Backup Copy Job scheduling

[baremetalbrisbane]

Morning

I am hoping i can get a bit more traction here(02622569)

Single repository server with SSD's SOBR and HDD SOBR attached

We have around 50 or so backup copy jobs that start after the tape job, so primary to SSD, To Tape and then GFS to mechanical

At some point in our GFS run backup copy run all active jobs (normally 15 running/active) will go to a state of "waiting for backup resource availability". This can happen with any activity i.e merges, health, copying data

The end result is that we have 15 jobs all waiting 10-15 hours later with the same status. On the repository server there is zero IO or CPU activity and no open connections to any disks that i can see

Server is not CPU or memory constrained (8vcpu 20GB memory). As its scale out the mechanical side is limited to 1 task per extent (6 extents)

Support have said its a concurrent tasks issue, i have asked for any proof of this. I fail to believe that an enterprise grade product like Veeam cant work out the scheduling for overlapping GFS jobs, or jobs that start at the same time. Primary files arent locked as the primary backup and tape job has finished

Please help

[PTide]

Hi,

The end result is that we have 15 jobs all waiting 10-15 hours later with the same status. On the repository server there is zero IO or CPU activity and no open connections to any disks that i can see

That is, once your active GFS jobs get into the "waiting for resource" state, then neither source, nor target repository will show any I/O activity for many hours, seemingly stuck at doing nothing? If so, then I suggest you to escalate the case.

Thanks