This post is more than 5 years old
1 Rookie
•
73 Posts
0
1169
January 17th, 2013 01:00
SRDF Data Corruption at DR site if Production Site has a POD failure
Hi all
Could someone provide some input into the scenario described in the subject.
If I have SRDF replicating data synchronously from my Production site to the DR Site and suddendly I lose a POD in the production site affecting some of my Raid Groups due to a PSU failure.
Does the SRDF link "split" automatically and will the SPS allow the data to be sent across to the DR site.
Is it possible you could get data corruption, I/O errors or inconsistent data at the DR site.
Many Thanks
No Events found!
sauravrohilla
859 Posts
0
January 17th, 2013 02:00
Hi Steve,
I believe when Symm is on stand by power, it only focus on vaulting than on transmitting data across the symm and I also believe that in any case,SRDF /S or SRDF/A your data on DR will not be corrupted.
Zikas is correct when a symm goes down, the SRDF link goes in to partitioned state.
regards,
Saurabh Rohilla
Zikas
278 Posts
1
January 17th, 2013 02:00
Hi Steve,
at least from my experience in SRDF/S when you have a POD the SRDF link will not be split automatically.
The link is going to be partitioned only if you have the SRDF link down unexpectedly and you have to resume it.
And yes it is possible if you data corupted on the Production Site to be replicated on the DR Site.
Steve0_2012
1 Rookie
•
73 Posts
0
January 17th, 2013 03:00
Thanks for your replies guys.
Seems like you disagree though that SRDF will/will not replicate corrupted data.
One last thing; could you describe what unrebuild is. I have seen the term and a little unsure.
I believe this is related to invalid tracks between R1 and R2 devices but would appreciate additional input.