repquota Once I get a couple of thousand users transferred repquota starts segfaulting [It's not really a backup, don't worry about the mountpoint name] [root@backup users]# ls -l /backup/aquota.* -rw------- 1 root root 8192 May 12 23:22 /backup/aquota.group -rw------- 1 root root 149504 May 12 23:22 /backup/aquota.user [root@backup users]# repquota /backup *** Report for user quotas on device /dev/hdc1 Block grace time: 7days; Inode grace time: 7days Block limits File limits User used soft hard grace used soft hard grace ---------------------------------------------------------------------- Segmentation fault Is the the same fault? It gets worse as the number of users goes up
Comment 3 Need Real Name 2003-05-13 20:29:46 UTC
Actually the segfault stops when I used the -n flag to repquota But still not all user-id's show up.
Comment 4 Mark J. Cox 2004-01-05 09:38:53 UTC
have you tried this with a later distribution (RHEL 3 or RHL 9)?
Comment 5 Need Real Name 2004-01-05 10:34:35 UTC
I've not, its not a box we can update easily. We since had such bad problems of users reportedly using a large quantity of disk space when a "find" for that user reports otherwise that we had to rebuild quotas entirely.