Peter,
I have had similar issues running Linux RH where the system was reaching avg load of > 3 also. I did some work on perlcc and was able to compile all polling jobs (perlcc job -o job.executable). The load has dropped to < 1 now.... hth paul *speaking for himself. [chop] Problem: Currently monitoring about 50 hosts and the limited server we use (a 8 year old Sun Netra-i) has an avg load of 3. And is nearly passing its 5 minute round time to read the information from the routers and server. Question: I want to increase the speed of reading out the systems. Now I'm not the world greatest coder, but I think a lot of speed can be gained by NOT opening a seperate socket for every snmpcall. I can do this by using snmpwalk, storing the info in an temporary array and taking some of that information into RRD. Is this a good idea? Does anybody else have any ideas to increase the performance? Thanks, [chop] -- Unsubscribe mailto:[EMAIL PROTECTED] Help mailto:[EMAIL PROTECTED] Archive http://www.ee.ethz.ch/~slist/rrd-users WebAdmin http://www.ee.ethz.ch/~slist/lsg2.cgi
