Tape Jobs Start and Stop

[JaxIsland7575]

I have multiple tap jobs scheduled each week and I try and space them out but they run overnight so most weeks they overlap. If backup Job A is running then the schedule for Job B starts, it will stop Job A and start Job B. Then this repeats if Job C starts. Then through the jobs they stop and switch back to Job A. This causes all the jobs to run much longer because of swapping tapes rather than just finishing the running job then moving on.

Is there a way to control how this works so I can keep the original jobs running in order of start time?

Thank you!

[Rick.Vanover]

Hey Jax

It may be better to put this in the tape forum vs. Vanguard forum. Over there - we have the R&D tape leads with access (where here we do not).

[JaxIsland7575]

Rickatron!

I did, or at least it appears that it is in the Main -> Products -> Veeam Backup & Replication -> Tape forum, is it showing differently for you?

Thanks!

[Dima P.]

Hi Ryan,

if backup Job A is running then the schedule for Job B starts, it will stop Job A and start Job B. Then this repeats if Job C starts.

Can you please clarify the type of this jobs and how they are scheduled (is that after this job type of schedule)? Do you have anything to share from the job log (specifically the indication of one backup job is being interrupted by another)? Thank you!

[Rick.Vanover]

Oh one better - Popov is in here. Nevermind this is good

[JaxIsland7575]

So here is the layout of the jobs:
Job A is a Backup to Tape that begins every Sunday at 0430 (weekly)
Job B is a Backup to Tape that begins the last Sunday of every month at 1130 (monthly)
Job C is a File to Tape that begins every Monday at 0500 (weekly)
Job D is a File to Tape that begins the last Sunday of every month at 2100 (monthly)
The Tape Library is configured with separate pools for each of the 4 jobs.

On Sunday 2019/02/03 Job A began at 0430 as scheduled and ran for approximately 24 hours rather than completing in less than 12 hours on a normal week. The same day later in the morning Job B begins at 1130 as scheduled but instead of going onto a queued phase it stops Job A and Job B begins and completes before Job A resumes and finishes.

Here is a snippet from the log file associated with Job B from when it started. Let me know if there is a more specific log file or snippet that will help.

Code: Select all

[03.02.2019 11:30:17] <60> Info     
[03.02.2019 11:30:17] <60> Info         [Scheduler] Ready to start jobs:
[03.02.2019 11:30:17] <60> Info         Name: [Monthly_Tape], Type: [VM Tape Backup], UID: [e3cedc15-57ea-466c-bfdc-762ac0fa6e69]
[03.02.2019 11:30:17] <60> Info     Job [Monthly_Tape] starting by scheduler. Type: [VmTapeBackup]
[03.02.2019 11:30:17] <60> Warning  Next run time for job [Monthly_Tape] is not updated.
[03.02.2019 11:30:17] <60> Info     Checking for running backup job bound up VM Tape Backup job 'Monthly_Tape'.
[03.02.2019 11:30:17] <60> Info         Job event 'started' was created for session '0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42'. Event full name: 'Global\startedvmjobevent0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42'
[03.02.2019 11:30:17] <60> Info         Job event 'busy' was created for session '0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42'. Event full name: 'Global\busyvmjobevent0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42'
[03.02.2019 11:30:17] <60> Info         Starting process. Path: [C:\Program Files\Veeam\Backup and Replication\Backup\Veeam.Backup.Manager.exe], Arguments: ["STARTTAPEVMBACKUP" "e3cedc15-57ea-466c-bfdc-762ac0fa6e69" "0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42" ], DesktopName: [VeeamBackupServiceDesktop]
[03.02.2019 11:30:18] <34> Info         Job event 'ready2work' was opened for session '0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42'. Full event name: 'Global\ready2workvmjobevent0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42'.
[03.02.2019 11:30:18] <34> Info         Acquiring job process lock (e3cedc15-57ea-466c-bfdc-762ac0fa6e69).
[03.02.2019 11:30:18] <34> Info         Job event 'finished' was opened for session '0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42'. Full event name: 'Global\finishedvmjobevent0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42'.
[03.02.2019 11:30:18] <34> Info         Job process lock was disposed.
[03.02.2019 11:30:18] <34> Info     [RTS] Notified that job was started. SessionId: [0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42], Notifier: [Service]
[03.02.2019 11:30:18] <34> Info         Job event 'ready2work' was disposed. Session: '0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42'.
[03.02.2019 11:30:18] <34> Info     Job [Monthly_Tape] was started. SessionId: [0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42]
[03.02.2019 11:30:18] <34> Info         Job event 'busy' was disposed. Session: '0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42'.
[03.02.2019 11:30:18] <34> Info         Job event 'started' was disposed. Session: '0ca8b9d4-8c28-4aea-bf2f-4f5c68176c42'.
[03.02.2019 11:30:23] <160> Info     Checking identity rights as S-1-5-21-1111111111-1111111111-111111111-1111
[03.02.2019 11:30:23] <160> Info     BUILTIN Administrator SID: S-1-5-21-1111111111-1111111111-111111111-1111
[03.02.2019 11:30:33] <118> Info     Opening client session...
[03.02.2019 11:30:33] <118> Info     Checking compatibility between client and server (Client='9.5.4.2615', Server='9.5.4.2615', Components=2)
[03.02.2019 11:30:33] <118> Info     Current user Domain\Admin
[03.02.2019 11:30:33] <118> Warning  [SatelliteGateway] Failed to register new route
[03.02.2019 11:30:33] <118> Info     CClientSessionStub [s687] was created by [N/A]. SessionUID: [797711a8-d69b-4e60-8baf-1cb657e7ddf0]. Total sessions: [51]
[03.02.2019 11:30:40] <114> Info         Checking access rights, principal [Domain\Admin], permission [JobsMgmt]
[03.02.2019 11:30:40] <114> Info         Access is granted. User is in group [BUILTIN\Administrators]

[JaxIsland7575]

Another piece to this is it appears that when the tape is full during a job it changes the tape and then starts a different job rather than continuing with the one that was running. We use a separate pool for all jobs, there are no shared tape pools between jobs.

Maybe that helps point this in the right direction.

Thanks!

[Dima P.]

Hi Ryan,

Sorry for the delay. Described behavior looks unexpected, so please open a case and share the case ID with me. Thank you in advance!

[JaxIsland7575]

I finally got around to opening a case. I am upgraded to U4a and it has not made any difference in the issue.

Case #03503460

Thanks Dima!