Attachment uploads are currently disabled. This is a temporary situation and will resume as normal in the coming days.
Solved!
Go to SolutionSRDF/A
Hi All,
We have 10 rdfg groups(rdfg/server) in a vmax for srdf/a.They are replicating from nj to ny now on one exchange server admins wants to run jetstress test(that means lot of io's) will it impact only one rdfg group to go down or all groups.
Thanks in advance..
RobertDudley
2 Intern
•
448 Posts
472
0
Posted January 21st, 2010 08:00
It may not impact only that one group as there is a "drop priority" in SRDF/A. Check Primus EMC195411.
RDF group attributes allow you to assign priorities to SRDF/A sessions, and to set the minimum amount of time before attempting an SRDF/A cycle switch.
To set RDF group attributes, use the following command:
set [ra | rdf] group GroupNum [session_priority = SessionPriority] | [minimum_cycle_time = CycleTime];
where:
GroupNum = the RDF (RA) group number.
session_priority = the priority used to determine which SRDF/A sessions to drop if cache becomes full. Values range from 1 to 64, with 1 being the highest priority (last to be dropped).
minimum_cycle_time = the minimum time to wait before attempting an SRDF/A cycle switch. Values range from 5 to 59 seconds.
RobertDudley
2 Intern
•
448 Posts
472
0
Posted January 21st, 2010 08:00
If you drop it out of SRDF/A mode then it will no longer be consistent. Acp_disk mode allows the array to accept data and destage to local disk prior to sending over the link. When bandwidth is available it will then re-read from disk to cache to send over the link.
You could run the command to set the drop priority for that one SRDF/A group to be the first group dropped during the testing period.
You could also just take that one group out of SRDF/A mode for the testing, go into acp_disk, then reactivate SRDF/A let it catch up, enable consistency.