1.3.5 wrong host status.

[KenvM]

Did you make sure to leave the agent disabled long enough for zabbix to realize and report that it was not working?

If so, then I'm out of ideas..

-Ken

[azurit]

when fixed ?

having the same problem with 1.4.3 (not yet stable). will this be fixed soon ?

[brainsik]

I've been able to fix it by shutting down the agent long enough for zabbix to realize that its off and then starting the agent back up.

Doing this "fixed" the problem for me as well. Zabbix 1.4.4 on Ubuntu 7.10.

[Philippe]

hello,

what agent do you mean ?
I have this with hosts that are not attached to an agent.
simple check (icmpping) for example.

grtz,
Ph.

[Idriel]

I got some problem with icmpping on router.
Version 1.4.4

Status: Monitored
Availability: Unknown
Error:

[Jason]

Same problem on 1.4.5 when talking to windows hosts behind a firewall... It was fine when I had the agent in passive mode with port 10050 opened in firewall, but stopped working when I moved to active checks... Now I just get

Timeout while connecting to [XXX.XXX.XXX.XXX:161]

I've tried the fix of stopping the agent, but no joy... I've only active checks on these servers, no snmp checks.

Jason

[mohamed_al_gad]

no data

Hi, I had similar problem with the host status (with Agent), after some tests I noticed different behaviour depending on the acknowledgement of the events.

Example :
Zabbix agent stopped
Zabbix notice this and raise a trigger (Server {HOSTNAME} is unreachable)
I acknowledged it
...some minutes later...
Zabbix agent started
...some minutes later...
Zabbix shows "unknown" for "Server {HOSTNAME} is unreachable"
...some minutes later...
No change...

Zabbix agent stopped
Zabbix notice this and raise a trigger (Server {HOSTNAME} is unreachable)
I !!! DIDN'T !!! acknowledged it
...some minutes later...
Zabbix agent started
...some minutes later...
Zabbix shows "Server {HOSTNAME} is unreachable" in green, ok

I did the same test twice and got different results, problematic items were the only difference, first time it was "Server {HOSTNAME} is unreachable", then it was "Too many processes running on {HOSTNAME}" and another, but not the unreachable one...

I dont explain (as I am a Zabbix-noob :-) ), but just noticed that...

So KenvM's remark solved my problem as long as I didn't acknowledge the event.

I run zabbix 1.6.1 for information

[ashuji]

Similar Problem

Hi,

I am also facing similar problem, in my case, ITEM for 'Host status' is being updated correctly but this trigger does not go. I stopped/started xinetd on monitored host and than after triggers went away. I am using zabbix 1.62, is this bug still there in new release as well.

Ashwani Jain