Page 2 of 4
Re: HA device timeout error
Posted: Thu Feb 08, 2024 11:18 am
by yaroslav (staff)
MPIO can be the Least Queue Depth for witnesse HA too.
Are other devices (i.e., data), also connected from both servers?
Re: HA device timeout error
Posted: Sat Feb 10, 2024 8:25 am
by bertram
logitech wrote:
2. StarWind VSAN 8.0.15260 installed
an old version installed, we will plan for the upgrade soon.
What version do you running? I can remember we had the same issue after installing Version V8 (build 14947) but I can not say from which version. The seperate active/active cluster was running fine after the update for ~2 weeks but out of nothing your mentioned issues appeared. After updating to the successor Version V8 (build 15020) the problem disappeared. I assume a bug was fixed with this version when I read the notes: "Synchronous Replication - Fixed spurious replication failure."
I hope you can solve your problem.
Regards
betram
Re: HA device timeout error
Posted: Mon Feb 12, 2024 7:10 pm
by logitech
Hi,
We updated the policy settings for all connections from virtual hosts to the storage nodes to "Least Queue Depth" .
Do we need to update the policy settings from Storage Node A to Storage Node B too?
Re: HA device timeout error
Posted: Mon Feb 12, 2024 9:33 pm
by yaroslav (staff)
You need to update for both nodes.
Re: HA device timeout error
Posted: Tue Feb 13, 2024 9:54 am
by logitech
I already updated the connection policy on vh01 and vh02. I don't see any iSCSI connection between storage01 and storage02 to update the policy settings.
what do you mean by both nodes? the hypervisor or storage?
Re: HA device timeout error
Posted: Tue Feb 13, 2024 11:19 am
by yaroslav (staff)
I mean it for hypervisor. It has to be on all hypervisor nodes for all ha devices.
Re: HA device timeout error
Posted: Tue Feb 13, 2024 1:13 pm
by logitech
done, does it require any restart to take effect?
Re: HA device timeout error
Posted: Tue Feb 13, 2024 1:47 pm
by yaroslav (staff)
Nope.
Re: HA device timeout error
Posted: Wed Feb 14, 2024 8:50 pm
by yaroslav (staff)
Updates are installed manually.
See
https://knowledgebase.starwindsoftware. ... d-version/. This procedure describes how to avoid downtime.
For CVM, press the Gear button at the top -> Update. The general prerequisites: check storage connectivity, check HA devices synchronization, etc. are the same.
Re: HA device timeout error
Posted: Sat Feb 17, 2024 6:18 pm
by logitech
Hi,
What is the difference between MCS policy and MPIO policy? both should be Least Queue Depth.
Re: HA device timeout error
Posted: Sat Feb 17, 2024 9:44 pm
by yaroslav (staff)
We recommend the Least Queue Depth as it grants the best performance.
Re: HA device timeout error
Posted: Sun Feb 18, 2024 8:15 am
by logitech
should we apply Least Queue Depth policy on MPIO and MCS policy settings?
Re: HA device timeout error
Posted: Mon Feb 19, 2024 4:03 am
by yaroslav (staff)
Could you please let me know what MCS is?
Re: HA device timeout error
Posted: Mon Feb 19, 2024 6:54 pm
by logitech
- mcs2.png (33.49 KiB) Viewed 1087 times
- mcs1.png (12.7 KiB) Viewed 1087 times
Re: HA device timeout error
Posted: Mon Feb 19, 2024 8:21 pm
by yaroslav (staff)
Thank ypu.
For one path it does not matter for multiple paths please use Least Queue Depth.