read() returned error 11

[arenist]

Hi,

those strange things DrydenK described in https://support.nagios.com/forum/viewto ... =7&t=39856 are happening to me, too. I'm running nagios 4.2.4 on Oralce Linux Server release 6.8.

First of all, I don't understand why nagios is writing to /var/log/messages. This should be forbidden by use_syslog=0 in nagios.cfg. After setting debug_level=4 nagios does not write anything into nagios.debug. A possible reason for this behavior might be, that nagios has detected the service to be flapping.

The service that produces this output to /var/log/messages should check the syslog for the occurence of the string "error". Nagios is allowed to read syslog:
-rw-r--r-- 1 root root 3903114 Dec 15 11:33 /var/log/messages

While nagios is writing continously this blocks of 12 errors
Dec 15 11:21:53 nagsrv01 nagios: job 157 (pid=11009): read() returned error 11
Dec 15 11:21:53 nagsrv01 nagios: job 157 (pid=11013): read() returned error 11
Dec 15 11:21:53 nagsrv01 nagios: job 157 (pid=11015): read() returned error 11
Dec 15 11:21:53 nagsrv01 nagios: job 158 (pid=11019): read() returned error 11
Dec 15 11:21:53 nagsrv01 nagios: job 157 (pid=11010): read() returned error 11
Dec 15 11:21:53 nagsrv01 nagios: job 158 (pid=11016): read() returned error 11
Dec 15 11:21:53 nagsrv01 nagios: job 157 (pid=11014): read() returned error 11
Dec 15 11:21:53 nagsrv01 nagios: job 157 (pid=11011): read() returned error 11
Dec 15 11:21:53 nagsrv01 nagios: job 158 (pid=11018): read() returned error 11
Dec 15 11:21:53 nagsrv01 nagios: job 158 (pid=11017): read() returned error 11
Dec 15 11:21:53 nagsrv01 nagios: job 157 (pid=11008): read() returned error 11
Dec 15 11:21:53 nagsrv01 nagios: job 157 (pid=11012): read() returned error 11
it will find the string "error" of course. And so nagios makes the service flapping itself.

install environment:
Oralce Linux Server release 6.8, fully updated
Nagios Core 4.2.4 installed from tar-ball
Server with 16GB RAM, 8 CPU cores

How can I prevent nagios from writing into my syslog???

Thanks in advance,
arenist

[avandemore]

First of all, I don't understand why nagios is writing to /var/log/messages. This should be forbidden by use_syslog=0 in nagios.cfg

That may be a seperate bug.

For your problem, see here:
https://github.com/NagiosEnterprises/na ... issues/172

[dwhitfield]

Also, just to clarify, are you using mod_gearman?

[arenist]

Hi avandemore,

sorry, but I can't find a solution of my problem in the link you sent me. jfrickson says, he has

changed it to LOG_NOTIC in branch 'maint'

. Whatever this means, I actually DO see an error in my syslog.

@dwhitfield: I'm not using mod_gearman. I had to google what mod_gearman is ...

Best regards,
arenist

[avandemore]

It means the issue has been fixed. You can either wait for the next version of Nagios Core to be released, or install from source using the maint branch.

[arenist]

Ah, thanks for your explanation avandemore. I will try to install from source. So you can close this thread. But it was really ugly that nagios is writing exactly that string into a file which I want to be noticed of.

Best regards,
arensit

[dwhitfield]

I will try to install from source.

be sure to use the maint branch...or just wait. Unfortunately, I don't have an ETA on when that will be. I know the devs are busy getting the Core version in XI updated to 4.2.4.

[avandemore]

We encourage community participation on GitHub so that things like this can be avoided.