Hi John,
I have snip out the ipaddress, I'm using cephfs clients. And as of now using 7 clients, it will increase. The kernel version i'm using is 3.10.0-514.16.1.el7.x86_64 in all ceph nodes(mds/osd/mon/clients). And the os version i'm using is Centos 7.3 Regards, Suresh ---- On Mon, 14 Aug 2017 19:54:25 +0530 John Spray <jsp...@redhat.com> wrote ---- On Mon, Aug 14, 2017 at 3:13 PM, psuresh <psur...@zohocorp.com> wrote: > Hi, > > Suddenly I've faced cache pressure on my production ceph cluster. mds > running on 8GB vm. > > After restarted the mds service, HEALTH come to OK. What is the issue? > > root@ceph-admin~#ceph -s > cluster 2074b31b-7965-4244-8390-a64f3b038f3e > health HEALTH_WARN > mds0: Client 192.168. failing to respond to cache pressure > mds0: Client 192.168. failing to respond to cache pressure > mds0: Client 192.168. failing to respond to cache pressure > mds0: Client 192.168. failing to respond to cache pressure > mds0: Client 192.168. failing to respond to cache pressure > mds0: Client 192.168. failing to respond to cache pressure > mds0: Client 192.168. failing to respond to cache pressure This is typically associated with use of older kernel clients, or it could indicate a bug if you're using a more recent client. What client is in use? By the way, are those the actual messages or did you snip out the second half of the IP address for privacy? If that's how they really appeared it's probably a bug. John > > > Regards, > Suresh > > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com >
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
