• EN
Question

surebackup and sql server always on

F
Userlevel 1

Hello,

 

I’m actually testing surebackup with a sql server always on cluster.

Setting to test default SqlChecker script.

Unfurtunatelly, it seems not to be working well.

All the databases (exepts system db) are in RECOVERY_PENDING state in the mounted vm lab.

Is surebakup working for always on clusters ?

Am I missing something ?

 

thanks for your help.

Fred

10 comments

marco_s
Userlevel 7
Badge +7

Hi, what error do you receive?

Does the script work outside sure backup job session?

https://helpcenter.veeam.com/docs/backup/vsphere/custom_verification_scripts.html?ver=120#sql

Marco Sorrentino - Italy | System Engineer | VMCE & Veeam Lover
F
Userlevel 1

Thanks for your comment.

The script is working outside the lab with no error (all dbs are online and reachable)

The surebackup job fail with error logs : 

06/04/2023 11:20:06] Error Database 'DBNAME' can't be accessed. Execution of the query 'use [DBNAME]' failed with the error: -2147217900 Database 'DBNAME' cannot be opened due to inaccessible files or insufficient memory or disk space.  See the SQL Server errorlog for details.
[06/04/2023 11:20:06] Error     Database 'DBNAME' is unavailable. State: 3 - RECOVERY_PENDING

 

everything’s good for system db

[06/04/2023 11:20:00] Info Trying to access each database...
[06/04/2023 11:20:00] Info     Executing query: use [master]
[06/04/2023 11:20:01] Info     Database 'master' is available. State: 0 - ONLINE
[06/04/2023 11:20:01] Info     Executing query: use [tempdb]
[06/04/2023 11:20:02] Info     Database 'tempdb' is available. State: 0 - ONLINE
[06/04/2023 11:20:02] Info     Executing query: use [model]
[06/04/2023 11:20:03] Info     Database 'model' is available. State: 0 - ONLINE
[06/04/2023 11:20:03] Info     Executing query: use [msdb]
[06/04/2023 11:20:04] Info     Database 'msdb' is available. State: 0 - ONLINE

 

thanks

T

Hello Marco,

i can confirm your topic. Same problem in our SureBackup-Job with the Always-On Cluster.

 

Info Trying to access each database...
[19.04.2023 17:21:33] Info     Executing query: use [master]
[19.04.2023 17:21:33] Info     Database 'master' is available. State: 0 - ONLINE
[19.04.2023 17:21:33] Info     Executing query: use [tempdb]
[19.04.2023 17:21:33] Info     Database 'tempdb' is available. State: 0 - ONLINE
[19.04.2023 17:21:33] Info     Executing query: use [model]
[19.04.2023 17:21:33] Info     Database 'model' is available. State: 0 - ONLINE
[19.04.2023 17:21:33] Info     Executing query: use [msdb]
[19.04.2023 17:21:33] Info     Database 'msdb' is available. State: 0 - ONLINE
[19.04.2023 17:21:33] Info     Executing query: use [Docusnap_DB]
[19.04.2023 17:21:33] Error Database 'Docusnap_DB' can't be accessed. Execution of the query 'use [Docusnap_DB]' failed with the error: -2147217900 Database 'Docusnap_DB' cannot be opened due to inaccessible files or insufficient memory or disk space.  See the SQL Server errorlog for details.
[19.04.2023 17:21:33] Error     Database 'Docusnap_DB' is unavailable. State: 3 - RECOVERY_PENDING

dips
Userlevel 7
Badge +7

If you haven’t already done so, it might be worth logging this with Veeam Support. Out of interest, what version of Veeam are you using?

Dipen N. K. | IT Security Specialist | Veeam Legend 2022 - 2023 | www.dipen.co.uk
T

I´ve already get in touch with Veeam Support, i will keep you informed about the feedback.

We´re using Veeam V12.

 

 

Scott
Userlevel 7
Badge +8

I haven't yet enabled app aware SQL backups here so I can’t try the Always ON to see if mine is the same. 

 

Hard to convince the SQL guys they shouldn’t do their own backups. I’m looking into the agents now for V12

 

 

MicoolPaul
Userlevel 7
Badge +20

Are they all in the application group or a linked job? Do they have any CSV disks as a cluster or is it share-nothing design?

Michael Paul - #VeeamLegend | #VeeamVanguard | Veeam Certified Architect | VMware vExpert | https://micoolpaul.com | Twitter: @MicoolPaul | Mastodon: @micoolpaul@masto.nu | Bluesky: @micoolpaul.bsky.social
F
Userlevel 1

Hello,

 

Also on the last v12 version.

They are not sharing disks.

 

T

Hello Michael,

 

the are all in an application group

  • 1 Domain Controller
  • 2 Nodes Always-On Cluster

They´ve got a file share witnss.

Boot-Up sequenz now is the following - perhaps this will work.

  • Domain Controller → Check is ok
  • SQL-Node 2 - “inactive” node in the backup window → no sql check
  • SQL-Node 1 - “active” node in the bakcup windows → sql check enabled
I

You could edit the Veeam.Backup.SqlChecker.vbs

 

Change the line

gaDbSuccessStates = Array( DB_STATE_ONLINE, DB_STATE_OFFLINE )

to

gaDbSuccessStates = Array( DB_STATE_ONLINE, DB_STATE_OFFLINE, DB_STATE_RECOVERY_PENDING )

So the recovery pending state will also be considered as a successful state

Comment


Terms & Conditions