Count number of hosts not reachable in group

[[email protected]]

Hello,

We want to count the number of hosts in a certain group that are not reachable; and raise a trigger if that number exceeds a threshold.

We want to use a groupfunction (grpsum) for that.

The item would be like (this worked fine in 1.8.2)
grpsum["SERVERS.LINUX","status","last","0"]
The trigger would be like
{xxxxx:grpsum["SERVERS.LINUX","status","last","0"].min(#2)}/2>5

I noticed that field status got value 2 (in zabbix 1.8.2) if a host is not reachable. But in 1.8.3 this is not the case anymore.

Is there another way to reach this goal ?

[richlv]

status should work pretty much the same with 1.8.3 - what issues are you seeing with it ?

[[email protected]]

You are right; that is what I expected when we upgraded to 1.8.3:
Status 2 means not reachable
Status 0 means reachable

But now in 1.8.3, I have a lot of hosts that are reachable but have status 2.
Also, a lot of hosts that are reasonable have status 0 (what I would expect).
I haven't figured out what the reason for this strange behaviour.

During the upgrade from 1.8.2 (24 hours ago) all hosts could not be reached during a short period of time. All hosts got status 2.
But now all hosts have been reachable since 24 hours (zabbix is retreiving data from those hosts), but not all of them have been reset to 0.
I would expect them to be reset to 0 within minutes.

[richlv]

i actually recall such a problem being fixed for 1.8.3...
are you sure it's 1.8.3 server that's running ?
what does the server logfile say ?

[[email protected]]

In the webinterface/GUI I see 1.8.3

On the server:
svr-zabbix:/etc/zabbix/bin# ./zabbix_server -h
Zabbix Server v1.8.3 (revision 13928) (16 August 2010)

I am sure that must be 1.8.3 running.

[richlv]

just to be completely sure, can you check the logfile ?

[[email protected]]

An extract of the logfile

4981:20100901:152542.457 Expression [{138714}>0&{138713}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.463 Evaluation failed for function: nodata
4981:20100901:152542.464 Expression [{138734}>0&{138733}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.474 Evaluation failed for function: nodata
4981:20100901:152542.474 Expression [{140811}=1&{140810}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.478 Evaluation failed for function: nodata
4981:20100901:152542.479 Expression [{168100}=1&{168099}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.482 Evaluation failed for function: nodata
4981:20100901:152542.483 Expression [{175741}=1&{175740}=0&{175739}>010000] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.486 Evaluation failed for function: nodata
4981:20100901:152542.487 Expression [{140821}=1&{140820}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.490 Evaluation failed for function: nodata
4981:20100901:152542.491 Expression [{168102}=1&{168101}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.494 Evaluation failed for function: nodata
4981:20100901:152542.494 Expression [{175744}=1&{175743}=0&{175742}>010000] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.500 Evaluation failed for function: nodata
4981:20100901:152542.501 Expression [{141002}>0&{141001}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.508 Evaluation failed for function: nodata
4981:20100901:152542.508 Expression [{141004}>0&{141003}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.515 Evaluation failed for function: nodata
4981:20100901:152542.515 Expression [{141085}>0&{141084}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.524 Evaluation failed for function: nodata
4981:20100901:152542.525 Expression [{141088}+{141087}>0&{141086}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.531 Evaluation failed for function: nodata
4981:20100901:152542.531 Expression [{141090}>0&{141089}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.539 Evaluation failed for function: nodata
4981:20100901:152542.539 Expression [{155228}=1&{155227}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.546 Evaluation failed for function: nodata
4981:20100901:152542.546 Expression [{155157}>0&{155156}=0&{155155}>0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.557 Evaluation failed for function: nodata
4981:20100901:152542.557 Expression [{155348}=1&{155347}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.567 Evaluation failed for function: nodata
4981:20100901:152542.567 Expression [{180828}>100&{180828}<200&{180827}=0] cannot be evaluated: Evaluation failed for function: nodata
4981:20100901:152542.575 Evaluation failed for function: nodata

[richlv]

no, i meant server version - what is it for the last entry of server startup ?

[[email protected]]

We have restarted the zabbix_server 1/2 hour ago.
where can I find the last entry of server startup ??

The extract is from /tmp/zabbix_server.log

[[email protected]]

I thnink that this is what you asked for

21437:20100901:211944.186 Starting Zabbix Server. Zabbix 1.8.3 (revision 13928)
.
21437:20100901:211944.186 **** Enabled features ****
21437:20100901:211944.186 SNMP monitoring: YES
21437:20100901:211944.186 IPMI monitoring: NO
21437:20100901:211944.186 WEB monitoring: YES
21437:20100901:211944.186 Jabber notifications: YES
21437:20100901:211944.186 ODBC: NO
21437:20100901:211944.187 SSH2 support: NO
21437:20100901:211944.187 IPv6 support: NO
21437:20100901:211944.187 **************************
21627:20100901:211957.492 server #1 started [DB Cache]
21629:20100901:211957.590 server #3 started [Poller. SNMP:YES]
21628:20100901:211957.604 server #2 started [Poller. SNMP:YES]
21630:20100901:211957.628 server #4 started [Poller. SNMP:YES]
21631:20100901:211957.656 server #5 started [Poller. SNMP:YES]
21633:20100901:211957.708 server #7 started [Poller. SNMP:YES]
21634:20100901:211957.750 server #8 started [Poller. SNMP:YES]
21635:20100901:211957.759 server #9 started [Poller. SNMP:YES]
21632:20100901:211957.761 server #6 started [Poller. SNMP:YES]
21650:20100901:211957.770 server #24 started [Trapper]
21636:20100901:211957.779 server #10 started [Poller. SNMP:YES]
21648:20100901:211957.793 server #22 started [Trapper]
21649:20100901:211957.794 server #23 started [Trapper]
21651:20100901:211957.805 server #25 started [Trapper]
21637:20100901:211957.813 server #11 started [Poller. SNMP:YES]
21652:20100901:211957.814 server #26 started [Trapper]
21638:20100901:211957.833 server #12 started [Poller. SNMP:YES]
21653:20100901:211957.837 server #27 started [ICMP pinger]
21655:20100901:211957.838 server #29 started [Housekeeper]
21655:20100901:211957.838 Executing housekeeper
21657:20100901:211957.839 server #31 started [HTTP Poller]
21640:20100901:211957.840 server #14 started [Poller. SNMP:YES]
21639:20100901:211957.845 server #13 started [Poller. SNMP:YES]
21654:20100901:211957.857 server #28 started [Alerter]
21656:20100901:211957.858 server #30 started [Timer]
21641:20100901:211957.864 server #15 started [Poller. SNMP:YES]
21643:20100901:211957.884 server #17 started [Poller. SNMP:YES]
21644:20100901:211957.894 server #18 started [Poller. SNMP:YES]
21659:20100901:211957.895 server #33 started [DB Syncer]
21661:20100901:211957.896 server #35 started [DB Syncer]
21437:20100901:211957.899 server #0 started [Watchdog]
21642:20100901:211957.904 server #16 started [Poller. SNMP:YES]
21647:20100901:211957.906 server #21 started [Poller. SNMP:YES]
21664:20100901:211957.907 server #38 started [Proxy Poller]
21662:20100901:211957.907 server #36 started [DB Syncer]
21663:20100901:211957.908 server #37 started [Escalator]
21660:20100901:211957.913 server #34 started [DB Syncer]
21646:20100901:211957.929 server #20 started [Poller. SNMP:YES]
21645:20100901:211957.930 server #19 started [Poller. SNMP:YES]
21658:20100901:211957.940 server #32 started [Discoverer. SNMP:YES]

[[email protected]]

I have added a graph with the item that contains the sum of all statusses of a set of windows servers.
As you can see, since the upgrade from 1.8.2 -> 1.8.3 the sum of statusses has increased from almost zero to 38, indicating that 19 servers would be down (which is not the case fortunately !)

[richlv]

well, i'm a bit out of ideas. in host configuration, does availability column have any red entries for hosts reported as down ?

[[email protected]]

All machines are monitored with a zabbix agent.
All Z icons are green.
We receive data from the machines that have status 2.
But the fact that the status is still 2 is the only thing that is wrong.

[[email protected]]

I am out of options.
It worked in 1.8.2 and doesn't work anymore in 1.8.3.
Should I report this as a bug ?

[walterheck]

Yep, please do. Then also post a link here to teh bug report for potential followers.