You are viewing limited content. For full access, please sign in.

Question

Question

Veeam backup of Laserfiche servers - causing workflow events to be missed and running workflows to terminate

Workflow Laserfiche Import Agent Version 9 Backup and Recovery
Updated June 21, 2015
asked on June 16, 2015

We have discovered a problem with our backup process. When it runs on a repository server it seems to causes a communication problem between the workflow/subscriber server and the repository server. 
The repository server is not shut down during backup but I am seeing workflows terminating and errors indicating network communication problems. Unfortunately my catch clause in my workflow also fails to send me an email notification. 


I am also discovering documents that are imported, via Laserfiche Import agent, into the repository around the same time as backup is happening, are being skipped by the workflow that is meant to be monitoring the ‘create’ event for the given folder.


Researching into how workflow works, so I can get a better idea how we can improve our backup procedure, I read in the workflow administration help documentation that ‘The Workflow Subscriber receives notification from a Laserfiche Server when a change is made to a Laserfiche entry’.  Can someone confirm it ‘receives’ the event notifications?  If this is the case surely the Laserfiche repository server will know to reissue the notification once communications between the servers resumed so no events are missed? Can someone throw some light on the subject of what might be causing the events to be lost? 


We are planning to stop the workflow service on our WorkFlow server and the import agent service on our Import Agent server before commencing backup.  The thinking behind this approach is to stop workflows running so they won’t terminate due to communication problems between servers. Stopping the import agent will prevent documents being put in the repository during backup to prevent events being mysteriously lost. Can someone advise if this is the right approach to the problem or give me a better solution to the problem?


The background on our backup system is we use Veeam B&R to replicate our production servers to our DR site. As part of this replication a VMware snapshot is performed by Veeam when the backup starts and is deleted when the backup finishes. VM's go into a stun state when the snapshot is removed, normally this stun or freeze takes at most couple of seconds. We believe this small outage is causing the issue with Laserfiche Workflow as the workflow server loses connectivity to the Laserfiche repository.

 

FYI we are running Workflow 9.2 and Laserfiche 9.2.

 

0 0

Replies

replied on June 17, 2015

Can you post the terminating errors so we have a better understanding of what you're seeing?

0 0
replied on June 17, 2015

This error occurred on a few nights at the same time Veeam runs: 

     

"A network-related or instance-specific error occurred while establishing a connection to SQL Server. The server was not found or was not accessible. Verify that the instance name is correct and that SQL Server is configured to allow remote connections. (provider: Named Pipes Provider, error: 40 - Could not open a connection to SQL Server)"


The other thing I have seen is a workflow at the time the veeam is running starts a query activity and rather than completing it in a timely manner it instead errors 4 hours later when the server it is suppose to connect to, to perform the query on, is shutdown. Why couldn't it complete in the 4 hours before the server went down?

The ones that error at least I can find them and rectify the problem by rerunning the workflow on the entry in question. By the problem with the  problem I reported about the email notification was a problem with how I had set up this activity so I should om future I will get these email notifications.

I want to understand how the imported documents, that got created in the repository around the time the backup was happening, failed to trigger the workflow. I would have thought if there was a communication error the repository server would resend the notification to the workflow server about the creation event  once communication was reestablished between the two servers. 

 

 
0 0
replied on June 21, 2015

Can someone from Laserfiche please tell me how events (eg. changed, moved, added, deleted) are handled between a repository and workflow so I can understand why some events were never acted upon by workflow? I really want to understand how this is possible because even if there is communication problem surely once it is resolved the events would be processed and none would be skipped or lost? 

0 0
You are not allowed to follow up in this post.

Sign in to reply to this post.

Details
Related Posts
Loading ...