We've had the OTD server, now workflow server, for almost 5 years but did not register all ~50 endpoints to it for calendar until yesterday. Typically we have about 5 endpoints registered to it without issue.
Every 4-6 hours, group series calendar registrations drop and when I RDP into the workflow Win2016 server, I can no longer connect to the admin web ui of the workflow app.
The only fix is to restart the "Polycom Workflow Server" service in windows. The issue is resolved for a few hours and then comes back. My thought is this is due to too many registrations but the guide says it supports 500.
Any insight or experience with a similar issue? We previously had all ~50 endpoints registered to the RPRM for calendar but switched them all over to the workflow server for maintenance.
Solved! Go to Solution.
Its ok. I finally got an answer. Memory Leak. I was advised to do the following
The most important and relevant thing is that this WFS is configured on pass-through authentication mode. There is a memory leak bug in WFS for pass-through mode, and that is not fixed yet. See details in EN-154169.
Both items can cause Node.JS memory utilization to build up and eventually service can either restart or simply hang, requiring a manual restart.
The only option (workaround) at the moment is to ask the customer to enable delegated mode following the official documentation available here https://support.polycom.com/content/dam/polycom-support/products/uc-infrastructure-support/polycom-w...
Which version are you running? I had the same exact isssue until I upgraded to 1.7.0 HF5.444445
Thanks for the response.
First 2 versions failed in the same way, 1-7-1 didn't work at all for me.
Did Polycom support provide you a hotfix, as notated in the version name you listed?
I was provided an ftp location by a polycom engineer who was working the case with me. When I visit support.polycom.com, Documents and Software, UC Infrastructure, Workflow Server - in order to download the software, you will need a password provided to you by Polycom.
I would suggest openeing a support ticket so you can get the latest HF
Ok, thanks a lot.
I do have a ticket open and will share this info.
Much appreciated. My "workaround" was a script that restarted the windows service every hour, not great.
Can you share the case number so we can link this?
Well this happened - I loaded up the server with ALL of my endpoints (32) and this exact thing happens. I too have a ticket now. I have spent 2 days staring at the WFS console and I have correlated that everytime Cylance scans the file system (this happens like 5 + times a day) The WFS server Service does NOT stop but from what I can see I THINK file scanning messes with the data base lock and WFS cannot recover. So I set a task to stop/start the wfs service ever time cylance logs that it is done scanning. This creates about 2 mins of downtime for wfs - which is better than totally down but still not really acceptable.
I will post back here when I get further. Best of luck!
Great, thanks for the update and sorry you are dealing with issues again. I definitely think this is # of endpoints related.
Good job on the workaround, same here.
Polycom support has thus far told me to update to 3 different versions, 2 of which I provided logs from when I opened the ticket initially. Not making much progress.
On to Tier3....
I'll update if I get anywhere on my case as well.
Check this out - can you look at the Environments section and go to the properties of the advanced options, look for Max Current Sessions and increase it from the defult of 10 to a value that is more than the number of endpoints you have. I set mine to 50 and I have 32 endpoints. I attached a screen shot of the location