Hi Lloyd, all,

Am 10.11.2026 um 20:42 schrieb Lloyd Brown via Bacula-users:
On 11/10/26 12:28, Lloyd Brown via Bacula-users wrote:
On 11/10/26 11:48, Arno Lehmann via Bacula-users wrote:
This really gets to the crux of my original question.  Why is the cycle bad? Suboptimal performance, perhaps, though I'd argue that for disks the difference is negligible.

We stumbled over that because performance was so bad that users actually couldn't complete their copy and virtual backups in decent time, couldn't saturate their tape drives, and found that SD threads, doing such jobs, wasted more than 90% (iirc) of a CPUs time going through bootstrap records. Negligible only if you accepted a whole university's backups to simply not take place any more 🙂


Interesting.  This was occurring during the run of the VF?

It's been a while, and I did not look up the actual case in our tickets, but given that it involved large jobs and involved writing to tape, I would assume it's been copy or migration jobs in that particular case. Unfortunately the users has since disabled their publicly visible monitoring system, otherwise I would have been able to point to very respectable performance today :-)

We certainly didn't experience that.  I wonder if the fact that the old 9.6.6 build had many more jobs per day, and had 100G disk volumes vs 500G currently on bacula 15, meant we avoided this effect without realizing it.

That is quite likely -- if you just observe the size of the temporarily created bootstrap files, it may become obvious. In the case I'm thinking of, these were quite long due to exactly those topics you mentioned -- interleaved jobs and lots of volumes involved -- were surely the reason the problem became aparent, and the fact that tape drives are somewhat naturally limited in number and thus deserve proper monitoring of usage had a lot to do with the user noticing the issues.

Anyway -- if the current situation, the available workarounds, and the fact that Bacula could just be better lead us to see improvements in that area, I'd be quite happy to qualify and verify any improvements; actually developing is most likely not something I should do :-)

Cheers,

Arno


Lloyd


This is additional speculation on my part, but I wonder also if it depends on how interleaved the jobs' data are on the disk volumes.  A long time ago we had performance problems during a restore when it had to do a lot of seeks within the disk volumes. We switched to having many `Device` entries in the SD's config, each with `Maximum Concurrent Jobs = 1`.  The result is that interleaving within a specific job run almost entirely disappeared.  And that would mean many fewer transitions between volumes when running the VF, and therefore fewer and likely shorter cycles, if they occurred at all.

Lloyd



_______________________________________________
Bacula-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bacula-users

--
Arno Lehmann

IT-Service Lehmann
Sandstr. 6, 49080 Osnabrück



_______________________________________________
Bacula-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bacula-users

Reply via email to