You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@httpd.apache.org by Paul Querna <pa...@querna.org> on 2011/11/11 21:55:46 UTC

2.3.15: hung worker during gracefu childl shutdown

I noticed in www.apache.org/server-status that there was a worker with
a single connection open, all the others had been gracefully closed.

However, it'd been in that state for an hour.

GDB attached to it, backtrace here:
   https://gist.github.com/be22714685f1e370f19e

No really sure why this would happen.

Ideas?

Re: 2.3.15: hung worker during gracefu childl shutdown

Posted by Stefan Fritsch <sf...@sfritsch.de>.

On Fri, 11 Nov 2011, Paul Querna wrote:

> I noticed in www.apache.org/server-status that there was a worker with
> a single connection open, all the others had been gracefully closed.
>
> However, it'd been in that state for an hour.
>
> GDB attached to it, backtrace here:
>   https://gist.github.com/be22714685f1e370f19e
>
> No really sure why this would happen.
>
> Ideas?

It seems to be a bug in mod_mbox combined with async write completion. I 
have tried downloading that URL and have aborted at 180MB of received 
data. A slower client may well take an hour for that.