VMware Cloud Community
aevenot
Contributor
Contributor

MSA 1500cs firmware v7 with ESX 3.0.2

Hi,

I'm currently looking for upgrading my HP MSA1500 SAN from v5 to v7 and switch my SAN to active/active configuration. For that, I've searched in several KB from both HP and VMware website and I have some interrogations.

The HP Compatibility

http://bizsupport.austin.hp.com/bc/docs/support/SupportManual/c00868010/c00868010.pdf

Only reference to ESX 3.0.1 support

The VMware San guide

Hewlett Packard MSA‐1000 V4 Active‐passive MRU - Most Recently Used MSA1000_VOLUME

Hewlett Packard MSA‐1500 V5 Active‐passive MRU - Most Recently Used MSA1000_VOLUME

Hewlett Packard MSA‐1500 V6 Active‐active Fixed MSA VOLUME

No reference to firmware v7

Does anybody have already done that?

Is it supported by HP and VMware?

Reply
0 Kudos
6 Replies
aldin
Enthusiast
Enthusiast

we have MSA 1500cs firmware v7 with ESX 3.5 and it works fine.

We have firmware v5 and we upgade it to v7 and it goes smoothly, by book.

aevenot
Contributor
Contributor

Great but do you have an experience in ESX 3.0.2 version? The ESX upgrade is no longer planned.

Other question : Do you have lost array configuration in your firmware upgrade ?

Reply
0 Kudos
aldin
Enthusiast
Enthusiast

no, we have not lost array configuration in upgrade process, as i already said it goes smoothly.

i have to say that we also was a little bit scared about that upgrade, we could not find in compatibility guides is it supported configuration (v7 and esx 3.5), and it was small installation with only 4 VM so we take a risk.

Reply
0 Kudos
jhanekom
Virtuoso
Virtuoso

We recently did an upgrade to v7 on an MSA 1000. ESX detects it as an active-active controller system afterwards. We set up the same preferred paths on all our hosts and configured manual LUN balancing on the MSA that matched that configuration.

We did encounter one glitch and still cannot explain why it occurred. For some reason, all our LUNs were detected as snapshots and we had to resignature them. This was odd, since the LUN ordering/numbering didn't change (we checked.) The result was that we had to fix the registrations for all our VMs and edit a few VMX files to point to the correct data stores.

I think what I'm trying to say is just: budget some extra time for unforeseen problems.

Reply
0 Kudos
Moynul_Ali
Contributor
Contributor

Hi jhanekom,

Please could you tell me how you got this working?

I have a custome who has got 4 x ESX Server 3.0.2 connected via 2 x HP Switches to an MSA1500cs with Dual Controllers. The FW are 7.00 and are active/active.

They have set the redundancy mode to manual and split the luns across both controllers on the MSA. All the ESX HBA's have been told to use MRU as the failover policy.

However, they recently had a failure of the controllers and the LUNs on the MSA did not failover automatically. Do you know why this could be? Everything has been configured correctly as in SSP, LUNs, Zoning, we just could not get the LUNS to failover to the other controller in an active/active sitatuion.

Any help is appreciated.

Please email me on

Thanks

Reply
0 Kudos
daniel-san
Enthusiast
Enthusiast

Hi.

I´d the same problem.. I hade to resignature all disks after the v5 -> v7 upgrade.. I´m not really sure why?

Regards

Daniel

Reply
0 Kudos