V6 Waiting for Resources

[odge]

We've asked support, but think its probably a good one for forum as someone is going to ask it or look it up.

What does it generally mean if it sits waiting to do the backup with "V6 Waiting for Foreign Virtual Disks to be released".

Also, if you have 2 replication running, both using the same proxy as target and source, does that mean that it requires 4 "simultaneous" sessions?

Thanks

[foggy]

Matthew, it's definitely better to proceed with support on this as this single message is obviously not sufficient and full logs are required for investigation.

Regarding your second question, what kind of sessions are you referring to? If you are talking about multithreading, then within one backup session Veeam B&R can open up to 5 parallel TCP/IP connections. You can control this with the help of the 'Use multiple TCP/IP connections per job' check box in the throttling rules dialog of the proxy settings.

Thanks.

[odge]

Yip, have done so first (several hours before), but thought its probably something reasonably simple, it just not giving a very helpful directive error in the logs or GUI. Will post cause here, maybe for the next guy, its the same issue, and he can save your tech department some time.

[foggy]

Would be much appreciated!

[Gostev]

V6 waiting for foreign virtual disks to be released

Hey, I know this one - I actually wrote this resource. This means the job is trying to backup a VM that serves as Veeam backup proxy, and that proxy currently has other VM disks hot added to it. Obviously you would not want to backup the VM at this time, as this would put whole bunch of other VMs' disks into the backup file as well. Restore from such a backup would be lots of fun too (not to mention the size of the backup).

[odge]

perfect sense! ok, so effecively we should rather look at using any other windows server attached to the cluster for this backup to do the work.
OR
I guess its kind of tricky, (ok, not that tricky ! But we need to always ensure the veeam proxies are backed up, when they are not backing anything else up.

Thanks for the help.

[Gostev]

You might as well do nothing, since the job will simply wait until that proxy VM finishes processing the task, and unmounts those disks. As soon as this happens, the job will then go ahead and back it up.

[LucW]

seems i have a same issue. running a local fast daily backup doing 1.2 TB in around 3 hours.
i have now started a WAN based offsite backup , running much slower , and will take around +/- 5 days (i hope/estimate ) until 1 backup is ready.

now my daily backups of all productions VM is waiting on the VM of the backup proxy server because of this WAN backup running.

so should i exclude the backup VM or use backup timeout. anyway, both will result in a backup with errors

[Gostev]

I would recommend stopping the WAN job and seeding the full backup instead to the remote site instead. 5 days is too long anyway, it is too likely that the network connection will drop during this time, and the job would fail anyway.

[LucW]

This is currently running in a test environmont , so with simulated minimum bandwidth to see how long it will take.
at the moment i'm not able to seed the remote end , and was kind of expecting that it will retry any error during connection.

main goal is to replace current rsync setup and to try to backup to a Amazon EC2/ESB combination.
still not sure if this is possible because i noticed that beside the initial ssh connections, the backup agent is also
creating connections for backup data stream to the backup proxy, which makes it hard if you have NAT firewalls and no tunnel between the 2 sides.
i was hoping it would use only ssh or only connections initiated from the backup server .

luc

[Gostev]

Wow, that is interesting project for sure! Please keep us posted on your progress.

[roniva]

Has anyone tried going into the Exclusions and for the proxy VM's telling it to only backup the system disk (or selected disks)? I would hope this would get you around this issue and I'll be trying it out later today or tomorrow (whenever my 13TB Exchange backup finishes).

-Ron

[roniva]

It looks like the answer to my question is no, telling it to only backup the system disk does not get you around this issue. It looks like I'll be creating a separate backup job for my backup servers so if a job runs log, I won't risk missing the next window.

If anyone does have another way, let us know.

-Ron

[kingpin1]

There is another related issue that significantly affects performance of SQL Server hosting B&R DB:
while task is in "Waiting for processing resources availability" state there are a lot of DB requests (my system can handle ~160 request per second).

There are 2 flooding requests:

Code: Select all

exec GetSessionControl @row_id='8EC77D49-A087-41A0-BC34-92BBA643388B'

exec UpdateSessionXmlLog @row_session_id='8EC77D49-A087-41A0-BC34-92BBA643388B',@row_log_id=5,@row_time=N'2012-03-04T17:32:56.4380000+04:00',@row_utc_time=N'03/04/2012 13:32:56',@row_status=N'ENone',@row_title=N'Waiting for processing resources availability',@row_desc=N''