Unsolved
This post is more than 5 years old
27 Posts
0
2121
Presenting CLARiiON CX500 and CX4-240 to same ESXi host
Hi,
Just wondering if anyone had any issues with Presenting CLARiiON CX500 and CX4-240 to same ESXi host and mixing Failover modes 1 & 4?
I'm trying to present 2 different arrays to the same ESXi host to migrate data from old array to new...
I get some failed to initialize vmfs3 distributed locking errors which result in storage becomming unresponsive.
CX500 is presented to 3 ESXi hosts with independent storage groups configured
CX4-240 is presented to3 ESXi hosts in the same storage group (intend to move to cluster based storage group for easier management)
ESXi hosts are configured on the CX500 with Failover mode 1 and Failover mode 4 for CX4-240.
Host LUN ID's start at 0 for CX500 and 100 for CX4-240.
Can anybody spot an issue here?
I've checked the Switch Zoning and everything is the same ESXi HBA is seperately zoned to each SP for both arrays.
Don't understand it, if anyone has any suggestions I'm keen to hear. Look forward to your response.
Craig
Stephanie_P
180 Posts
0
April 12th, 2011 07:00
Hi Craig, It sounds like you are in a production environment. We'd like to move this thread over to the Support Forums Community where your question can be addressed.
Thanks,
Stephanie
dynamox
1 Rookie
1 Rookie
•
20.4K Posts
0
April 12th, 2011 14:00
if your CX500 running on flare 26 ..then you could use failover mode 4 on both arrays.
dynamox
1 Rookie
1 Rookie
•
20.4K Posts
0
April 12th, 2011 15:00
not sure if you have access to this document
Storage Target Revisions and Adoption Rates
by here is a snapshot
cswaters1
27 Posts
0
April 12th, 2011 15:00
I thought that anything pre CX4 didn't support Failover mode 4?
I'm of the understanding that anything prior to CX4 (CX3 and CX500) only supports Failover mode 1.
Can someone else confirm this?
In the meantime, thanks for your response dynamox I'll see if I can test this.
thanks again for your response.
Craig.
cswaters1
27 Posts
0
April 12th, 2011 15:00
CX500 - 02.26.500.5.031
CX4-240 - 04.30.000.5.509
ESXi - 4.1
There are a lot of the following errors on an ESXi hosts when I try to present a lun from CX4-240: -
Reservation error: Timeout
Failed to initilise VMFS3 distributed locking
Failed to reserve volume ...
Any suggestions?
cswaters1
27 Posts
0
April 12th, 2011 16:00
Just looked at Release notes for FLARE 26 for CX500 and it does support ALUA.
Can you give me more details on what you have suggested on your previous post?
Also, can I just move all the ESXi hosts to Failover mode 4 or is an outage required?
Thanks for your response.
dynamox
1 Rookie
1 Rookie
•
20.4K Posts
0
April 12th, 2011 16:00
can you start with host id 0 on both arrays ...there are different targets as far as ESX is concerned.
dynamox
1 Rookie
1 Rookie
•
20.4K Posts
0
April 12th, 2011 18:00
you mentioned that on CX4 you started from host id 100 ? Any reason why ? I was suggesting to start from host id 0 on each array. I have not moved ESX from failover mode 1 to 4, on windows i simply changed the mode and rebooted the system.
dynamox
1 Rookie
1 Rookie
•
20.4K Posts
0
April 12th, 2011 19:00
just for test, present a LUN from CX4 with host id 0 ..see if it generates the same error message.
cswaters1
27 Posts
0
April 12th, 2011 19:00
Appologies for the delayed response, just has a meeting.
we started at 100 as there are luns on the CX500 starting at 0, so I started all the LUNs on the CX4-240 at 100.
cswaters1
27 Posts
0
April 13th, 2011 21:00
I think I've fix it, it looked as though one fo the ESXi servers had a lock on one of the VMFS volumes on the CX4-240 array, once all the ESXi hosts were restarted the lock was removed and the VMFS volume was mountable and visible.
EMC engineering are checking over the dump files, so I should have some concert info soon, I'll post what I find. Thanks for your help.