Re: monit (latest CVS checkout) crashes

[Martin Pala]

It could help, when you have core of crashed process (note that you can 
use 'ulimit -c unlimited' command before actual test attempt to unset 
possible core size limit). When you have core, you can use following 
sequence to trace the stack:

1.) run gdb with monit binary and core as arguments:
  gdb /path/to/monit /path/to/core

2.) show backtrace - on gdb prompt:
(gdb) bt

... now you will see what preceeded the crash ... this output may help 
us to fix the problem quickly.

(if you are running solaris, you can just use pstack command on core)

You can also can look on short debug howto in monit FAQ.txt. It 
describes how to trace the system calls - these are useful informations 
as well (solaris alternative is truss).

Thanks,
Martin


Nowak, Bernhard wrote:
> Hi all,
>
> I just compiled monit from CVS because I wanted to test the new URL and HTTP 
> content monitoring feature.
>
> I use monit to monitor two services (aaa and bbb) and restart them max two 
> times. If that happens I want to switch over to standby system (by stopping 
> heartbeat, monit was started before by heartbeat..I know, the manual tells me 
> to do it the other way around...but I want it that way).
> As long as the "Event ratio dependant action rules" are not yet supported I will do a 
> workaround monitoring the monit service pages for the words "not monitored" (ccc_mont and 
> ddd_monit).
>
>
> Unfortunately monit crashes.
>
> /etc/init.d/monit status gives me a "monit dead but pid file exists".
> Last line of monit.log is: monit HTTP server started"
>
> I also tried to use monit -c /etc/monit.conf -v. This one started monit (with 
> all the debug messages), but also crashed after some seconds.
>
> I attached my config as well.
>
> I'm no developer, just user.
>
> Thanks,
>
> Bernhard
>
> ***************************************************
>
> set daemon  30     
> set logfile /var/log/monit.log
> set mailserver localhost
> set mail-format         
>   { from: address@hidden 
>  set alert address@hidden 
> set httpd port 2812 and  
>      allow localhost      
>      allow 192.168.9.41
>      allow 192.168.10.171
>
> #------services to 
> watch-----------------------------------------------------------
>   check host aaa with address 127.0.0.1
>      start program = "/etc/init.d/ABC start"
>      stop program  = "/etc/init.d/ABC stop"
>      if failed port 20001 with timeout 60 seconds then restart
>      if 2 restarts within 20 cycles then timeout
>
>
>   check host bbb with address 192.168.10.173
>      start program = "/etc/init.d/DEF start"
>      stop program  = "/etc/init.d/DEF stop"
>      if failed host 192.168.10.173 port 8080 protocol http with timeout 60 
> seconds then restart
>      if 2 restarts within 20 cycles then timeout
>
> #-----use mont http content feature to check for "not monitored"  -------
>
>   check host ccc_monit with address 192.168.10.173
>   if failed url http://192.168.10.173:2812/aaa
>      and content != "not monitored" with timeout 60 seconds
>      then exec "/etc/init.d/heartbeat stop"
>
>
>   check host ddd_monit with address 192.168.10.173
>   if failed url http://192.168.10.173:2812/bbb
>      and content != "not monitored" with timeout 60 seconds
>      then exec "/etc/init.d/heartbeat stop"
>
> ***************************************************************************************************
> The information in this email is confidential and may be legally privileged.  
> Access to this email by anyone other than the intended addressee is 
> unauthorized.  If you are not the intended recipient of this message, any 
> review, disclosure, copying, distribution, retention, or any action taken or 
> omitted to be taken in reliance on it is prohibited and may be unlawful.  If 
> you are not the intended recipient, please reply to or forward a copy of this 
> message to the sender and delete the message, any attachments, and any copies 
> thereof from your system.
> ***************************************************************************************************
>
>
> --
> To unsubscribe:
> http://lists.nongnu.org/mailman/listinfo/monit-general