Unsolved
This post is more than 5 years old
1 Rookie
•
45 Posts
0
2591
January 26th, 2011 07:00
VSS Errors when snapshotting in RM
All,
We have VSS errors in all our SQL LUN's that we snapshot using Replication Manager, the errors are as follows:
26-Jan-2011 14:54:27 FLGN64 000546 ERROR: Microsoft Volume Shadow Copy error: Unexpected provider error. Please verify the load and configuration of the storage array and try again.
26-Jan-2011 14:54:27 FLGN64 026202 ERROR: Storage Services operation waitVSSCompletion failed with an error as follows: Wait Snapshot completion failed.
26-Jan-2011 14:54:27 FLGN64 026084 ERROR: Storage Services operation processExecute failed with an error as follows: Storage Services has terminated with an error.
Is there anything that can be done to increase the 10 second time out for VSS or is there anything that can be done to get around these issues.
Thanks in advance.
ACN960
1 Message
0
May 28th, 2013 21:00
Does anybody knows how to resolve this?
FoolInTheRain
123 Posts
0
May 29th, 2013 06:00
I agree with Adolfo, but if this is a new job/setup, you may want to:
1) make SURE you have all required MS hot fixes applied to the prod host...especially the VSS hot fixes :-)
2) try setting up the RM job option to do "retries". I suggeest 3 retries waiting 60 seconds between retry
3) double check your overall RM Job schedule and space jobs out 15 minutes (if possible)
4) I've seen issues that were related to Solutions Enabler, you may want to consider upgrading to SE 7.5 or 7.5.1
to see if that helps (assuming your version of RM supports those versions of SE!)
5) Just throing this out there, but if the prod host is a VMware host...try increasing the cores/cpu
(ya, I've been saying this alot lately for any performance issue!)
There could be other reason's as well, like network or array performance that could impact the VSS 10 second limit, but i would suggest you rule out as many of the above before trying anything else.
FoolInTheRain
123 Posts
0
May 30th, 2013 08:00
Just tossing in a #6 for my list from above:
6) If you have VSS 10 second issue, and you have a Clariion or VNX, what level of "security" do you have for your Naviseccli? The RM Admin Guide states:
Note: EMC requires you to run the Naviseccli commands to confirm acceptance of a security certificate when in medium security mode. Low security mode does not use security certificates, so it is not necessary to use the commands to setup certificates. use of medium mode can result in fewer LUNs splitting during the 10 seocnd VSS freeze windows used by Windows replications (most apparent on the IA64 platforms).
so, you may want to have the naviseccli security mode set to low.
BigMark1
1 Message
0
October 18th, 2013 03:00
I also have this issue.
This is with Replication Manager 5.4 SP4, Recoverpoint as the Snapshot Mechanism, and Vplex as the storage provider on top of VMAX10k Arrays.
EMC have acknowledged this as a bug in the Recoverpoint code, to be fixed in a future release of recoverpoint
currently developer trying to add a fix to RP v4.0.2 tentative GA 22/11/2013.
Mark