PDA

View Full Version : Traceback from amcheck



DaveAtFraud
February 5th, 2008, 03:49 PM
Everything appears to be working except amcheck when my tape changer needs to have the tapes changed. That is, amcheck runs just fine if the next tape is found. amdump runs just fine. amtape doesn't seem to have a problem. But when it's time to swap tapes I get:

*** glibc detected *** /usr/sbin/amcheck: double free or corruption (fasttop): 0x0868d998 ***
======= Backtrace: =========
/lib/libc.so.6[0x1edaa6]
/lib/libc.so.6(cfree+0x90)[0x1f0fc0]
/usr/sbin/amcheck[0xd4ce03]
/usr/sbin/amcheck(main+0xc2d)[0xd4e91d]
/lib/libc.so.6(__libc_start_main+0xdc)[0x19cdec]
/usr/sbin/amcheck[0xd49a01]
======= Memory map: ========
0016c000-00185000 r-xp 00000000 09:02 2378079 /lib/ld-2.5.so
00185000-00186000 r-xp 00019000 09:02 2378079 /lib/ld-2.5.so
00186000-00187000 rwxp 0001a000 09:02 2378079 /lib/ld-2.5.so
00187000-002c1000 r-xp 00000000 09:02 2378081 /lib/libc-2.5.so
002c1000-002c3000 r-xp 0013a000 09:02 2378081 /lib/libc-2.5.so
002c3000-002c4000 rwxp 0013c000 09:02 2378081 /lib/libc-2.5.so
002c4000-002c7000 rwxp 002c4000 00:00 0
00564000-00565000 r-xp 00564000 00:00 0 [vdso]
009e8000-00a09000 r-xp 00000000 09:02 4675888 /usr/lib/libamanda-2.5.0p2.so
00a09000-00a0a000 rwxp 00021000 09:02 4675888 /usr/lib/libamanda-2.5.0p2.so
00a0a000-00a29000 rwxp 00a0a000 00:00 0
00a63000-00a83000 r-xp 00000000 09:02 4953736 /usr/lib/libamserver-2.5.0p2.so
00a83000-00a85000 rwxp 0001f000 09:02 4953736 /usr/lib/libamserver-2.5.0p2.so
00a85000-00a87000 rwxp 00a85000 00:00 0
00b2e000-00b3d000 r-xp 00000000 09:02 2378128 /lib/libresolv-2.5.so
00b3d000-00b3e000 r-xp 0000e000 09:02 2378128 /lib/libresolv-2.5.so
00b3e000-00b3f000 rwxp 0000f000 09:02 2378128 /lib/libresolv-2.5.so
00b3f000-00b41000 rwxp 00b3f000 00:00 0
00bcb000-00bd6000 r-xp 00000000 09:02 2378249 /lib/libgcc_s-4.1.2-20070626.so.1
00bd6000-00bd7000 rwxp 0000a000 09:02 2378249 /lib/libgcc_s-4.1.2-20070626.so.1
00c82000-00c85000 r-xp 00000000 09:02 2378258 /lib/libtermcap.so.2.0.8
00c85000-00c86000 rwxp 00002000 09:02 2378258 /lib/libtermcap.so.2.0.8
00c9f000-00ca8000 r-xp 00000000 09:02 2378088 /lib/libnss_files-2.5.so
00ca8000-00ca9000 r-xp 00008000 09:02 2378088 /lib/libnss_files-2.5.so
00ca9000-00caa000 rwxp 00009000 09:02 2378088 /lib/libnss_files-2.5.so
00d48000-00d51000 r-xp 00000000 09:02 4560646 /usr/sbin/amcheck
00d51000-00d52000 rwxp 00008000 09:02 4560646 /usr/sbin/amcheck
00d60000-00d85000 r-xp 00000000 09:02 2378087 /lib/libm-2.5.so
00d85000-00d86000 r-xp 00024000 09:02 2378087 /lib/libm-2.5.so
00d86000-00d87000 rwxp 00025000 09:02 2378087 /lib/libm-2.5.so
00ece000-00ee1000 r-xp 00000000 09:02 2378125 /lib/libnsl-2.5.so
00ee1000-00ee2000 r-xp 00012000 09:02 2378125 /lib/libnsl-2.5.so
00ee2000-00ee3000 rwxp 00013000 09:02 2378125 /lib/libnsl-2.5.so
00ee3000-00ee5000 rwxp 00ee3000 00:00 0
00f46000-00f50000 r-xp 00000000 09:02 4674895 /usr/lib/libamtape-2.5.0p2.so
00f50000-00f51000 rwxp 00009000 09:02 4674895 /usr/lib/libamtape-2.5.0p2.so
08687000-086a8000 rw-p 08687000 00:00 0
b7e00000-b7e21000 rw-p b7e00000 00:00 0
b7e21000-b7f00000 ---p b7e21000 00:00 0
b7f19000-b7f1c000 rw-p b7f19000 00:00 0
b7f25000-b7f27000 rw-p b7f25000 00:00 0
bf8d2000-bf8e8000 rw-p bf8d2000 00:00 0 [stack]
Amanda Tape Server Host Check
-----------------------------
Holding disk /share2/stage: 90535524 kB disk space available, using 90433124 kB


Amanda Backup Client Hosts Check
--------------------------------
Client check: 4 hosts checked in 0.356 seconds, 0 problems found

(brought to you by Amanda 2.5.0p2)

I guess it works in the sense that, if I get a traceback, I know I need to swap tapes. This seems a little inelegant. This is with CentOS 5.1 (RHEL clone) and all updates applied. The tape changer is a:

# cat /proc/scsi/scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 04 Lun: 00
Vendor: HP Model: C1557A Rev: U709
Type: Sequential-Access ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 04 Lun: 01
Vendor: HP Model: C1557A Rev: U709
Type: Medium Changer ANSI SCSI revision: 02

Cheers,
Dave

dustin
February 5th, 2008, 03:57 PM
You may want to try upgrading to 2.5.2 or, if you can wait a week or two, 2.6.0.

DaveAtFraud
February 6th, 2008, 06:37 AM
You may want to try upgrading to 2.5.2 or, if you can wait a week or two, 2.6.0.Thanks, Dustin. Actually, that translates as I should file a Red Hat bug so they know they need to either move forward or backport the fix. See https://bugzilla.redhat.com/show_bug.cgi?id=431688 (which is a cut and paste from my post).

I just went through a number of hardware and software upgrades (described at http://davenjudy.org/wordpress). My first thought was that I had botched something.

Thanks again,
Dave

DaveAtFraud
March 22nd, 2008, 06:56 AM
Problem "went away" as of at least last weekend. No idea which library changed in the distro.

Cheers,
Dave