No data from HP Insight agent

[kostya2]

I have a number of servers that report no data from HP Insight agent. Output from snmpget is as follows (ran from a proxy):
snmpget -v 2c -c public 10.96.61.139 .1.3.6.1.4.1.232.2.1.3.0
SNMPv2-SMI::enterprises.232.2.1.3.0 = INTEGER: 2

Graphs for the metric show that the data was collected up until a few days ago. After that I have a few short periods when the data is received surrounded by long periods of no data. Last period with data was yesterday and lasted for about 30 min.
Where do I start troubleshooting this issue? I recently assumed the role of a zabbix admin after the previous admin left the company. Any help would be much appreciated.

[richlv]

which version of zabbix ?
are those hosts monitored by zabbix proxy ?
is the zabbix proxy running ?
what does proxy logfile say ?
do other hosts and items, monitored by the same proxy, return data ?

[kostya2]

zabbix version 1.8.2
All these hosts that are missing this data are monitored via same proxy. Other hosts monitored by this proxy are OK. Proxy processes are running.
Here is a snippet from the zabbix_proxy.log (filetered for one of the hosts with missing data):

27737:20110220:141931.808 Item [ny3wdev010.arbitrade.com:HPenclosureRollupStatus1. 2] error: SNMP error [(noSuchName) There is no such variable name in this MIB.]
27737:20110220:141931.808 Parameter [ny3wdev010.arbitrade.com:HPenclosureRollupStatus1. 2] is not supported by agent Old status [0]
27745:20110220:141948.515 Item [ny3wdev010.arbitrade.comerf_counter[\Physical Disk(_Total)\Avg. Disk Read Queue Length]] error: Not supported by Zabbix Agent
27745:20110220:141948.515 Parameter [ny3wdev010.arbitrade.comerf_counter[\Physical Disk(_Total)\Avg. Disk Read Queue Length]] is not supported by agent Old status [0]
27774:20110220:141951.229 Sending list of active checks to [10.96.61.139] failed: host [ny3wdev010] not found
27744:20110220:142039.756 Item [ny3wdev010.arbitrade.com:HPsystemStateStatusMessag e] error: SNMP error [(noSuchName) There is no such variable name in this MIB.]
27744:20110220:142039.756 Parameter [ny3wdev010.arbitrade.com:HPsystemStateStatusMessag e] is not supported by agent Old status [0]
27774:20110220:142151.229 Sending list of active checks to [10.96.61.139] failed: host [ny3wdev010] not found
27773:20110220:142351.230 Sending list of active checks to [10.96.61.139] failed: host [ny3wdev010] not found

[richlv]

first, it looks like you have lots of minor misconfiguration issues - active agents connecting in but having no items, incorrect oids, incorrect perfcounters... these all add up.

second, you are using zabbix 1.8.2 - version, which doesn't have unreachable pollers. thus it is possible that proxy is overloaded by having too many incorrect things to check. try to fix all the errors reported by proxy, and once you have done that, consider increasing amount of pollers (even if only temporary to see whether that helps)