[Gluster-users] glusterfs using large amount of ram

[Jim Kusznir]

Hi:

My monitoring system notified me that one of my glusterfs nodes is low on
swap. When I logged in, a glusterfs process for one of my volumes was
using 66% of the systems' ram, top reported 28.4g virt, 20.8g RES, and 3704
SHR.

As far as I can tell, this should not be considered "normal". What does
this mean / how do I fix it?

Thanks!
--Jim

[Niels Hendriks]

Hi Jim,

I have a similar issue I reported here:
http://lists.gluster.org/pipermail/gluster-users/2018-June/034169.html
However, I have not yet received a response. It's probably one of the
settings since I've seen this on 3 separate gluster environments but I'm
not sure what's causing it...

Niels

Post by Jim Kusznir
My monitoring system notified me that one of my glusterfs nodes is low on
swap. When I logged in, a glusterfs process for one of my volumes was
using 66% of the systems' ram, top reported 28.4g virt, 20.8g RES, and 3704
SHR.
As far as I can tell, this should not be considered "normal". What does
this mean / how do I fix it?
Thanks!
--Jim
_______________________________________________
Gluster-users mailing list
http://lists.gluster.org/mailman/listinfo/gluster-users

[Nithya Balachandran]

Hi Jim,

Which process is using up the memory? Please take a statedump of it at
intervals and send those across. See [1] for details on how to take and
read a statedump.

Regards,
Nithya

[1] https://docs.gluster.org/en/v3/Troubleshooting/statedump/

Post by Jim Kusznir
My monitoring system notified me that one of my glusterfs nodes is low on
swap. When I logged in, a glusterfs process for one of my volumes was
using 66% of the systems' ram, top reported 28.4g virt, 20.8g RES, and 3704
SHR.
As far as I can tell, this should not be considered "normal". What does
this mean / how do I fix it?
Thanks!
--Jim
_______________________________________________
Gluster-users mailing list
http://lists.gluster.org/mailman/listinfo/gluster-users