Date: Fri, 11 Sep 1998 20:39:47 -0500 (CDT) From: Greeves ROOT <root@mfn.org> To: FreeBSD-Questions@FreeBSD.ORG Subject: Panic during backups Message-ID: <Pine.BSF.3.96.980911202259.11522A-100000@greeves.mfn.org>
next in thread | raw e-mail | index | archive | help
Greetings... Three days ago we attempted to automate our backups rather then having the night folks do it. A simple shell script (below) is the vehicle. On the first night out, the system ran the job (from cron, as root), and ran out of file desriptors. So I built a new kernel with maxusers=30, and waited for the next night. Next night's backup crashed but left no reason why. Yesterdays backups at least left a trail 8=> This is being retyped, as I couldnt cut/paste it: swap_pager: indefinite wait on buffer: device: 131073, blkno: 6320, size: 8192 swap_pager: indefinite wait on buffer: device: 131073, blkno: 86104, size: 12288 fatal trap 12: page fault while in kernel mode fault virtual address = 0x40 fault code = supervisor write, page not present instruction pointer = 0x8:0xf012eba3 stack pointer = 0x10:0xefbffef4 frame pointer = 0x10:0xefbfff58 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL=0 current process = 2213 (mount) interrupt mask = panic: page fault syncing disks... It may have "synced disks", but there were broken inodes all over the place when it came back up. Also, as an interesting (?) aside, the IP stack was still answering ping and connection requests when we finally put this box out of it's misery... The first pass through the script completed succesfully, wd0a was backed up. wd0s1e was not even started: the email said it couldn't find the file system being backup up to: an nfs on "greeves". Any thoughts? This script seems to work when I use it from the command line, so why is it dying (and taking a machine down with it) when running from cron? TIA, John Blau Asst. Admin. jb214@mfn.org #!/bin/sh # Perfors nightly backups of all filesystems to a remotely # mounted backup server (greeves) # # setup the basics # BACKUPSERVER="greeves" BACKUPROOT="/backup" OLDBUDATE=`echo "/backup/"``date -v-1d '+%y%m%d_'``hostname -s` NEWBUDATE=`echo "/backup/"``date '+%y%m%d_'``hostname -s` # Kludge, what's a kludge :) /sbin/mount greeves:/backup /backup if [ $? != 0 ]; then echo "Mount of greeves:/backup filesystem failed! Aborting." exit 1 fi echo "This job will process the following filesystems:" mount | grep /dev echo " " # # Loop through the fstable, backing up only # those devices that start with "/dev"... # f_iles=`mount | grep /dev | cut -d" " -f1 | sed 's./._.g'` for eachfile in $f_iles do echo " " echo " * Now Processing `hostname -s`:$eachfile *" OLDBU=$OLDBUDATE$eachfile NEWBU=$NEWBUDATE$eachfile if [ -f "$OLDBU" ]; then echo "Old backup file found:" echo "$OLDBU" else echo "WARNING: No previous backup found!" echo " Expected: $OLDBU" fi dump -0auf $NEWBU `echo $eachfile | sed 's._./.g'` #echo "Dumping" if [ $? != 0 ]; then echo "ERROR! BACKUP FAILED: ABORTING!" exit 1 else echo "Backup successful, removing prior backup." #echo "removing any prior backup" rm "$OLDBU" 2> /dev/nul fi done To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.980911202259.11522A-100000>