[ClusterLabs] Could we use pacemaker configure two master/slave resources?

[范国腾]

Hello,



We use the following cluster now. the postgres resource agent PAF has its own mechanism to decide which node should be promoted to master.



But the performance test show the disk IO is the bottleneck. So we want to try to change the GFS2 resource（clusterf） to be master/slave mode too which could increase the performance.



So there will be two resources（pgsql and gfs）to be master/slave module and we hope their master are always in the same node.  If switchover happens and we hope the pgsql and gfs switchover at the same time.



I am studying how to implement this configuration. Could you please give any suggestion?



[cid:image001.jpg at 01D3D19F.C815E640]



By the way, Now we use the following method to create the clone GFS resource. Is there any way to configure to improve the share disk access performance?
pcs resource create dlm ocf:pacemaker:controld op monitor interval=30s on-fail=fence clone interleave=true ordered=true;
pcs resource create clvmd ocf:heartbeat:clvm op monitor interval=30s on-fail=fence clone interleave=true ordered=true;

pcs resource create clusterfs Filesystem device="/dev/sdb1" directory="/home/postgres/sharedisk" fstype="gfs2" "options=noatime" op monitor interval=10s on-fail=fence clone interleave=true;
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/users/attachments/20180411/f3e8ac40/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 31288 bytes
Desc: image001.jpg
URL: <https://lists.clusterlabs.org/pipermail/users/attachments/20180411/f3e8ac40/attachment-0001.jpg>