[Ganglia-general] webserver load

2009-11-19 Thread Markus Köberl
Hi! We use ganglia to monitor all our clients and servers. Everything works fine, but if a lot of people are watching the webinterface, the webserver gets overloaded while generating the pictures. Is there a simple way to cache the pictures till the rrd data changes? Regards, Markus Köberl --

Re: [Ganglia-general] webserver load

2009-11-19 Thread Daniel Pocock
Is there a simple way to cache the pictures till the rrd data changes? You could use a reverse proxy - just make sure the expiry header is set correctly on the graphs coming from Ganglia -- Let Crystal Reports

[Ganglia-general] segfault killing gmetad

2009-11-19 Thread Hardik Shah
Hi, I have gmetad/gmond running on ganglia farm master host. I have made Linux virtual machine as master host and it is serving monitoring for around 30-40 physical servers. But I do see once in day segfault is occurring and is killing gmetad on the master which is stopping ganglia farm

Re: [Ganglia-general] segfault killing gmetad

2009-11-19 Thread chifeng
Can you provides this error logs when gmetad stop? default path: /var/log/messages Chifeng On Fri, Nov 20, 2009 at 12:33 PM, Hardik Shah hardik.s...@synopsys.comwrote: Hi, I have gmetad/gmond running on ganglia farm master host. I have made Linux virtual machine as master host and it is

Re: [Ganglia-general] segfault killing gmetad

2009-11-19 Thread Hardik Shah
Thanks for your reply. Below are the logs when gmetad stops: Nov 19 20:29:59 ganglia-aix-vm xinetd[2038]: START: shell pid=18292 from=146.225.141.186 Nov 19 20:30:00 ganglia-aix-vm rshd[18293]: r...@ecsadmin as root: cmd='/etc/init.d/gmond restart' Nov 19 20:30:01 ganglia-aix-vm xinetd[2038]: