2

Has anyone setup Ganglia for metrics collection on a production system? We are using Ganglia version 3.1.7-3.el6 on RHL 6.3 on Amazon (in unicast mode) with RRD tool 1.4.7-1.el6.

We are facing major stability issues and Gmetad seems to go down quite frequently with following error.

kernel: gmetad[32277]: segfault at 4e47aa0a ip 00000037af00aee4 sp 00007fae22766230 error 4 in libganglia-3.1.7.so.0.0.0[37af000000+14000]

It could be because we are using incompatible or old versions of the above packages.If you are successfully running Ganglia on a RHL production setup without any issues and what versions of the above packages you are using?

snehaso
  • 51
  • 1

1 Answers1

0

I was facing a similar issue.

This resolution might help -> Ganglia - gmetad - process is getting terminated by SIGSEGV

In my case, to pin point a file name - 'part_max_used.rrd' was a file name under /path/to/ganglia/rrds/node_name was the root cause of SIGSEGV. After removing the file gmetad started running fine.

Thanks

Community
  • 1
  • 1
Rishi
  • 5,869
  • 7
  • 34
  • 45