Question Plesk Obsidian - 18.0.36 and nginx worker connection issues

[GregFowler]

Hi Guys,

I recently took a server from Plesk Obsidian 18.0.35 to .36, everything appeared to be fine but on a daily basis the sites go offline for 10/15 minutes at a time and return a 500 internal nginx error.

Looking at the nginx logs I can see the below entry continually in the error_log

2021/06/16 06:23:26 [alert] 54492#0: 1024 worker_connections are not enough
2021/06/16 06:23:26 [alert] 54492#0: 1024 worker_connections are not enough
2021/06/16 06:23:26 [alert] 54492#0: 1024 worker_connections are not enough
2021/06/16 06:23:26 [alert] 54492#0: 1024 worker_connections are not enough
2021/06/16 06:23:26 [alert] 54492#0: 1024 worker_connections are not enough
2021/06/16 06:23:26 [alert] 54492#0: 1024 worker_connections are not enough
2021/06/16 06:23:26 [alert] 54492#0: 1024 worker_connections are not enough
2021/06/16 06:23:26 [alert] 54492#0: 1024 worker_connections are not enough
2021/06/16 06:23:26 [alert] 54492#0: 1024 worker_connections are not enough
2021/06/16 06:23:26 [alert] 54492#0: 1024 worker_connections are not enough
2021/06/16 06:23:26 [alert] 54492#0: 1024 worker_connections are not enough

Looking through old logs the first occurence happened at:

2021/06/14 08:17:03 [alert] 37837#0: 1024 worker_connections are not enough

The only change that was made was updating Plesk to 18.0.36 which completed at 08:11 the same day so I can only conclude its related to this somehow.

Any pointers?

 

[Bitpalast]

What is worker_connections set to in /etc/nginx.conf? I think the default is 1024.

 

[GregFowler]

It is set to the default of 1024. I tried to double it but the same problem, still getting worked connections are not enough.

No errors in the log to this effect until the update, not sure if this this just a coincidence?

 

[Ehud]

Hi,

I also face problems since upgrade to latest version this morning. Some PHP now take 5-15 minutes to "run" instead of 10 seconds (PDF generation on the fly), and also have the BUG of not fetching local SERVER side images.

Issue might be related to a change of some memory configuration on latest Plesk version, not known to users.

I have contacted the Plesk Support about 2 hours ago, however still haven't received response.

Currently, Plesk support do not see a connection between my issue, to the latest Plesk version upgrade, however did not identify my root cause.

 

[mow]

Wait, isn't error_log the apache error log?

 

[GregFowler]

Sorry I meant error.log as opposed to error_log.

This is definitely nginx. I have no further outages but the error remains and still no further forward regarding root cause.

 

[GregFowler]

Anybody have any idea?

Could it be related to this:

Module ngx_stream_proxy_module

 

[SLOposum]

Anybody have any idea?

Could it be related to this:

Module ngx_stream_proxy_module

Hey @GregFowler, how did you solve this? Having same problems.