Re: client failing to respond to cache pressure

[Zhenshi Zhou <deaderzzs@xxxxxxxxx>]

Hi

The cluster recovers healthy after I set "mds_cache_memory_limit" from 4G to 8G.

Zhenshi Zhou <deaderzzs@xxxxxxxxx> 于2018年11月30日周五 上午11:04写道：

Hi

I used to get warning message claims client faling to respond 

to cache pressure. After I switch rockdb and wal data to ssd, the 

message seems disappeared.

However it shows again yesterday and the message looks a little 

different: MDS_CLIENT_RECALL_MANY: 1 MDSs have many 

clients failing to respond to cache pressure.

I added some fuse client and it has 48 clients mounting the cephfs 

at present. The cluster has only one MDS active and one standby. 

How can I get the cluster healthy again? I'm happy to provide any 

information of the cluster.

Thanks

_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com