SERVICE ALERT shows in nagios log sporadically

[jagden]

Yes, I do get the following:

[1446616754] EXTERNAL COMMAND: PROCESS_SERVICE_CHECK_RESULT;myhost;tst2;2;helloworld\n
[1446616755] PASSIVE SERVICE CHECK: myhost;tst2;2;helloworld

But no SERVICE ALERT or NOTIFICATION

Attached the log file, the service name I am checking is "tst2"

[jdalrymple]

Your debug log doesn't start until after the event:

Code: Select all

jdalrymple@localhost ~Desktop
$ head nagios.debug
[1446617385.710566] [001.0] [pid=24840] drop_privileges() start
[1446617385.710603] [004.0] [pid=24840] Original UID/GID: 501/501
[1446617385.710677] [064.1] [pid=24840] Making callbacks (type 9)...
[1446617385.710743] [064.1] [pid=24840] Making callbacks (type 9)...
[1446617385.710798] [064.1] [pid=24840] Making callbacks (type 9)...
[1446617385.710809] [064.1] [pid=24840] Making callbacks (type 7)...
[1446617385.715142] [064.1] [pid=24840] Making callbacks (type 9)...
[1446617385.720982] [064.1] [pid=24840] Making callbacks (type 9)...
[1446617385.721067] [064.1] [pid=24840] Making callbacks (type 9)...
[1446617385.721125] [064.1] [pid=24840] Making callbacks (type 9)...

jdalrymple@localhost ~Desktop
$ grep helloworld nagios.debug

Incidentally, I repeated my trial on a 3.5.0 system, I could not recreate your issue there either, although I will say that the SERVICE ALERT log entries were slow to arrive as compared to my 4.0.8 box.

Any chance you could run a current version?

[jagden]

do grep on tst2
which was the string I've sent.
I am running a production system which will take some time to upgrade therefore I am trying to figure out if possible to make it work on 3.5.0

[jdalrymple]

I simply can't find anything related to your problem in old forum posts, and I cannot recreate it on any of my test systems, including a 3.5.0 configured identically. I get consistent ALERT notifications in the nagios.log

Can you verify that your availability reports aren't reflecting proper data? It should not - if it does then there is some nagios.log trickery at play.

[jagden]

Everything looks ok.
We have around 300 servers.
Perhaps it got something to do with that ?
Availability reports and UI looks ok. The service status and text get updated correctly. It is just that the output of the nagios log does not reflect the alert/notification.
Anyhow, I think I will push for a version upgrade and re-test with that.
I also did tests on an isolated machine with newer version and could not reproduce.

Thank you.

I simply can't find anything related to your problem in old forum posts, and I cannot recreate it on any of my test systems, including a 3.5.0 configured identically. I get consistent ALERT notifications in the nagios.log

Can you verify that your availability reports aren't reflecting proper data? It should not - if it does then there is some nagios.log trickery at play.

[rkennedy]

It could be your system - what are the specs of it, and how many service checks are you running?