i'm trying to setup simple lab consisting of NNM+NA+NNM perfSPI. I'm using 3 servers (one for each application), Windows 2008. NNM and NA are working fine. But i have problems with running perf SPI. Installation went fine. Configuration also went fine (nnmenableperfspi.ovpl), but as soon as it starts running it starts filling up prspi.log with same error. Log is filling up very faste, like 1GB in half an hour.
Error is saying something like this. Bigger example is in attachment.
DB Transaction failed due to insufficient buffers. Retrying. Memory available: 2.64319229125977 GB. DB error: Could not execute statement. Insufficient buffers for 'QCPRun'. -- (db_sqlins.cxx 17044) SQLCODE=-1009134
In NNM i even disabled all performanse pulling temporarely, but that didn't help at all.
-What version of the PerfSPI are you running 9.20? a) how much memory does your machine have? b) what are the settings for -iqmc and -iqtc (in DATADIR/database/perfspi.cfg) c) was it being asked to process a lot of data.. ie was there a big backlog? e) are many other extensionPacks also running
This usually happens on very underpowered machines.. especially if we're asking it to load a lot of data at once.
HP Support If you find that this or any post resolves your issue, please be sure to mark it as an accepted solution.
Hi, It has only 6gb of RAM. However, in meantime i managed to sort the issue. You were on the right track, becuase of low RAM iqmc and iqtc values were i guess too low (around 1400). I changed them to 2500 and it seems to fix the issue. There was no data to process, because performance data monitoring was disabled, and that was just after fresh install of perf spi.