2

This ASP.NET Core Azure App Services website has been running in production (unchanged) for months with CPU averages of 1-3%. Starting yesterday I noticed that the CPU all of a sudden went to 100%:

cpu at 100%

I couldn't immediately figure out what was causing this, so I just rebooted the service. Rebooting the instance causes the CPU to go back to the normal 1-3% average, however, as time passes the CPU continually jumps by factors of 25%, stays at that level for 30-40 minutes, then jumps another 25%. This happens until the CPU eventually reaches the 100% mark again.

cpu climbing to 100%

I can restart the instance over and over and the same process shown above keeps happening.

Kudu Process explorer shows this, which I guess seems to indicate that my application (Api) is taking up the CPU:

Kudu process explorer

How do I troubleshoot this? I have monitored Application Insights (can't install New Relic on ASP.NET Core apps) and other app service charts and I do not see anything out of the ordinary happening at the different timestamps of CPU interest. HTTP traffic to the service seems normal and steady.

kspearrin
  • 463
  • 1
  • 4
  • 8

1 Answers1

1

I could never figure this issue out so I ended up creating an entirely new app service plan and web app, migrated my website over to that, changed DNS and I have not had a problem sense (24 hours now). I don't know what was wrong with that other instance, but it clearly was not my application code.

kspearrin
  • 463
  • 1
  • 4
  • 8