Hello everybody.
Thank you for having me.
I have searched the archives and found many references to timing and
looking at the script itself, but I am a bit stumped and maybe your
expert view can help.
OS: Centos 6.4
Bacula:
bacula-director-mysql-5.0.0-12.el6.x86_64
bacula-console-5.0.0-12.el6.x86_64
bacula-common-5.0.0-12.el6.x86_64
bacula-storage-mysql-5.0.0-12.el6.x86_64
bacula-director-common-5.0.0-12.el6.x86_64
bacula-client-5.0.0-12.el6.x86_64
bacula-storage-common-5.0.0-12.el6.x86_64
mtx:
mtx-1.3.12-5.el6.x86_64
mt:
mt-st-1.1-5.el6.x86_64
Mysql:
mysql-server-5.1.69-1.el6_4.x86_64 (and supporting)
All services come up as expected, bconsole status all shows that all is
well.
The tape loaders are Two IBM TS2900 with LT05 drive.
Executing
./mtx-changer /dev/sg2 listall /dev/nst0 0
lists all 8 slots as expected with the drive being in slot 0
(I had a tape already in the drive)
Executing
./mtx-changer /dev/sg2 unload 3 /dev/nst0 0
results in:
Unloading drive 0 into Storage Element 3...done
and mtx-changer successfully exits on the command line.
Executing
./mtx-changer /dev/sg2 load 3 /dev/nst0 0
results in:
Loading media from Storage Element 3 into drive 0...done
and mtx-changer successfully exits on the command line.
For all those operations the debug in mtx.log reads
20130529-18:41:58 Parms: /dev/sg2 listall /dev/nst0 0
20130529-18:41:58 Doing mtx -f /dev/sg2 -- to list all
20130529-18:42:31 Parms: /dev/sg2 unload 3 /dev/nst0 0
20130529-18:42:31 Doing mtx -f /dev/sg2 unload 3 0
20130529-18:45:45 Parms: /dev/sg2 load 3 /dev/nst0 0
20130529-18:45:45 Doing mtx -f /dev/sg2 load 3 0
mtx-changer.conf has only one variable set which is:
load_sleep=10
Executing bconsole I get the following status:
(for storage)
Device status:
Autochanger "Autochanger" with devices:
"TS2900server" (/dev/nst0)
Device "backupstore" (/tmp) is not open.
Device "TS2900server" (/dev/nst0) is not open.
Drive 0 status unknown.
mtx.log shows (after bacula-dir startup)
20130529-18:50:31 Parms: /dev/sg2 loaded 0 /dev/nst0 0
20130529-18:50:31 Doing mtx -f /dev/sg2 0 -- to find what is loaded
20130529-18:50:32 Parms: /dev/sg2 loaded 0 /dev/nst0 0
20130529-18:50:32 Doing mtx -f /dev/sg2 0 -- to find what is loaded
Since this is a fresh install with a cassette not yet catalogued I
wanted to run:
update slots scan
that is when the problems start.
Slot 1 has a cleaning tape
The system starts executing the first load trying to load the first Tape
into the drive, I can see it executing
./mtx-changer /dev/sg2 load 1 /dev/nst0 0
and after that is simply hangs until bconsole kills the command with a
Signal 15.
The same thing happens when mtx-changer is ran from the command line.
It will load, say done and then hang until I ctrl+c it.
It is as if mt cannot tell that the tape is ready when it is a cleaning
tape and then just hangs. The debug log shows
20130529-19:07:32 Device /dev/nst0 - not ready, retrying...
update slots in bconsole produces output:
3306 Issuing autochanger "slots" command.
Device "Autochanger" has 7 slots.
Connecting to Storage daemon tapestorage at backupibm.*.corp:9103 ...
3306 Issuing autochanger "list" command.
Volume "CLNU51L1" not found in catalog. Slot=1 InChanger set to zero.
Volume "432CEUL5" not found in catalog. Slot=2 InChanger set to zero.
Volume "434CEUL5" not found in catalog. Slot=3 InChanger set to zero.
Volume "488CFAL5" not found in catalog. Slot=4 InChanger set to zero.
So I am wondering if that is something anyone has experienced before.
What else could I be doing to debug this?
run mt status with a cleaning tape loaded versus a regular tape and
compare the output?
What would you suggest?
Thank you!
------------------------------------------------------------------------------
Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
Get 100% visibility into your production application - at no cost.
Code-level diagnostics for performance bottlenecks with <2% overhead
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap1
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
