Not suitable with glusterfs version < 3.8

[shaotingcheng]

Here is a memory leak issue of glusterd reported in 2015
https://bugzilla.redhat.com/show_bug.cgi?id=1287517

The memory leak is caused by running "gluster volume status all" thousands of times which is the similar behavior with gluster-exporter does.
This issue has been fixed in glusterfs-3.8

My experience with glusterfs-3.6.9, the exporter will cause the severe memory leak on the server.
(prometheus scrape period : 180s)
The glusterd will crashed from time to time due to out of memory.