HP-UX 11.23 & 11.31 some parameters implementation

[Sobakovod]

Provided "as is" for 1.6.6

put files from archive in src/libs/zbxsysinfo/hpux

compile

implemented proc.num, proc.mem, system.uptime|boottime, vm.memory

Best regards

PS: on demand I can upload compiled agents for itanium

[orogor]

Thanks a lot

I am always eager for compiled hp-ux agents and new functionalities, supposedly they re going to provide a recent agent version for hp-ux , but it s still not there (devs do you hear me?). Basically i have itanuim and risc hp-ux, both latest versions.

These counters were exactly the ones i wanted missed, most importantly proc.num because i have some specific bug related to duplicated instances of tnslistener on oracle.

I am also looking for a reliable way to monitor the print queues on these system , because it happens on a daily basis that a queue get stuck for np good reason (i guess a bugged protocol response from the printer), and the support from nor hp-ux (who want to support hp printers only) nor kyocera (who doesn t want to support hp-ux) are helping me.

If you have some monitoring stuff for oracle on hp-ux, specifically rman thats also good. (i am unsure tail -n 10|grep -i error is enought).

On the hp stuff i have a beva 4400 and 4000 to monitor but i have no idea where to start (snmp or perfmon to the manager or to a controller or yet another thing).

[NOB]

stuff deleted. [NOB]

If you have some monitoring stuff for oracle on hp-ux, specifically rman thats also good. (i am unsure tail -n 10|grep -i error is enought).

On the hp stuff i have a beva 4400 and 4000 to monitor but i have no idea where to start (snmp or perfmon to the manager or to a controller or yet another thing).

Hi orogor

we use the check_oracle_health Nagios plugin to monitor whether the DB
is alive and to check for the usage of the tablespaces.
ZABBIX supports external checks for this purpose.

Be warned: We have 51 Oracle databases and, I guess, more than 100 tablespaces.
Using external checks for all that immediately reach the limit of the number of pollers and the load on
the ZABBIX server.

HTH,

Norbert.

[orogor]

Using a simple tnsping isnt enought to know if the DB is up?
Yes adding a check for the tablespace is a good idea, i think we had twice the issue this year with a full tablespace.

[den_crane]

http://www.zabbix.com/wiki/howto/mon.../orcale/oracle