Re: signal related race conditions and deadlock

[Martin Pala]

Martin Pala wrote:

> Martin Pala wrote:
>
> > Hi,
> >
> > there are SIGTERM and SIGINT related problems:
> >
> > ------
> > FIRST:
> > ------
> >
> > 1.) monit receives one of above mentioned signals and calls do_destroy()
> > 2.) do_destroy() sets http engine stopped flag and returns - monit 
> > httpd then will stop in next cycle on 'while' condition 
Sorry again (i see the errors on second sigth everytime after i have 
send the message) - the mentioned return is from stop_http() (which sets 
the stopped flag) called inside do_destroy(), which will of course 
continue and call garbage collector and close files as described bellow 
(i have not mentioned that because of freed memory monit could crash 
when httpd will access it in the short timeframe before exit). In the 
case that monit will reach exit sooner that httpd will close socket, it 
could probably cause another problems (httpd will not end gracefully).

> > 3.) monit will close all files and logs (asynchronously of monit 
> > httpd task)
> >
> > Problem, is that monit httpd thread is able to log messages, while it 
> > is possible that the socket will be closed earlier. 
>
>
> Sorry - instead of closed "socket" i meant "filedescriptor" associated 
> with logs.
>
> ...
>
>
>
>
> _______________________________________________
> monit-dev mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/monit-dev