Product: PowerShell Universal
I upgraded my server last night for the CVE issue and found that my API became non-responsive fairly quickly. If I call it from PowerShell, it just sits there and never returns a timeout or any data. I had been on 2.7.4 prior to this (just hadn’t had a chance in recent months to upgrade it). I went back to 2.9.3 and it seems like it works for longer, but sometime in the middle of the night it started showing the same symptoms. If I restart the service (Windows Server 2019), it will work again for a time.
This has worked reliably in 2.7.4 and is really just doing a SQL call that is pretty straight forward using SQLServer module (not DBA…).
Any ideas on how to troubleshoot this and figure out what I need to do to move to the latest?
Can you provide what options you have selected for your API and API Environment? Timeouts, PS version etc
Also, I’m always open to reviewing a memory dump. When the API is in a hung state, if you capture one, I can take a look. You can upload them here: Dropbox - Submit files
@dkkazak - I can produce a 2.7.5. build in the meantime to ensure you are patched for this vuln and we can figure out the upgrade issue afterwards.
Sorry for not responding right away as I was trying to capture and do some more testing. What I discovered in my testing is that I moved to a different Azure server that has temp storage with it. When I did that, the API has been responsive once again. So, it seems like it was something related to having temp storage that seems to help and that we ran into in prior testing for other performance issues. I’m continuing to monitor, but I haven’t had it freeze yet on the new server.
No worries. Glad you’re back up and running. That’s good info and something we might need to include in the docs.
I did merge the fix into the 2.7 branch and am running a production build of that now. It will be available on our previous releases page in the next couple of hours.