Re: AW: AW: [Nagios-devel] Nagios 2.0b4 restart issue
Posted: Tue Nov 29, 2005 4:32 am
Husch Andreas wrote:
>>>there are many possible causes for Error 127, but in Nagios 2.x the
>>>most common problem is that two Nagios instances are running at the
>>>same time!
>>
>>
>>No it's not. Multiple instances has absolutely nothing to do with what
>
> the shell returns when asked to run a program. In this case, it does
>
>> result = execve(argv[0], argv, envp);
>> /* execve() only returns on errors, so hint what went wrong */
>> exit (-errno);
>>
>>and since ENOENT == 2 this results in a return code of -127 on all
>
> architectures that implement the two-complement bit pattern.
>
> Hi Andreas,
>
> thats right. But there are cases where the useres get Error 127 when
> they reload with the init script. If they do a clean stop and start, the
> error is gone away.
That's as may be, but unless the init-script does funny things with the
resource.cfg file it can't possibly be related.
> Im not really a C coder an i had no look at the source for this (which i
> should had), but maybe $USER1$ ist not filled right in this cases?
Supposing it reads config from the same files it will be set identically
(so long as no-one has made any changes, but then those changes would
show up even if just one process was running).
> So
> the Path to the plugin is wrong and as a result we get Error 127 from
> the shell. Just a stupid guess.
>
> btw: I've done a test on my 2.0b3 production system (if this is fixed in
> a newer version forget everything i wrote):
If you're going with the 2.0b series and want to report a bug, you
should make sure you're running the latest from CVS before doing so.
There's a pretty good chance that the problem has been fixed in the main
trunk after all.
> If i hit "/etc/init.d/nagios
> reload", Nagios immediately detects my hole network as DOWN because ALL
> checks result in an error 127. If i do a clean stop and start everything
> works ALLWAYS fine.
>
Something else is broken on that system. I can't say what it is without
looking at it fairly closely but I have neither interest nor intention
of doing so.
--
Andreas Ericsson [email protected]
OP5 AB www.op5.se
Tel: +46 8-230225 Fax: +46 8-230231
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]
>>>there are many possible causes for Error 127, but in Nagios 2.x the
>>>most common problem is that two Nagios instances are running at the
>>>same time!
>>
>>
>>No it's not. Multiple instances has absolutely nothing to do with what
>
> the shell returns when asked to run a program. In this case, it does
>
>> result = execve(argv[0], argv, envp);
>> /* execve() only returns on errors, so hint what went wrong */
>> exit (-errno);
>>
>>and since ENOENT == 2 this results in a return code of -127 on all
>
> architectures that implement the two-complement bit pattern.
>
> Hi Andreas,
>
> thats right. But there are cases where the useres get Error 127 when
> they reload with the init script. If they do a clean stop and start, the
> error is gone away.
That's as may be, but unless the init-script does funny things with the
resource.cfg file it can't possibly be related.
> Im not really a C coder an i had no look at the source for this (which i
> should had), but maybe $USER1$ ist not filled right in this cases?
Supposing it reads config from the same files it will be set identically
(so long as no-one has made any changes, but then those changes would
show up even if just one process was running).
> So
> the Path to the plugin is wrong and as a result we get Error 127 from
> the shell. Just a stupid guess.
>
> btw: I've done a test on my 2.0b3 production system (if this is fixed in
> a newer version forget everything i wrote):
If you're going with the 2.0b series and want to report a bug, you
should make sure you're running the latest from CVS before doing so.
There's a pretty good chance that the problem has been fixed in the main
trunk after all.
> If i hit "/etc/init.d/nagios
> reload", Nagios immediately detects my hole network as DOWN because ALL
> checks result in an error 127. If i do a clean stop and start everything
> works ALLWAYS fine.
>
Something else is broken on that system. I can't say what it is without
looking at it fairly closely but I have neither interest nor intention
of doing so.
--
Andreas Ericsson [email protected]
OP5 AB www.op5.se
Tel: +46 8-230225 Fax: +46 8-230231
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]