How did you make the build?
Swarm is unlike docker-compose: "a deploy is a deploy" - and NOT a build! (you need to build the images before you deploy them - and the (probably available) compose ones' might not meet the requirements...)
docker stack deploy -c your-compose-style-code.yml <name-of-your-swarm>
you are deploying an ...
Overriding [www] pool values through custom .conf files:
It is possible, and @breaking-not-so-bad explanation of how config files are loaded is great, I recommend you read it first.
But it is definitely worth emphasising the importance of config files naming (as it can be tricky):
As @hroj3e explained and @istranger exemplified in the comments above
If the ...
We had almost this exact same issue last year.
Raising the max children will only offset the problem to a later date.
It turned out to be a slow MySQL database hosted on a dedicated server on our network for a blog.
Our PHP was configured to try connecting for 30 seconds and whenever this database decided to act up then it would chew up 100 PHP children.
As you can see in your status output
you have a task waiting to be started (5 active, 0 idle, 6 tasks). As you posted in your own answer (and I'm glad it worked) increasing the number of allowed children can be a good solution - but there's a lot that goes into optimizing php-fpm, and certainly more thought should be given to the entire system before making ...
There may be several reasons for this behavior:
If this web server processes requests from the external network, then with the increase in the amount of traffic, the load could increase, which led to an increase in the server response time.
If your scripts use calls to external resources, then in this case the response time of your server may increase due ...
I think I've found solution, but still if you have any suggestions please let me know or post another answer.
I checked /var/log/php7.4-fpm.log and saw this many entries like this:
[30-Sep-2021 03:36:46] WARNING: [pool www] server reached
pm.max_children setting (5), consider raising it
So I raised the max_children to 15 and it seems it helps.
If you experience the same error on different public apps, it may indicate an issue with GKE. You can report it via Issue Tracker.
If this error is specific to this app, consider different logging aggregator, for example the sidecar pattern.
There's more information regarding logging on GKE in this document.
Additionally, if you are using cOS with Docker, ...
In situations like this, I find that it helps to throttle the number of connections higher up the stack.
When more than, say, 100 active processes are running, they stumble over each other. They are vying for resources (CPU, etc). The net effect is that all processes run slower, sometimes to the point where you feel like the only solution is to reboot the ...