Hello!

On Sat, Feb 02, 2019 at 01:21:12AM +1100, Mathew Heard wrote:

> >> If you've seen a
> >> percentage of connections being dropped for some time - likely
> >> there is another problem elsewhere.
>
> That's definitely what I observed. It was around 50% of this customers
> connections and strace on all workers (including the shutting down worker)
> did not show the missed connections at the accept level (grep on unique
> testing IP).

On Linux, "ss -nltp" can show important details on listening
sockets and how they are handled, including the number of
connectionion requests in each socket's backlog, as well as
processes having the socket open (note: run under root to get the
list). You may want to use it next time you'll see a problem.

> The only thing strange I was able to note was the one process remaining in
> "worker is shutting down state" (it's not uncommon for us to have a few
> workers hanging around for a while due to websocket or similar connections
> keeping workers open). This is why I formulated this theory.
>
> Further reloads did not resolve the issue, it took a restart of the nginx
> process to get everything back to normal.
>
> As far as I am aware no other nginx process was started on the server
> (systemd manages nginx).

Well, "ps -alx | grep nginx" output is much more definitive than
"systemd manages nginx".

Depending on the actual systemd configuration used, it certainly
can easily result in multiple nginx instances being started. Not
to mention that starting another instance directly is certainly
possible too.

--
Maxim Dounin
http://mdounin.ru/
_______________________________________________
nginx-devel mailing list
nginx-devel@nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx-devel