insufficient system memory in resource pool 'internal'

[connelp]

Hi all

Quick bit of background. We're running VEEAM B+R 7.0.0.839 with the bundled SQL Server 2008 R2 database (which is patched to SP2).

Several of my backups failed this weekend with VEEAM reporting:

Error: Microsoft SQL server hosting the configuration database is currently unavailable. Possible reasons are heavy load, networking issue, server reboot, or hot backup. Please wait, and try again later. Error information: SQL server is not available

A bit more investigation revealed SQL Server generating the following error:

There is insufficient system memory in resource pool 'internal' to run this query.

If you Google the above error you will find an avalanche of hits!

I have raised a call about this before (it happened in Feb 2014) and applying the service packs and a reboot fixed it.

However, given that when it happens it generally wipes out a full backup I'd like to stop it happening again!

So, has anyone else had this problem and how did you fix it?

Please note also that this server is specced above and beyond what is required for VEEAM B+R.

Thanks

Paul

[Vitaliy S.]

Hello Paul,

SQL Server errors cannot wipe out your backup files, if you had a ruining session, then the next job pass should fix everything that failed. Can you please elaborate it further?

There are some existing topics with the same error message, please take a look, might be helpful > http://forums.veeam.com/search.php?st=0 ... navailable.

Thank you!

[connelp]

Hi Vitaliy

Some of my jobs failed three times so I assumed that was that.

I do not run reverse incremental backups, incremental in the week and full backups at weekends. I don't want the full backups to run on a weekday for obvious reasons.

Maybe I should have mentioned the base OS is Windows Storage Server 2012.

Thanks

[Vitaliy S.]

Not sure it could be the reason, but it seems like Windows Storage Server can bring some issues with EULA > Windows Storage Server 2012 compatibility.

If I were you I would try to move your SQL Server installation to anther box and see is that fixes the issue or not. Even if you're running forward incremental mode, you do not need to run active full after the failed incremental job run.

[foggy]

If you have full scheduled for the weekend, it will not be performed on weekdays, even tough it failed last weekend. It will be just skipped and attempted next weekend, according to the schedule.

[tmckinnon]

Have you found a solution for your issue?

I've just had the same thing occur to me after I upgraded from SQL Express 2005 / Veeam 7 Patch 4 to SQL Express 2012 SP2 and Veeam v8.

All worked OK for a day or so but then a whole bunch of SQL errors are now popping up

I've raised a case 00678642.

While I wait for support I'm thinking I should install SQL 2012 SP2 CU2 (http://support.microsoft.com/kb/2983175) as it contains a fix to resolve a memory leak (http://support.microsoft.com/kb/2965758)

[tmckinnon]

The issue was resolved without the help of support

For others looking at this issue -

Solution (in this case) which made our MS SQL Express stable was to set the MAXMEM to 1024MB and to make sure the compatibility level is 2012.