Hi,
I occasionally run into the following warning during copy and migration jobs:
Warning: dird/catreq.cc:608 MD5 digest not same File=103468 as attributes=103467
The two indices change with each occurrence, but the file and attribute index are always off by one.
Does anybody know what this is about?
Thanks and best regards,
Samuel
Hi Samuel,
would it be possible to run bscan with the bsr of that job and the -r/--list-records option set ?bscan: stored/bscan.cc:494-0 Record: SessId=1 SessTim=1708427201
FileIndex=-4 Stream=1 len=164
bscan: stored/bscan.cc:494-0 Record: SessId=1 SessTim=1708427201
FileIndex=1 Stream=1 len=182
bscan: stored/bscan.cc:494-0 Record: SessId=1 SessTim=1708427201
FileIndex=1 Stream=29 len=4250
bscan: stored/bscan.cc:494-0 Record: SessId=1 SessTim=1708427201
FileIndex=1 Stream=1998 len=80
bscan: stored/bscan.cc:494-0 Record: SessId=1 SessTim=1708427201
FileIndex=1 Stream=40 len=16
bscan: stored/bscan.cc:494-0 Record: SessId=1 SessTim=1708427201
FileIndex=2 Stream=1 len=183
bscan: stored/bscan.cc:494-0 Record: SessId=1 SessTim=1708427201
FileIndex=2 Stream=29 len=10008
bscan: stored/bscan.cc:494-0 Record: SessId=1 SessTim=1708427201
FileIndex=2 Stream=1998 len=80
bscan: stored/bscan.cc:494-0 Record: SessId=1 SessTim=1708427201
FileIndex=2 Stream=40 len=16
This would help us greatly in trying to debug this issue.
Kind Regards
nto the following warning during copy and migration jobs:
Warning: dird/catreq.cc:608 MD5 digest not sa
-- Sebastian Sura sebasti...@bareos.com Bareos GmbH & Co. KG Phone: +49 221 630693-0 https://www.bareos.com Sitz der Gesellschaft: Köln | Amtsgericht Köln: HRA 29646 Komplementär: Bareos Verwaltungs-GmbH Geschäftsführer: Stephan Dühr, Jörg Steffens, Philipp Storz
Hi Samuel
thanks for the gathering this info. Ill look into the bscan
issue as well though i think this might be a known issue.
Regardless, regarding your actual issue: The bscan output shows
that File 3 is missing its attributes (it has no stream 3).
This confuses the director as it never got told that file 3 now
started getting backed up and this is why you get the warning
message from the director.
The same is happening during the restore: the filedaemon never
got told that now a new file was started (because it never got the
attribute stream)
so it basically merged both File 2 and File 3 into one file. Even
with this in mind, i think the restore should have caught that and
issues a warning.
Ill look into why this did not happen.
If you add up all the data records (stream=2) with fileid=2 and
fileid=3, you will get 738212473 bytes, which is exactly what the
filedaemon reported
as size for file 2. Since the log contains the size it expected i
imagine that you could manually restore file 3 by splitting file 2
into two.
Can you check which file fileid 3 corresponds to ? Is it a special kind of file or just a normal one ?
Ill try to see if i can reproduce your issue in the copy system test. Do you do a local copy (so copy to the same sd) or a remote one?
Kind Regards
Sebastian Sura
--
You received this message because you are subscribed to the Google Groups "bareos-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bareos-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bareos-users/e766e8c6-5b58-424f-816f-f7c5d7252180n%40googlegroups.com.
I meant to say File 3 has no stream 1, sorry for the confusion!
To view this discussion on the web visit https://groups.google.com/d/msgid/bareos-users/111cecfe-d3e1-4eb9-8a0e-f053a2777d0f%40bareos.com.
Enter SQL query: select *, decode_lstat(lstat) from file where jobid=94511 and fileindex=2;
+-------------+-----------+-------+---------+----------+--------+--------+--------+----------------------------------------------------------------+------------------------+--------------------------+----------------------------------------------------------------------------------------------+
| fileid | fileindex | jobid | pathid | deltaseq | markid | fhinfo | fhnode | lstat | md5 | name | decode_lstat |
+-------------+-----------+-------+---------+----------+--------+--------+--------+----------------------------------------------------------------+------------------------+--------------------------+----------------------------------------------------------------------------------------------+
| 484,575,603 | 2 | 94511 | 910,901 | 0 | 0 | 0 | 0 | P0D CgA4 IGk B A i A sADp5 BAA FgAo BlxAt1 BlxASF BlxVYH A A d | D/VouduY5TF4KFacjRE7Hw | backup_www_1100CC.tar.gz | (64771,655416,33188,1,0,34,0,738212473,4096,1441832,1707346805,1707345029,1707431431,0,0,29) |
+-------------+-----------+-------+---------+----------+--------+--------+--------+----------------------------------------------------------------+------------------------+--------------------------+----------------------------------------------------------------------------------------------+
Enter SQL query: select *, decode_lstat(lstat) from file where jobid=94511 and fileindex=3;
+-------------+-----------+-------+---------+----------+--------+--------+--------+---------------------------------------------------------------+------------------------+-----------------------------+-------------------------------------------------------------------------------------------+
| fileid | fileindex | jobid | pathid | deltaseq | markid | fhinfo | fhnode | lstat | md5 | name | decode_lstat |
+-------------+-----------+-------+---------+----------+--------+--------+--------+---------------------------------------------------------------+------------------------+-----------------------------+-------------------------------------------------------------------------------------------+
| 484,575,604 | 3 | 94511 | 910,901 | 0 | 0 | 0 | 0 | P0D CgA5 IGk B A i A BA9gA BAA IHw BlxAt8 BlxASH BlxVYH A A d | J9h+OdR/6XCqhqpJdJvYtw | backup_databases_1100CC.tar | (64771,655417,33188,1,0,34,0,17029120,4096,33264,1707346812,1707345031,1707431431,0,0,29) |
+-------------+-----------+-------+---------+----------+--------+--------+--------+---------------------------------------------------------------+------------------------+-----------------------------+-------------------------------------------------------------------------------------------+
Restore of this job (jobid=94511) works just fine:
23-Feb 13:44 bareos-sd JobId 94559: Releasing device "FileDevice-ReadOnly-0005" (/backup_1/bareos).
23-Feb 13:44 bareos-dir JobId 94559: Max configured use duration=82,800 sec. exceeded. Marking Volume "AI-Consolidated-1771" as Used.
23-Feb 13:44 bareos-dir JobId 94559: Bareos bareos-dir 23.0.2~pre32.0a0e55739 (31Jan24):
Build OS: Ubuntu 20.04.5 LTS
JobId: 94559
Job: Restore-Files.2024-02-23_13.43.47_38
Restore Client: "bareos-fd" 23.0.2~pre32.0a0e55739 (31Jan24) Ubuntu 20.04.5 LTS,ubuntu
Start time: 23-Feb-2024 13:43:49
End time: 23-Feb-2024 13:44:20
Elapsed time: 31 secs
Files Expected: 33
Files Restored: 33
Bytes Restored: 9,143,514,060
Rate: 294952.1 KB/s
FD Errors: 0
FD termination status: OK
SD termination status: OK
Bareos binary info: Bareos community build (UNSUPPORTED): Get professional support from https://www.bareos.com
Job triggered by: User
Termination: Restore OK
I'll also try restoring its copy from tape once it has been copied.
Best regards,
Samuel
Hi Samuel
i just wanted to make sure i completely understand the
situation. You have an always incremental job with compression
turned on (which algorithm do you use ?) and then you have a copy
job that copies the consolidated full job (on the same sd)
to tape with a device that auto inflates on write.
Do i have that right ?
Does the device where the job gets consolidated onto allow concurrent writes from multiple jobs ?
What selection type has your copy job ?
Kind Regards
Sebastian Sura
To view this discussion on the web visit https://groups.google.com/d/msgid/bareos-users/3823a22d-b49e-4892-8b45-6c932ce7e699n%40googlegroups.com.
Hi Samuel,
i think i managed to reproduce your Problem. I created a fix here if you want to check it out: https://github.com/bareos/bareos/pull/1745
Kind Regards
Sebastian Sura
To view this discussion on the web visit https://groups.google.com/d/msgid/bareos-users/67455f46-6182-4fb6-9be1-e8969ae0fb3dn%40googlegroups.com.