We are having an issue like this. When we do 'iostat -x' on a solaris box, we see a particular metadevice is showing High Read cycle and %b column is more than 90% always.
root@maty1 # iostat -x 2
extended device statistics
device r/s w/s kr/s kw/s wait actv svc_t %w %b
ssd19 555.0 8.5 4667.8 68.0 0.0 2.5 4.4 0 97.
This particular metadevice 'ssd19' is a part of metaset of a shared disk storage between a two node cluster. Since it is a part of metaset, there is no specific mount point available for this metadevice so that we can identify the process which is causing the high read cycles. So we tried to find out the actual disks which constitute the creation of this metadevice 'ssd19' and finally we found those.
Now my question is.....I have 4 disks which are part of a metaset, having this information how can I find the processes which are accessing it and which are causing high read cycles?
(Please note that I have no mount point details).
Please get me a solution asap.
Thanks in advance !