2 hyperv nodes and 2 starwind nodes HA VM Migration problem

[Trololoshka]

Our company deploy failover cluster based on Starwind SAN and Hyper-V according this manual (Starwind v6, Hyperv-v 2012)
http://www.starwindsoftware.com/styles- ... -V_6.0.pdf

When we turn off one hyperv node, VM on HA storage entering in save state, then migrate to alive node (Quick migration), after trying to start and getting error

Code: Select all

Cluster resource 'Virtual Machine Test' of type 'Virtual Machine' in clustered role 'Windows Test' failed. The error code was '0x80048007'.

if VM placed on simple starwind storage image (not HA) there is no problem with migration.

Any suggestion?

[Trololoshka]

After lot of test we concluded.. Starwind HA not usable for failover cluster system. After one of two nodes is fail with VM on HA storage, migration failed with '0x80048007' or '0x80048005'. Seeking another software solution...

[lohelle]

I can say with many years of Starwind experience, that this should work. We are using Starwind for both VMware and Hyper-V clusters (HA).

[Trololoshka]

I can say with many years of Starwind experience, that this should work. We are using Starwind for both VMware and Hyper-V clusters (HA).

Glad to here but in our environment, accordingly by manual setup this doesn't work. Planning to buy 3tb HA licence, but seems not Starwind solution

[anton (staff)]

This is absolutely primary production scenario for us and what you have is misconfiguration. I've brought this thread to attention of our engineers and they will be ready to move on with you including remote session (if required).

I can say with many years of Starwind experience, that this should work. We are using Starwind for both VMware and Hyper-V clusters (HA).

Glad to here but in our environment, accordingly by manual setup this doesn't work. Planning to buy 3tb HA licence, but seems not Starwind solution

[thefinkster]

I echo that with both VSphere (5.5U1/U2) and Hyper-V 2008 R2 or 2012 R2; StarWind v8 (not v6) works for HA and failover.

After lot of test we concluded.. Starwind HA not usable for failover cluster system. After one of two nodes is fail with VM on HA storage, migration failed with '0x80048007' or '0x80048005'. Seeking another software solution...

Can we assume you verified that the active and passive paths for MPIO are set and are pointed to the proper IPs (localhost first) with the alternative node setup as well? And that you've verified this works by turning off StarWind on the 'local' system and making sure your CSV/Witness stays online to the other box via iSCSI before doing anything else with it?

I have had zero issues with just turning off one (just don't ever turn off both storage systems!) of my StarWind nodes in any HA setup. Hyper-V and VMWare just keep on chugging; and on 10gbit, I don't even notice a speed difference. I actually have to verify specifically that it's using the other system's target/image since everything appears seamless.

You should try version 8. Or did you just mean the manual states v6 but you're actually on v8?

[Anatoly (staff)]

@Trololoshka, support is here! As Anton said - we will be glad to assist! Please drop quick email with installation assistance request to suport@ and ut reference to this topic and one of our engineers will schedule the remote session for you.

[Trololoshka]

Thank u very much Now we update to starwind v8 and playing around and have first problems..
Configuration
512mb FLAT, 32mb WT L1, L2 NaN, 2 MPIO links to Hyperv, 10G link
3tb LSFS Thin, no dedup, 3gb WB L1, 440gb WT L2 SSD, 4 MPIO links to Hyperv, 10G link

data transmission stuck very often in log we see

12/10 14:13:12.377 cf0 CACHE: CacheChain::getSettings: Returning settings for the first cache in chain only.
12/10 14:13:12.393 c50 CACHE: CacheChain::getSettings: Returning settings for the first cache in chain only.
12/10 14:13:14.784 1d6c Sp: --- CStarPackSpacePageMemorySegment::SearchEmptyPage(...) (line #484) 38

if cache is WT only there is no transmission stuck, but overall storage speed is poor. And not able to resize lsfs image it's sad. LSFS is ready for production?

What this log means?

[Anatoly (staff)]

First two strings are diagnostic, we will hide them in L1 logging.
The third indicates some problem in the LSFS engine. Can I ask you to upload the following information to some file hosting and share the link with us?
· StarWind service logs at the time of the issue from all problematic SAN boxes
· Windows Application and System logs (in *.csv format) on the time of the issue from all problematic SAN boxes
· Detailed network diagram of SAN system
· Description of the actions that were performed before/at the time of the issue
· Approximate time frames when the issue happened
 
I`d appreciate if you`ll separate the logs from different servers into the different folders

Thank you



Before we will dive into the logs, will it be possible for you to update the StarWind to the Beta, that fixes LSFS problems:
http://www.starwindsoftware.com/forums/ ... 838#p22838