commvault.com commvault.com

Media Agents automatically updated to 11.2801.22 and VMWare jobs fail

S
Rank
Userlevel 3
Badge +8
  • Shane Commvault Certified Expert
  • Commvault Certified Expert
  • 74 replies

Is there a way to prevent these automatic updates to the Media Agents from running?

Version 11.2801.22 breaks VMware backups and I have to keep manually reverting to 11.2801.11. Very frustrating!

Ryan Carr 1 year ago

Hi Shane 

Looking at the thread and case I can see Development backported a fix. 

The fix is to stop invoking APIs on VDDK versions lower than 6.5.2

So the best path forward is that we get the environment on 11.2801.22

Crash dumps have been enabled so if there is an issue then we can get the dumps and logs to Dev and confirm if the API’s is not being invoked and it could be something else that the team need to help you resolve. 



 

View original
If you have a question or comment, please create a topic

26 replies

Onno van den Berg
Rank
Userlevel 7
Badge +19

@Shane by default there is an automatic schedule tasks that runs to install updates. Easiest way to achieve this is by disabling the task through the Commcell console. 

S
Rank
Userlevel 3
Badge +8

@Shane by default there is an automatic schedule tasks that runs to install updates. Easiest way to achieve this is by disabling the task through the Commcell console. 

On Metallic? How do I access the console?

Onno van den Berg
Rank
Userlevel 7
Badge +19

You can’t access this console when you are using Metallic. Have you checked out the Metallic documentation? If you are hitting an issue with the MAs than I would open a ticket to get it sorted!

S
Rank
Userlevel 3
Badge +8

Well Metallic’s down at the moment, for a change.

But I did log a Critical over 3 hours ago, no response as yet.

Onno van den Berg
Rank
Userlevel 7
Badge +19

Ok, we’ll in that case it might be that they are installing a maintenance release that also addresses your issue. So I would recommend to see if this this is the case once it is back online and see what happens once you update your MAs.

S
Rank
Userlevel 3
Badge +8

It’s possible, but it’s not in “Maintenance”, jobs are just Waiting and some pages don’t load. Call logged for that too. 

Going to have some pretty unhappy customers when their TLog drives fill up.

Onno van den Berg
Rank
Userlevel 7
Badge +19

Mmmm that doesn’t sound as online….. Curious to hear the outcome! Keep us posted!

S
Rank
Userlevel 3
Badge +8

Will do, thanks.

Mike Struening RETIRED
Rank
Userlevel 7
Badge +23

@Shane , I found your case and want to split this into its own thread.  Looks like your MA was auto updated which certainly warrants some attention.

I’ll move this to the Metallic Q&A section so it gets better attention and tracking.

https://www.linkedin.com/in/michael-struening
Onno van den Berg
Rank
Userlevel 7
Badge +19

Thanks mr moderator @Mike Struening !

Mike Struening RETIRED
Rank
Userlevel 7
Badge +23

Thanks mr moderator @Mike Struening !

I try!

https://www.linkedin.com/in/michael-struening
Onno van den Berg
Rank
Userlevel 7
Badge +19

Thanks mr moderator @Mike Struening !

I try!

You’re doing great!

L
Rank
Userlevel 1
Badge +4

Hi,

You can check in device manager at MA / proxy under SCSI if Commvault driver is enabled. But there could be dozen of other root causes. 

 

Tbh, support is.. average.. and everything takes ages..

 

 

Mike Struening RETIRED
Rank
Userlevel 7
Badge +23

Hi,

You can check in device manager at MA / proxy under SCSI if Commvault driver is enabled. But there could be dozen of other root causes. 

 

Tbh, support is.. average.. and everything takes ages..

 

 

@LukasM , I’m sorry to hear you’ve had a bad experience.  I would like to have someone reach out to you to discuss and see how we can remedy the situation.

If that sounds good to you, can you send me a pm with some details on the case numbers and what could have been done better?

 

https://www.linkedin.com/in/michael-struening
S
Rank
Userlevel 3
Badge +8

@Shane , I found your case and want to split this into its own thread.  Looks like your MA was auto updated which certainly warrants some attention.

I’ll move this to the Metallic Q&A section so it gets better attention and tracking.

Hi Mike, can I please ask you to get involved here, it seems my message is not getting through.

220713-177

I’ve tried to explain that on version 11.2801.11 all is well, but on versions 11.2801.22 and 11.2801.23 vsbkp.exe sometimes crashes and the job fails. Different days, different Gateways. But I am banging my head against a brick wall.

Many thanks,

Onno van den Berg
Rank
Userlevel 7
Badge +19

So the issue remains….. @Mike Struening  Can't you just disable the auto updating of the MAs or will Metallic update clients or specifically MAs automatically because the normal advice has always been to make sure the MAs run the same version as the CommServe.

S
Rank
Userlevel 3
Badge +8

I downgraded this morning to .11 and before I could rerun the failed jobs it auto-updated to .23.

Pulling my hair out here.

Onno van den Berg
Rank
Userlevel 7
Badge +19

What do you see when you lookup the job history on server level? Do not forget to select show admin jobs. It really appears they have a backend job running that constantly runs and performs forced updates on clients running the MA role. 

Ryan Carr
Rank
Userlevel 2
Badge +5

Hi Shane 

Looking at the thread and case I can see Development backported a fix. 

The fix is to stop invoking APIs on VDDK versions lower than 6.5.2

So the best path forward is that we get the environment on 11.2801.22

Crash dumps have been enabled so if there is an issue then we can get the dumps and logs to Dev and confirm if the API’s is not being invoked and it could be something else that the team need to help you resolve. 



 

S
Rank
Userlevel 3
Badge +8

Hi Shane 

Looking at the thread and case I can see Development backported a fix. 

The fix is to stop invoking APIs on VDDK versions lower than 6.5.2

So the best path forward is that we get the environment on 11.2801.22

Crash dumps have been enabled so if there is an issue then we can get the dumps and logs to Dev and confirm if the API’s is not being invoked and it could be something else that the team need to help you resolve. 



 

I have uploaded 2 or 3 of these process dumps on the ticket.

Ryan Carr
Rank
Userlevel 2
Badge +5

Hi Shane 

Thanks for the dumps it was appreciated. Those dumps were used to help identify that the API was causing the issues. 

As per the engineer’s findings lets get everything on the same version and then we can monitor and make sure the jobs are running successfully. 

If any jobs fail then notify the engineer of the time frame and they can help you and diagnose the logging. 

 

S
Rank
Userlevel 3
Badge +8

Hi Shane 

Thanks for the dumps it was appreciated. Those dumps were used to help identify that the API was causing the issues. 

As per the engineer’s findings lets get everything on the same version and then we can monitor and make sure the jobs are running successfully. 

If any jobs fail then notify the engineer of the time frame and they can help you and diagnose the logging. 

 

Thanks Ryan

I’ve updated all the gateways, will monitor and advise. 

Mike Struening RETIRED
Rank
Userlevel 7
Badge +23

So the issue remains….. @Mike Struening  Can't you just disable the auto updating of the MAs or will Metallic update clients or specifically MAs automatically because the normal advice has always been to make sure the MAs run the same version as the CommServe.

It’s part of a schedule, yes.  Otherwise they wouldn’t get the right updates on time (consistently).

https://www.linkedin.com/in/michael-struening
S
Rank
Userlevel 3
Badge +8

All jobs went well last night, except for one unrelated failure, so far so good!

Ryan Carr
Rank
Userlevel 2
Badge +5

Good to hear Shane 

I’ve notified the engineer assigned to the ticket to let him know. 

As Mike mentioned the Gateways get updated as part of a weekly schedule. This includes critical patches such as the one we had from your site with the API so other customers don't get affected. 

But again, glad to see the jobs are going through successfully. 

let me know if you need anything further on this thread. 

Ryan  

Terms & ConditionsCookie settings