Hi there. I've had some tape problems, and hopefully have cleared them up.
So now since I've had tape problems, my tapes have old backups, I'm down two tapes and my holding disk (primary) is full. So, I want to run amflush to get things to the tapes, right? I try to run amflush and it seems to be instantly done. I look at the logs (/var/log/amanda/amflush and /var/log/amanda/log) which don't indicate anything useful. I run amflush with -f to get more. Ah ha! A seg fault! That's better - now I know that something is being broken. I can't tell what though. Perhaps an amand guru can enlighten me, or offer some advice on where to look next. Also, after running the amflush, it (amanda) doesn't respond with it's normal email message about finishing the flush. I have to run amcleanup in order to run another flush. And then all the existing backups are STILL on the list. Here's some of what I'm seeing on the screen with amflush -f <conf> Some domain names as indicated in the output have been changed to protect the guilty. backup@backup:~$ /usr/sbin/amflush -f DailySet1 amflush: /var/log/amanda/DailySet1/log exists: amdump or amflush is already running, or you must run amcleanup backup@backup:~$ amcleanup Usage: amcleanup conf backup@backup:~$ amcleanup DailySet1 amcleanup: processing outstanding log file. backup@backup:~$ /usr/sbin/amflush -f DailySet1 Scanning /var/amanda2... lost+found: skipping cruft directory, perhaps you should delete it. Scanning /var/amanda... lost+found: skipping cruft directory, perhaps you should delete it. 20030126: found Amanda directory. 20030127: found Amanda directory. ymir: skipping cruft directory, perhaps you should delete it. 20030209: found Amanda directory. 20030210: found Amanda directory. 20030129: found Amanda directory. 20030130: found Amanda directory. 20030202: found Amanda directory. 20030203: found Amanda directory. 20030204: found Amanda directory. 20030205: found Amanda directory. 20030206: found Amanda directory. 20030207: found Amanda directory. 20030211: found Amanda directory. Multiple Amanda directories, please pick one by letter: A. 20030126 B. 20030127 C. 20030129 D. 20030130 E. 20030202 F. 20030203 G. 20030204 H. 20030205 I. 20030206 J. 20030207 K. 20030209 L. 20030210 M. 20030211 Select directories to flush [A..M]: [ALL] a Today is: 20030211 Flushing dumps in 20030126 to tape drive "/dev/nst0". Expecting tape Daily02 or a new tape. (The last dumps were to tape Daily01) Are you sure you want to do this [yN]? y amflush: datestamp 20030211 Segmentation fault backup@backup:~$ driver: pid 815 executable driver version 2.4.3 driver: send-cmd time 0.002 to taper: START-TAPER 20030211 driver: adding holding disk 0 dir /var/amanda2 size 37528732 driver: adding holding disk 1 dir /var/amanda size 1501432 reserving 9757541 out of 39030164 for degraded-mode dumps driver: start time 0.004 inparallel 6 bandwidth 10000000 diskspace 39030164 dir OBSOLETE datestamp 20030211 driver: drain-ends tapeq LFFO big-dumpers ttt taper: pid 816 executable taper version 2.4.3 taper: page size is 4096 taper: buffer size is 32768 taper: buffer[00] at 0x40215000 taper: buffer[01] at 0x4021d000 taper: buffer[02] at 0x40225000 taper: buffer[03] at 0x4022d000 taper: buffer[04] at 0x40235000 taper: buffer[05] at 0x4023d000 taper: buffer[06] at 0x40245000 taper: buffer[07] at 0x4024d000 taper: buffer[08] at 0x40255000 taper: buffer[09] at 0x4025d000 taper: buffer[10] at 0x40265000 taper: buffer[11] at 0x4026d000 taper: buffer[12] at 0x40275000 taper: buffer[13] at 0x4027d000 taper: buffer[14] at 0x40285000 taper: buffer[15] at 0x4028d000 taper: buffer[16] at 0x40295000 taper: buffer[17] at 0x4029d000 taper: buffer[18] at 0x402a5000 taper: buffer[19] at 0x402ad000 taper: buffer[20] at 0x402b5000 taper: buffer[21] at 0x402bd000 taper: buffer[22] at 0x402c5000 taper: buffer[23] at 0x402cd000 taper: buffer[24] at 0x402d5000 taper: buffer[25] at 0x402dd000 taper: buffer[26] at 0x402e5000 taper: buffer[27] at 0x402ed000 taper: buffer[28] at 0x402f5000 taper: buffer[29] at 0x402fd000 taper: buffer[30] at 0x40305000 taper: buffer[31] at 0x4030d000 taper: buffer structures at 0x40315000 for 384 bytes taper: read label `Daily02' date `20030208' taper: wrote label `Daily02' date `20030211' driver: result time 16.242 from taper: TAPER-OK driver: state time 16.242 free kps: 10000000 space: 39030164 taper: idle idle-dumpers: 6 qlen tapeq: 0 runq: 0 roomq: 0 wakeup: 86400 driver-idle: not-idle driver: interface-state time 16.242 if : free 10000000 driver: hdisk-state time 16.242 hdisk 0: free 37528732 dumpers 0 hdisk 1: free 1501432 dumpers 0 driver: QUITTING time 16.243 telling children to quit driver: send-cmd time 16.243 to taper: QUIT taper: DONE [idle wait: 0.000 secs] taper: writing end marker. [Daily02 OK kb 0 fm 0] driver: FINISHED time 22.714 backup@backup:~$ cat /var/log/amanda/DailySet1/log DISK amflush backup /dev/hda2 DISK amflush backup /dev/hdb1 DISK amflush www /dev/sda2 DISK amflush www /dev/sda4 DISK amflush www /dev/md0 DISK amflush www /dev/md1 DISK amflush skuld.bin /dev/ida/c0d0p5 DISK amflush skuld.bin /dev/ida/c0d0p7 DISK amflush skuld.bin /dev/ida/c0d0p1 DISK amflush skuld.bin /dev/ida/c0d0p2 DISK amflush yggdrasil /dev/ida/c0d0p1 DISK amflush yggdrasil /dev/ida/c0d0p6 DISK amflush yggdrasil /dev/ida/c1d0p1 DISK amflush yggdrasil /dev/ida/c0d0p5 DISK amflush yggdrasil /dev/ida/c0d0p7 DISK amflush misc-shell /dev/hda2 DISK amflush admin-nfs /dev/hda2 DISK amflush admin-nfs /dev/hdc2 DISK amflush misc-shell /dev/hda2 DISK amflush www-apache /dev/hda2 DISK amflush www-apache /dev/hdc1 DISK amflush auth-radius /dev/hda2 DISK amflush auth-radius /dev/hda3 DISK amflush loghost-nfs /dev/hda2 DISK amflush loghost-nfs /dev/hdc1 DISK amflush ns1 /dev/hda2 START amflush date 20030211 START driver date 20030211 STATS driver startup time 0.004 START taper datestamp 20030211 label Daily02 tape 0 INFO taper tape Daily02 kb 0 fm 0 [OK] FINISH driver date 20030211 time 22.714 If you point to things in the Faq-o-matic, would you offer any advice as well on how to get the faq-o-matic to NOT display things that are 10 pages across in the web browser? 'segmentation fault' doesn't have a hit in the faq-o-matic though. TIA, John
