Description of problem (please be detailed as possible and provide log
snippests):
Cluster wide performance is slow where customer is unable to start all of the applications (using both cepfs and rbd pvs) without encountering timeouts.  We are seeing some MDS slow requests reported over past few days but no slow requests reported on the OSDs. We performed fio tests on cephfs mount to worker node and storage node and observed slow writes using both 4k and 4 block sizes.  Performing the same test on storage node using local disk using same back-end storage (different luns) on Dell EMC powerflex storage performed much better.

Version of all relevant components (if applicable):
ODF 4.10.12 / rhcs 16.2.7-126

Does this issue impact your ability to continue to work with the product
(please explain in detail what is the user impact)?
Yes, the customer is unable to bring up all of the applications

Is there any workaround available to the best of your knowledge?
None at the moment

Rate from 1 - 5 the complexity of the scenario you performed that caused this
bug (1 - very simple, 5 - very complex)?
5

Can this issue reproducible?
Yes, at the customer site.


Additional info:
See comment #1 for fio results