[hobbit] No alerts propagating...

[David Gore]

RULES:

HOST=rchimpm1,rsoimpm1 COLOR=red,yellow
          MAIL somebody at somehost.com DURATION>15

HOST=lismore
         SCRIPT /export/home/rbaldwin/hobbit.sh PAGE RECOVERED REPEAT=8h

DUMP:

HOST=rchimpm1,rsoimpm1 COLOR=yellow,red
         MAIL page_somebody at somehost.com REPEAT=30 COLOR=yellow,red 
DURATION>15

HOST=lismore
         SCRIPT /export/home/rbaldwin/hobbit.sh PAGE FORMAT=SCRIPT 
REPEAT=480 RECOVERED

debug:

2005-01-29 22:48:41 criteriamatch rchimpm1:procs 
rchimpm1,rsoimpm1:(NULL):(NULL)
2005-01-29 22:48:41 Checking explicit color setting 10000000060 against 
4 gives 1
2005-01-29 22:48:41 Found a first matching rule
2005-01-29 22:48:41 criteriamatch rchimpm1:procs (NULL):(NULL):(NULL)
2005-01-29 22:48:41 failed minduration 0<900

so far so good...

2005-01-29 23:32:38 Got page message from cosimpm1:rds
2005-01-29 23:32:38 0 alerts to go
2005-01-29 23:33:38 0 alerts to go

Nothing in between, so it started well and didn't go anywhere

bummer.


David Gore wrote:
> COLOR=red
>   MAIL somebody at somehost.com REPEAT=8h DURATION>15
> 
> Is this a possible rule, or do you need BOTH a HOST and a SERVICE for 
> ANY rule?  Or do you need at least one, HOST or a SERVICE?  Can I have a 
> rule for ALL red alarms?  This is running on Solaris 9.
> 
> 2005-01-29 22:48:41 criteriamatch rchimpm1:procs 
> rchimpm1,rsoimpm1:(NULL):(NULL)
> 2005-01-29 22:48:41 Checking explicit color setting 10000000060 against 
> 4 gives 1
> 2005-01-29 22:48:41 Found a first matching rule
> 2005-01-29 22:48:41 criteriamatch rchimpm1:procs (NULL):(NULL):(NULL)
> 2005-01-29 22:48:41 failed minduration 0<900
> 
> David Gore (v965-3670)
> Enhanced Technology Support (ETS)
> Network Management Systems (NMS)
> IMPACT Transport Team Lead - SCSA, SCNA
> Page: 1-800-PAG-eMCI pin 1406090
> Vnet: 965-3676
> 
> 
> David Gore wrote:
> 
>>
>> Ok, the alerts are driving me mad. What I would like to do is send out 
>> a  mail/page for any alert that turns red and is red for 15 minutes or 
>> longer.  I assumed DURATION>15 would work?  DURATION>15m results in 
>> rather large number by the way that may indicate a bug, 'h' works and 
>> leaving off 'm' seems to parse OK.  None of these work, and there is 
>> nothing in ~/log/page.log.  Here are our rules and a dump.
>>
>> COLOR=red
>>         MAIL somebody at somehost.com REPEAT=8h
>>
>> SERVICE=conn,procs,disk,vmem EXHOST=dante,cayhp2,cayhp3
>>         MAIL somebody at somehost.com REPEAT=8h DURATION>15
>>
>> HOST=dante,cayhp2,cayhp3 COLOR=red
>>         MAIL somebody-different at somehost.com REPEAT=1d
>>
>> HOST=nmsbb-get SERVICE=pri-1 COLOR=red
>>         MAIL somebody at somehost.com REPEAT=8h DURATION<29
>>
>> COLOR=red
>>         SCRIPT /export/home/rbaldwin/hobbit.sh
>>
>> hobbit at nmsbb-get ~/server> ./bin/bbcmd --env=etc/hobbitserver.cfg 
>> hobbitd_alert --dump-config
>>
>> COLOR=red
>>         MAIL somebody at somehost.com REPEAT=480
>>
>> EXHOST=dante,cayhp2,cayhp3 SERVICE=conn,procs,disk,vmem
>>         MAIL somebody at somehost.com REPEAT=480 DURATION>15
>>
>> HOST=dante,cayhp2,cayhp3 COLOR=red
>>         MAIL somebody-different at somehost.com REPEAT=1440
>>
>> HOST=nmsbb-get SERVICE=pri-1 COLOR=red
>>         MAIL somebody at somehost.com REPEAT=480 COLOR=red DURATION<29
>>
>> COLOR=red
>>
>> We don't mind re-simplyfying just for paging on ALL red alarms that 
>> are RED for 15 minutes or longer and wacking all the other rules that 
>> do not work.  Thanks for any help.
>>
>>
>> ~ David Gore
>>
> 
> To unsubscribe from the hobbit list, send an e-mail to
> hobbit-unsubscribe at hswn.dk
> 
>