nop 4.1 keeps restarting

Howdy,

I've now run out of ideas and really hope someone here can point me in a direction.

My production nop 4.1 installation on one of my stores restarts sometimes every 60 seconds. While the application starts up again it will often time out for admins using it (we are soft live only so no customers are using it yet). It only seems to happen when I have admin users online capturing products. It does't stop after hours, so its not a scheduled task issue.

I am hosting in azure (serverless)
Always on is turned on and working, it doesnt shut down at night when no one is using it.

There are no errors in the log, just a lot of "Application Started" logs. No errors are getting logged in Azure Application Insights either. I've checked the "diagnose and solve problems" section on my web app in azure and the only restarts that show up there are manual restarts of the web app, so I dont think the application pool is crashing.

It seems a lot of people usually suggest turning off the shadowCopy setting, I've tried that with no success.

The annoying part is I cant reliably reproduce the issue, i'm not sure which action actually shuts nop down.
I've even backed up the database and file system to run on my local machine and cant get it to die there while attached to the debugger.

I've thought maybe something is calling the RestartAppDomain method, upon investigation I saw that works by changing the LastWriteDateTime of the web.config file. I've confirmed that the file on disk does not get updated when these restarts happen so its not that either.

We are running a few stores and this is the only one that has this issue. The only real difference is that this store has a purchased theme (Urban) instead of a hand built one.

Any ideas for things I can check?

Seems like IIS w3pw process restarting itself when timeout is happeing. Check CPU percentage at the time of restart.

mhsjaber wrote:


Servers are idling, both compute and sql. Ive scaled both up just in case with no joy either.

HarveySaayman wrote:

Did you enable the Log from the webconfig?
stdoutLogEnabled="true".
Also, run bellow query and share the both result
select * from [Log]  where LogLevelId=40

sina.islam wrote:

I'll give this a try, configured it. Just have to wait for it to happen again.

Thanks

HarveySaayman wrote:

Keep the log enabled for dome days. The log file size increased very rapidly, even by running 10-15 days the log can be reached at some GB .

Hi Harvey,

Just come across your issue with your application keeps restarting, I had a similar issue with my 4.2 installation I found changing some of the settings within the application pool helped alot in-fact made the site run a bit quicker:

• Enable 32-Bit applications: True
• Identity: NetworkService
•  Idle timeout: 10
• Load user profile: False
• Disable overlapping recycle: True

Hope this helps