Yes I got that one.
The problem is the new release of Tavve is doing something rather intensive
with interface descriptions. (I think they are substituting in the local
ifAlias on top of the normal ifDesc). Anyway, they have a fix NSN220011 that
will revert the way Tavve works back to the way the previous version worked.
-----Original Message-----
From: owner-nv-l@lists.us.ibm.com [mailto:owner-nv-l@lists.us.ibm.com]On
Behalf Of Paul
Sent: Tuesday, January 20, 2004 3:27 PM
To: nv-l@lists.us.ibm.com
Subject: Re: [nv-l] High CPU Usage in ovwdb
Scott,
Have you installed IY50663, I just wanted to be sure.
Paul
Barr, Scott wrote:
> (this note is being sent to tavve support + netview mailing list - I
> appeciate any input)
>
> Solaris 2.8
> Tavve ENMS 2.2
> NetView v7.1.4
>
> Greetings, I am seeing high cpu utilziation (i.e. above 50% of a CPU)
> for ovwdb. I can find no netview cause. Can you speculate on the
> possibility of Tavve causing this situation? The NetView FFDC (first
> failure data capture) feature has been disabled.
>
> Performance of the web interface for Preview and NetView is horribly
> slow. If I cycle ovwdb, the CPU comes down for a while but eventually
> ramps back up. Any suggestions?
>
> last pid: 21454; load averages: 1.20, 1.27,
> 1.31
> 13:55:45
> 113 processes: 110 sleeping, 1 running, 2 on cpu
> CPU states: 49.1% idle, 38.5% user, 12.2% kernel, 0.2% iowait, 0.0% swap
> Memory: 2048M real, 427M free, 1293M swap in use, 4212M swap free
>
> PID USERNAME THR PRI NICE SIZE RES STATE TIME CPU COMMAND
> 479 root 1 0 0 19M 15M cpu/0 34.9H *49.01% ovwdb
> * 569 root 1 40 0 88M 76M sleep 992:49 0.92% snmpCollect
> 2230 root 1 48 0 677M 674M sleep 33.4H 0.38% tdatad
> 1345 root 1 58 0 29M 17M sleep 132:38 0.12% netmon
> 498 root 1 58 0 47M 36M sleep 11:49 0.08% nvcold
> 21275 root 1 58 0 2688K 1696K cpu/1 0:00 0.07% top
> 21453 root 1 58 0 11M 4672K sleep 0:00 0.07% getIfDescr
> 29515 root 11 59 0 85M 41M sleep 230:54 0.06% java
> 21454 root 1 58 0 11M 4648K sleep 0:00 0.05% getIfDescr
> 10902 root 1 44 0 1920K 1312K sleep 0:02 0.03% ksh
> 823 root 23 59 0 46M 15M sleep 7:35 0.02% java
> 7945 root 12 58 0 8424K 7104K sleep 3:05 0.01% mibiisa
> 15710 root 1 58 0 23M 16M run 7:19 0.00% nvcorrd
> 500 root 14 59 0 50M 15M sleep 0:06 0.00% java
> 61 root 7 32 0 3000K 2144K sleep 31:01 0.00% picld
>
>
>
>
> *Scott Barr*
>
> *Distributed Network Engineer*
>
> *CSG Systems Inc.*
>
> *scott_barr@csgsystems.com*
>
> *Voice: 402-431-7939*
>
> *Fax: 402-431-7413*
>
> * *
>
>
|