[BUG] Frequent 502 Bad Gateway

[beto-Sibileau]

Describe the bug
Usual 502 response when using GeoRepo UI or API

• May have either Azure Application Gateway description:

• Or nginx:

To Reproduce
Difficult to say what may be triggering the behavior, given below datetimes to help searching in Log:

• Friday, 10 May 2024 13:29 (CET)
• Thursday, 9 May 2024 19:58 (CET)
• Wednesday, 8 May 2024 16:21 (CET)
• Tuesday, 7 May 2024 11:55 (CET)
• Monday, 6 May 2024 18:42 (CET)

Service is restablished after waiting, sometimes around 10/15 minutes

[danangmassandy]

Hi @beto-Sibileau, I will check on this, I have discussed with @fcproj about several reason that can cause the pod is failing. I will keep you updated on this ticket.

[danangmassandy]

Updates: after production deployment on May 21, Django and Nginx in production are running fine without any crash.
I only improved the cache configuration for Nginx container (ticket #1161). There is no configuration change for django since I didn't find the root cause for out of ephemeral storage.

cc: @fcproj