RE: [cobalt-users] parseReport.pl and huge web.cache

["Steven Depuydt - BeNe.WS" <steven@xxxxxxx>]

I had the same problem, here is an UNSUPORTED solution:

The cache file is generated by analog when invoked from
/usr/local/sbin/split_logs which is invoked from
/etc/logrotate.d/apache.

You do want split_logs to sort through /var/log/httpd/access and create
all the per-site logs as it does now, but you don't want it to invoke
analog.

There are a few ways of resolving this problem.

*** The following suggestions are unsupported and are issued as free,
friendly, useful help ***

You could disable analog, the application and create a dummy in its
place:

  mv /usr/local/sbin/analog /usr/local/sbin/analog.disabled
  echo "#!/bin/sh" > /usr/local/sbin/analog
  echo "exit 0" >> /usr/local/sbin/analog
  chmod +x /usr/local/sbin/analog

Now be aware that ALL built-in site and server Cobalt statistics will no
longer generate.  The changes listed above appear to work on one of my
development RaQs.

To undo this change, should it cause a problem a few months down the
road:

  mv /usr/local/sbin/analog.disabled /usr/local/sbin/analog

...Remember that any OS Update package files that update analog will
overwrite /usr/local/sbin/analog so you'd have to reapply this hack.



-----Oorspronkelijk bericht-----
Van: cobalt-users-admin@xxxxxxxxxxxxxxx
[mailto:cobalt-users-admin@xxxxxxxxxxxxxxx]Namens Anthony Cook
Verzonden: donderdag 17 januari 2002 11:18
Aan: cobalt-users@xxxxxxxxxxxxxxx
Onderwerp: [cobalt-users] parseReport.pl and huge web.cache


Hi,

We've got a RaQ4r with 512meg of memory. Every night, run from the
cron.daily jobs, logrotate is causing problems with our server. We're
getting daily messages from the RaQ software that the server is low on
memory at about 5am. After further investigation, it appears that a script
called parseReport.pl is using most of the memory on the server. It's also
causing the load on the server to increase, and because of the high load has
caused sendmail to shut down temporarily. Here is output from top showing
this:

5:25am up 63 days, 17:12, 0 users, load average: 1.66, 2.35, 2.16 95
processes: 92 sleeping, 3 running, 0 zombie, 0 stopped CPU states: 0.0%
user, 0.6% system, 0.7% nice, 0.2% idle Mem: 517188K av, 515072K used, 2116K
free, 84192K shrd, 10484K buff Swap: 131448K av, 122500K used, 8948K free
11708K cached

PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND 23206 root 11
0 453M 406M 556 R 0 54.8 80.4 11:52 parseReport.

This shows that parseReport is using 80.4% of memory on the server. This
occurs for at least 30 minutes and seems to be the cause of the problem.

This script is Cobalt software, located:
/usr/admserv/cgi-bin/.cobalt/webUsage/parseReport.pl It appears to parse
analog stats to print some reports for the web gui.

I have seen other people post asking for help with this. Does anyone know
what is wrong with this Cobalt script? Does anyone know how to fix it?

It is my guess that this has something to do with the huge logfiles that the
Cobalt is creating (called web.cache and web.cache.new). These are both
around 250meg each now on my server, so I wonder if I can delete these to
solve the problem?

Does anyone know how to disable the stats that the Cobalt does for you? They
are pretty useless, and seem to be causing huge problems.

Thanks in advance for any help.

Anthony Cook



_________________________________________________________________
Join the world?s largest e-mail service with MSN Hotmail.
http://www.hotmail.com

_______________________________________________
cobalt-users mailing list
cobalt-users@xxxxxxxxxxxxxxx
To Subscribe or Unsubscribe, please go to:
http://list.cobalt.com/mailman/listinfo/cobalt-users