From: Buchan M. <bg...@st...> - 2008-01-14 17:36:41
|
On Wednesday 19 December 2007 19:25:55 Buchan Milne wrote: > On Wednesday 19 December 2007 17:19:34 xbgmsharp wrote: > > Hi all, > > > > After a day of DEBUG. I works. > > I modify the file dm_snmp.pm which do snmp pooling. > > So now instead of pooling all leaf oid in one request, 1 oid equal 1 > > request. So now than i am pooling each oid separitly devmon should not go > > purple. I try to make it per pakets like 5 or 10 but devmon get lost when > > corresponding data with test done in dm_tests.pm > > > > I will let it working on some PIX and F5 devices and commit by the end of > > the week. > > > > All of this, is for not getting SNMP Error: "error status: tooBig". > > Because of this error all my leaf test was going clear. > > > > Comments? > > I have a number of servers using the compaq-server template, and on some, > all the tests are always green. On the others, the are all clear, and I > have error messages such as this: > > Missing repeater data for primary OID cpqHeFltTolFanIndex > > There is no apparent difference between the servers. One DL380 works, one > does not. Two DL580s work, three don't. > > I'm wondering if there is a similar issue here. If I snmpwalk the whole > Compaq OID, or each branch, I get the data I expect on the servers that > aren't working. I have resolved most of the issues here. In some cases it may have been that the Insight agent for the device type was not running (I have two servers that currently aren't showing any info on their raid controllers). In the other case, the IML log was empty, which results in a "No Such Object available on this agent at this OID" for the parent of all the branches in the log test. When the IML log is emptied with the "hpasmcli -s 'clear iml'" command, the IML is empty, and this error occurs. When emptying the log from the iLO (integrated lights out) card's web interface, it puts one entry in the log, as follows: Informational Unknown 2008-1-14 13:14 2008-1-14 13:14 1 IML Cleared (iLO user:admin) So, I have updated the text on the log page to indicate that the IML should be cleared from the web interface. Note that if the IML log is empty and the log test is enabled on the server, most likely most of the other tests will be clear (in the case where I upped the limit for failed branches in the code, around line 560 in dm_snmp) or purple. I'm undoing the change on the "Failed too many queries" limit I had done. However, it would have been nicer if we could just skip the test in question (rather than the whole device). Regards, Buchan |