Re: Potential Bug
Hi Rhys,
It's a bug, can you try the attached patch?
Jean-Louis
Rhys Rhaven wrote:
Using amanda 3.2.0 on everything. I have 2 hosts and a server
currently. 1 host and the server are ubuntu 10.04 and the other host
(that I'm having issues with) is 8.04.
I can run amdump for each host individually with no problem. Running
amdump for the config (daily) will fail with the following. Talking to
djmitche in IRC, told me to post to the list. My tapes are 200GB, so
they shouldn't be filling up. I'm using vtapes and no holding disks.
FAILURE DUMP SUMMARY:
asterisk-it0.mydomain.com / RESULTS MISSING
asterisk-it0.mydomain.com / lev 1 FAILED [can't dump in degraded mode]
diff --git a/client-src/sendbackup-gnutar.c b/client-src/sendbackup-gnutar.c
index 3b718ae..5e423c7 100644
--- a/client-src/sendbackup-gnutar.c
+++ b/client-src/sendbackup-gnutar.c
@@ -575,6 +575,7 @@ start_backup(
g_ptr_array_add(argv_ptr, stralloc("."));
}
g_ptr_array_add(argv_ptr, NULL);
+sleep(16);
dumppid = pipespawnv(cmd, STDIN_PIPE, 0,
&dumpin, &dumpout, &mesgf,
(char **)argv_ptr->pdata);
diff --git a/client-src/sendbackup.c b/client-src/sendbackup.c
index dacb961..80bc5e6 100644
--- a/client-src/sendbackup.c
+++ b/client-src/sendbackup.c
@@ -652,6 +652,7 @@ main(
} else {
safe_fd(3, 1);
}
+sleep(16);
execve(cmd, (char **)argv_ptr->pdata, safe_env());
exit(1);
break;
diff --git a/server-src/driver.c b/server-src/driver.c
index 9a0bb27..8bf5119 100644
--- a/server-src/driver.c
+++ b/server-src/driver.c
@@ -1044,6 +1044,7 @@ allow_dump_dle(
enqueue_disk(&directq, diskp);
diskp->to_holdingdisk = HOLD_NEVER;
}
+ if (empty(*rq)) force_flush = 1;
}
} else if (client_constrained(diskp)) {
free_assignedhd(holdp);
@@ -1326,6 +1327,7 @@ start_some_dumps(
dumper_cmd(dumper, PORT_DUMP, diskp, NULL);
}
diskp->host->start_t = now + 15;
+ if (empty(*rq)) force_flush = 1;
if (result_argv)
g_strfreev(result_argv);
@@ -3137,6 +3139,7 @@ read_schedule(
log_add(L_WARNING, _("WARNING: got empty schedule from planner"));
if(need_degraded==1) start_degraded_mode(&runq);
schedule_done = 1;
+if (empty(runq)) force_flush = 1;
start_some_dumps(&runq);
startaflush();
}
Re: Potential Bug
Sorry bout that. amdump: start at Fri Nov 5 00:00:01 CDT 2010 amdump: datestamp 20101105 amdump: starttime 2010110501 amdump: starttime-locale-independent 2010-11-05 00:00:01 CDT driver: pid 3102 executable /usr/libexec/amanda/driver version 3.2.0 planner: pid 3101 executable /usr/libexec/amanda/planner version 3.2.0 planner: build: VERSION="Amanda-3.2.0" planner:BUILT_DATE="Mon Oct 18 09:20:30 PDT 2010" BUILT_MACH="" planner:BUILT_REV="3545" BUILT_BRANCH="3_2_0" planner:CC="i486-linux-gnu-gcc" planner: paths: bindir="/usr/bin" sbindir="/usr/sbin" planner:libexecdir="/usr/libexec" planner:amlibexecdir="/usr/libexec/amanda" mandir="/usr/share/man" planner:AMANDA_TMPDIR="/tmp/amanda" planner:AMANDA_DBGDIR="/var/log/amanda" CONFIG_DIR="/etc/amanda" planner:DEV_PREFIX="/dev/" RDEV_PREFIX="/dev/r" planner:DUMP="/sbin/dump" RESTORE="/sbin/restore" VDUMP=UNDEF planner:VRESTORE=UNDEF XFSDUMP="/sbin/xfsdump" planner:XFSRESTORE="/sbin/xfsrestore" VXDUMP=UNDEF VXRESTORE=UNDEF planner:SAMBA_CLIENT="/usr/bin/smbclient" GNUTAR="/bin/tar" planner:COMPRESS_PATH="/bin/gzip" UNCOMPRESS_PATH="/bin/gzip" planner: LPRCMD=UNDEF MAILER=UNDEF planner:listed_incr_dir="/var/lib/amanda/gnutar-lists" planner: defs: DEFAULT_SERVER="localhost" DEFAULT_CONFIG="DailySet1" planner:DEFAULT_TAPE_SERVER="localhost" DEFAULT_TAPE_DEVICE="" planner:NEED_STRSTR AMFLOCK_POSIX AMFLOCK_FLOCK AMFLOCK_LOCKF planner:AMFLOCK_LNLOCK SETPGRP_VOID ASSERTIONS AMANDA_DEBUG_DAYS=4 planner:BSD_SECURITY USE_AMANDAHOSTS CLIENT_LOGIN="amandabackup" planner:CHECK_USERID HAVE_GZIP COMPRESS_SUFFIX=".gz" planner:COMPRESS_FAST_OPT="--fast" COMPRESS_BEST_OPT="--best" planner:UNCOMPRESS_OPT="-dc" READING CONF INFO... planner: timestamp 2010110501 driver: tape size 20480 reserving 0 out of 0 for degraded-mode dumps planner: tape_length is set from tape length (20480 KB) * runtapes (1) == 20480 KB planner: time 0.000: startup took 0.000 secs SENDING FLUSHES... ENDFLUSH SETTING UP FOR ESTIMATES... planner: time 0.000: setting up estimates for asterisk-it0.mydomain.com:/ setup_estimate: asterisk-it0.mydomain.com:/: command 0, options: none last_level 0 next_level0 6 level_days 0getting estimates 0 (-3) 1 (-3) -1 (-3) planner: time 0.001: setting up estimates for git.mydomain.com:/ driver: started dumper0 pid 3104 driver: send-cmd time 0.003 to dumper0: START 2010110501 driver: started dumper1 pid 3105 driver: send-cmd time 0.003 to dumper1: START 2010110501 driver: started dumper2 pid 3106 driver: send-cmd time 0.004 to dumper2: START 2010110501 driver: started dumper3 pid 3107 driver: send-cmd time 0.004 to dumper3: START 2010110501 driver: send-cmd time 0.004 to taper: START-TAPER worker0 2010110501 driver: start time 0.004 inparallel 4 bandwidth 2048000 diskspace 0 dir OBSOLETE datestamp 2010110501 driver: drain-ends tapeq FIRST big-dumpers sssS driver: state time 0.004 free kps: 2048000 space: 0 taper: idle idle-dumpers: 4 qlen tapeq: 0 runq: 0 roomq: 0 wakeup: 0 driver-idle: not-idle driver: interface-state time 0.004 if default: free 2048000 driver: hdisk-state time 0.004 setup_estimate: git.mydomain.com:/: command 0, options: none last_level 0 next_level0 6 level_days 0getting estimates 0 (-3) 1 (-3) -1 (-3) planner: time 0.006: setting up estimates took 0.005 secs GETTING ESTIMATES... dumper: pid 3105 executable dumper1 version 3.2.0 dumper: pid 3104 executable dumper0 version 3.2.0 dumper: pid 3106 executable dumper2 version 3.2.0 dumper: pid 3107 executable dumper3 version 3.2.0 driver: state time 0.326 free kps: 2048000 space: 0 taper: idle idle-dumpers: 4 qlen tapeq: 0 runq: 0 roomq: 0 wakeup: 0 driver-idle: not-idle driver: interface-state time 0.326 if default: free 2048000 driver: hdisk-state time 0.326 driver: result time 0.326 from taper: TAPER-OK worker0 planner: time 6.213: got partial result for host asterisk-it0.mydomain.com disk /: 0 -> -3K, 1 -> -3K, -1 -> -3K planner: time 8.885: got partial result for host git.mydomain.com disk /: 0 -> -3K, 1 -> -3K, -1 -> -3K planner: time 12.423: got partial result for host git.mydomain.com disk /: 0 -> 1042740K, 1 -> -3K, -1 -> -3K planner: time 13.016: got partial result for host git.mydomain.com disk /: 0 -> 1042740K, 1 -> 15570K, -1 -> -3K planner: time 13.016: got result for host git.mydomain.com disk /: 0 -> 1042740K, 1 -> 15570K, -1 -> -3K planner: time 199.035: got p
Re: Potential Bug
Rhys Rhaven wrote: Using amanda 3.2.0 on everything. I have 2 hosts and a server currently. 1 host and the server are ubuntu 10.04 and the other host (that I'm having issues with) is 8.04. I can run amdump for each host individually with no problem. Running amdump for the config (daily) will fail with the following. Talking to djmitche in IRC, told me to post to the list. My tapes are 200GB, so they shouldn't be filling up. I'm using vtapes and no holding disks. FAILURE DUMP SUMMARY: asterisk-it0.mydomain.com / RESULTS MISSING asterisk-it0.mydomain.com / lev 1 FAILED [can't dump in degraded mode] log: START driver date 2010110501 amdump.1 amdump: starttime 20101105115136 The log and amdump.1 files are not for the same run, can you send both files for the same run. Jean-Louis
Potential Bug
Using amanda 3.2.0 on everything. I have 2 hosts and a server currently. 1 host and the server are ubuntu 10.04 and the other host (that I'm having issues with) is 8.04. I can run amdump for each host individually with no problem. Running amdump for the config (daily) will fail with the following. Talking to djmitche in IRC, told me to post to the list. My tapes are 200GB, so they shouldn't be filling up. I'm using vtapes and no holding disks. FAILURE DUMP SUMMARY: asterisk-it0.mydomain.com / RESULTS MISSING asterisk-it0.mydomain.com / lev 1 FAILED [can't dump in degraded mode] log: INFO amdump amdump pid 3100 INFO driver driver pid 3102 INFO planner planner pid 3101 START planner date 2010110501 START driver date 2010110501 DISK planner asterisk-it0.mydomain.com / STATS driver hostname styx-it0 DISK planner git.mydomain.com / STATS driver startup time 0.004 INFO dumper dumper pid 3105 INFO dumper dumper pid 3104 INFO dumper dumper pid 3106 INFO dumper dumper pid 3107 INFO taper taper pid 3103 INFO planner Full dump of git.mydomain.com:/ promoted from 6 days ahead. FINISH planner date 2010110501 time 199.580 INFO planner pid-done 3101 INFO dumper gzip pid 3121 START taper datestamp 2010110501 label daily-07 tape 1 PART taper daily-07 1 git.mydomain.com / 2010110501 1/-1 0 [sec 78.771226 kb 435153 kps 5524.263398] SUCCESS dumper git.mydomain.com / 2010110501 0 [sec 88.620 kb 435154 kps 4910.3 orig-kb 1042750] INFO dumper pid-done 3121 DONE taper git.mydomain.com / 2010110501 1 0 [sec 84.00 kb 435153 kps 5180.392857 orig-kb 1042750] STATS driver estimate git.mydomain.com / 2010110501 0 [sec 90 nkb 1042772 ckb 435360 kps 4828] FAIL driver asterisk-it0.mydomain.com / 2010110501 1 [can't dump in degraded mode] INFO dumper pid-done 3106 INFO dumper pid-done 3104 INFO dumper pid-done 3107 INFO dumper pid-done 3105 INFO taper tape daily-07 kb 435153 fm 1 [OK] INFO taper pid-done 3103 FINISH driver date 2010110501 time 289.614 INFO driver pid-done 3102 amdump.1 amdump: start at Fri Nov 5 11:51:36 CDT 2010 amdump: datestamp 20101105 amdump: starttime 20101105115136 amdump: starttime-locale-independent 2010-11-05 11:51:36 CDT planner: pid 3659 executable /usr/libexec/amanda/planner version 3.2.0 planner: build: VERSION="Amanda-3.2.0" planner:BUILT_DATE="Mon Oct 18 09:20:30 PDT 2010" BUILT_MACH="" planner:BUILT_REV="3545" BUILT_BRANCH="3_2_0" planner:CC="i486-linux-gnu-gcc" planner: paths: bindir="/usr/bin" sbindir="/usr/sbin" planner:libexecdir="/usr/libexec" planner:amlibexecdir="/usr/libexec/amanda" mandir="/usr/share/man" planner:AMANDA_TMPDIR="/tmp/amanda" planner:AMANDA_DBGDIR="/var/log/amanda" CONFIG_DIR="/etc/amanda" planner:DEV_PREFIX="/dev/" RDEV_PREFIX="/dev/r" planner:DUMP="/sbin/dump" RESTORE="/sbin/restore" VDUMP=UNDEF planner:VRESTORE=UNDEF XFSDUMP="/sbin/xfsdump" planner:XFSRESTORE="/sbin/xfsrestore" VXDUMP=UNDEF VXRESTORE=UNDEF planner:SAMBA_CLIENT="/usr/bin/smbclient" GNUTAR="/bin/tar" planner:COMPRESS_PATH="/bin/gzip" UNCOMPRESS_PATH="/bin/gzip" planner: LPRCMD=UNDEF MAILER=UNDEF planner:listed_incr_dir="/var/lib/amanda/gnutar-lists" planner: defs: DEFAULT_SERVER="localhost" DEFAULT_CONFIG="DailySet1" planner:DEFAULT_TAPE_SERVER="localhost" DEFAULT_TAPE_DEVICE="" planner:NEED_STRSTR AMFLOCK_POSIX AMFLOCK_FLOCK AMFLOCK_LOCKF planner:AMFLOCK_LNLOCK SETPGRP_VOID ASSERTIONS AMANDA_DEBUG_DAYS=4 planner:BSD_SECURITY USE_AMANDAHOSTS CLIENT_LOGIN="amandabackup" planner:CHECK_USERID HAVE_GZIP COMPRESS_SUFFIX=".gz" planner:COMPRESS_FAST_OPT="--fast" COMPRESS_BEST_OPT="--best" planner:UNCOMPRESS_OPT="-dc" READING CONF INFO... planner: timestamp 20101105115136 planner: tape_length is set from tape length (20480 KB) * runtapes (1) == 20480 KB planner: time 0.000: startup took 0.000 secs SENDING FLUSHES... ENDFLUSH SETTING UP FOR ESTIMATES... driver: tape size 20480 reserving 0 out of 0 for degraded-mode dumps planner: time 0.000: setting up estimates for asterisk-it0.mydomain.com:/ setup_estimate: asterisk-it0.mydomain.com:/: command 0, options: none last_level 0 next_level0 6 level_days 0getting estimates 0 (-3) 1 (-3) -1 (-3) planner: time 0.000: setting up estimates took 0.000 secs GETTING ESTIMATES... driver: started dumper0 pid 3662 driver: send-cmd time 0.001 to dumper0: START 20101105115136 driver: started dumper1 pid 3663 driver: send-cmd time 0.001 to dumper1: START 201
Re: snapshot already exists
On Fri, Nov 05, 2010 at 10:15:55AM -0400, Jean-Louis Martineau wrote: > Brian Cuttler wrote: > >Amanda 3.1.2 on Solaris 10x86. > > > >Occasionally we see this failure, even when the prior run > >ends cleanly. > > > Can you find in the debug files why the destroy failed in the prior runs? > >I'm not sure how amanda handles it, > > does sit fail to create the snapshot ? > > does it destroy and then create a new snapshot ? > > > It doesn't destroy and recreate the snapshot. > The backup use the snapshot, so it backup old data > > destroy and create is probably good in case of 'dataset already exists', > can you post a patch? It sounds simple enough... and I figured out how to apply a patch yesterday. I was so close, needed gpatch rather than patch, I was so proud. [else you get these weird errors] > Jean-Louis > >It did successfully backup the DLE. > > > > > >NOTES: > > planner: cascade /cascadep/nalm/alm2 20101104183001 0 "Script > > 'amzfs-snapshot' command 'PRE-DLE-ESTIMATE': ERROR cannot create > > snapshot 'cascadep/nalm/a...@amanda-_cascadep_nalm_alm2-current': > > dataset already exists" > > > > > >--- > > Brian R Cuttler brian.cutt...@wadsworth.org > > Computer Systems Support(v) 518 486-1697 > > Wadsworth Center(f) 518 473-6384 > > NYS Department of HealthHelp Desk 518 473-0773 > > > > > > > >IMPORTANT NOTICE: This e-mail and any attachments may contain > >confidential or sensitive information which is, or may be, legally > >privileged or otherwise protected by law from further disclosure. It > >is intended only for the addressee. If you received this in error or > >from someone who was not authorized to send it to you, please do not > >distribute, copy or use it or any attachments. Please notify the > >sender immediately by reply e-mail and delete this from your > >system. Thank you for your cooperation. > > > > > > > --- Brian R Cuttler brian.cutt...@wadsworth.org Computer Systems Support(v) 518 486-1697 Wadsworth Center(f) 518 473-6384 NYS Department of HealthHelp Desk 518 473-0773 IMPORTANT NOTICE: This e-mail and any attachments may contain confidential or sensitive information which is, or may be, legally privileged or otherwise protected by law from further disclosure. It is intended only for the addressee. If you received this in error or from someone who was not authorized to send it to you, please do not distribute, copy or use it or any attachments. Please notify the sender immediately by reply e-mail and delete this from your system. Thank you for your cooperation.
Re: snapshot already exists
Brian Cuttler wrote: Amanda 3.1.2 on Solaris 10x86. Occasionally we see this failure, even when the prior run ends cleanly. Can you find in the debug files why the destroy failed in the prior runs? I'm not sure how amanda handles it, does sit fail to create the snapshot ? does it destroy and then create a new snapshot ? It doesn't destroy and recreate the snapshot. The backup use the snapshot, so it backup old data destroy and create is probably good in case of 'dataset already exists', can you post a patch? Jean-Louis It did successfully backup the DLE. NOTES: planner: cascade /cascadep/nalm/alm2 20101104183001 0 "Script 'amzfs-snapshot' command 'PRE-DLE-ESTIMATE': ERROR cannot create snapshot 'cascadep/nalm/a...@amanda-_cascadep_nalm_alm2-current': dataset already exists" --- Brian R Cuttler brian.cutt...@wadsworth.org Computer Systems Support(v) 518 486-1697 Wadsworth Center(f) 518 473-6384 NYS Department of HealthHelp Desk 518 473-0773 IMPORTANT NOTICE: This e-mail and any attachments may contain confidential or sensitive information which is, or may be, legally privileged or otherwise protected by law from further disclosure. It is intended only for the addressee. If you received this in error or from someone who was not authorized to send it to you, please do not distribute, copy or use it or any attachments. Please notify the sender immediately by reply e-mail and delete this from your system. Thank you for your cooperation.
snapshot already exists
Amanda 3.1.2 on Solaris 10x86. Occasionally we see this failure, even when the prior run ends cleanly. I'm not sure how amanda handles it, does sit fail to create the snapshot ? does it destroy and then create a new snapshot ? It did successfully backup the DLE. NOTES: planner: cascade /cascadep/nalm/alm2 20101104183001 0 "Script 'amzfs-snapshot' command 'PRE-DLE-ESTIMATE': ERROR cannot create snapshot 'cascadep/nalm/a...@amanda-_cascadep_nalm_alm2-current': dataset already exists" --- Brian R Cuttler brian.cutt...@wadsworth.org Computer Systems Support(v) 518 486-1697 Wadsworth Center(f) 518 473-6384 NYS Department of HealthHelp Desk 518 473-0773 IMPORTANT NOTICE: This e-mail and any attachments may contain confidential or sensitive information which is, or may be, legally privileged or otherwise protected by law from further disclosure. It is intended only for the addressee. If you received this in error or from someone who was not authorized to send it to you, please do not distribute, copy or use it or any attachments. Please notify the sender immediately by reply e-mail and delete this from your system. Thank you for your cooperation.
