Hello,
Hello, Dan.
I am working with an LTO-4 tape library. It has two drives but I plan to write to only one for backups.
I will backup to disk first, on another SD. Later, I will copy the jobs to the tape library on this new SD
which is on another server. The copy jobs will be spooled to local SSD before being written to tape.
Now what I'm thinking of is streaming multiple concurrent jobs to a single drive.
Did you test if it is possible? (I can test for you in a dummy environment if you want). I don't remember if it is or if the bellow quote only applies to copy jobs...
"If the Migration control job finds a number of JobIds to migrate (e.g. it is asked to migrate one or more Volumes), it will start one new migration backup job for each JobId found on the specified Volumes. Please note that Migration doesn't scale too well since Migrations are done on a Job by Job basis. This if you select a very large volume or a number of volumes for migration, you may have a large number of Jobs that start. Because each job must read the same Volume, they will run consecutively (not simultaneously)." (http://www.bacula.org/5.2.x-manuals/en/main/main/Migration_Copy.html)
Sure, downside on restore is interleaving of blocks....
I also remember that I read somewhere that you don't need to concern about that when doing copy jobs, but I think it refers to lack of multiplexing quoted before.
I don't see any downsides to going down this path. I have yet to run any copy jobs to the new library,
but it may be ready this week.
Even without multiplexing, I think you will achieve a hight throughput since you are copying sequential data.
In this case, if you really want to multiplex the copy I think you would have to create two different source and destination pools.
Comments?
--
Dan Langille - BSDCan / PGCon
