Ranked Statistics: 12334963.32 percent
12334963.32 percent is a nice number if you talk about increase in wages, but not if you talk about mortgage rates or CPU load. Something is wrong with our "Ranked Statistics". Top 20 CPU Routers show 6 boxes with impossible values, but if i click on the "Target"-link, the graph shows normal values (but not high enough to put it into Top 20). The rest of the boxes show normal values. The other views shows just as erroneous behaviour. Is this a result of us using RRDtool version 1.2.23? --Ingeborg -- Ingeborg ?strem Hellemo -- ingeborg@cc.uit.no (Univ. of Troms?, Norway)
Ingeborg Hellemo <Ingeborg.Hellemo@cc.uit.no> writes:
12334963.32 percent is a nice number if you talk about increase in wages, but not if you talk about mortgage rates or CPU load.
Something is wrong with our "Ranked Statistics". Top 20 CPU Routers show 6 boxes with impossible values, but if i click on the "Target"-link, the graph shows normal values (but not high enough to put it into Top 20). The rest of the boxes show normal values.
The other views shows just as erroneous behaviour.
Is this a result of us using RRDtool version 1.2.23?
It probably is. As soon as we upgraded to Debian Etch, we noticed the same problems on our installations. Debian Etch features rrdtool 1.2.15. John-Magne has investigated the issue, and found a problem with the accompanying python-rrd 1.2.15. It seems that when one asks to fetch all the data points within a time period, an extra record with nonsensical numbers is added to the result. This didn't happen on Debian Sarge, which I think featured rrdtool 1.0.49. AFAIK, John-Magne is working on a workaround, and also on querying the rrdtool community about the problem. I just added it to the NAV bugtracker so can monitor it if you like: http://sourceforge.net/tracker/index.php?func=detail&aid=1822583&group_id=10... -- mvh Morten Brekkevold UNINETT
From Ingeborg.Hellemo at cc.uit.no Tue Oct 30 10:01:25 2007 From: Ingeborg.Hellemo at cc.uit.no (Ingeborg Hellemo) Date: Tue Oct 30 10:01:27 2007 Subject: [Nav-users] smsd is throwing exceptions at me In-Reply-To: Your message of "Fri, 26 Oct 2007 13:17:13 +0200." <200710261117.l9QBHDQ6017343@barnetv.cc.uit.no> Message-ID: <200710300901.l9U91Pvl076327@barnetv.cc.uit.no>
Ingeborg.Hellemo@cc.uit.no said:
How can I turn on more debugging?
I have put loglevel: DEBUG into etc/smsd.conf, but this only makes a difference in logging from /usr/local/nav/bin/smsd.py. I still don't see any log messages from dispatcher.py --Ingeborg -- Ingeborg ?strem Hellemo -- ingeborg@cc.uit.no (Univ. of Troms?, Norway)
participants (2)
-
Ingeborg.Hellemo@cc.uit.no -
morten.brekkevold@uninett.no