VSAN Cluster not synching

Software-based VM-centric and flash-friendly VM storage + free version

Moderators: anton (staff), art (staff), Max (staff), Anatoly (staff)

Post Reply
WolfR1der
Posts: 13
Joined: Tue Apr 24, 2018 8:30 pm

Tue Jun 30, 2020 5:38 pm

We have had a two node VSAN cluster running for a year now and were supposed to go to a hardware solution well before the license expired. Because of the pandemic all incoming hardware was very slow to arrive and we still haven't been able to implement it. In the mean time the license for our Starwinds setup expired. We were able to obtain new licenses and update the software for the short term however neither of our servers are synchronizing or even attempting to synch. Both show that they can reach each other via their 10Gig synch channel but nothing is happening. This cluster is used for remote desktops so our RDS setup connected to this cluster is down.

How do I force a synch?
yaroslav (staff)
Staff
Posts: 2279
Joined: Mon Nov 18, 2019 11:11 am

Tue Jun 30, 2020 5:56 pm

Could you tell me how long the servers are hanging in that state?
So, HA devices on both servers are hanging in not synchronized state? If that's the case, this article might be helpful https://knowledgebase.starwindsoftware. ... -blackout/.
Could you collect the logs? Share them via Google Drive or DropBox. For quicker and easier log collection from StarWind nodes please do not hesitate using StarWind Log Collector from our knowledge base article below:
https://knowledgebase.starwindsoftware. ... collector/. I want to see the logs anyway to assist you.
SyncHaDevice script resembles the "Mark as Synchronized" button.
WolfR1der
Posts: 13
Joined: Tue Apr 24, 2018 8:30 pm

Tue Jun 30, 2020 6:28 pm

Not completely sure when they fell out of synch, sometime over the weekend so could be up to 48 hours. This is little to no IO occurring between these servers. IOPs is 0 and network traffic is generally in the KB over a 10Gig link.

I am making a copy of the drive files before I implement any forced synch. I will run the log collector as soon as the copy is completed and before any forced synch.
yaroslav (staff)
Staff
Posts: 2279
Joined: Mon Nov 18, 2019 11:11 am

Tue Jun 30, 2020 7:20 pm

I would like to have the logs anyway. Please share them as soon as possible so that I could assist you.
Yes, it is good to have a backup in such a situation.

Please share logs with me as soon as possible. I would also recommend opening a support case by filling in this form https://www.starwindsoftware.com/support-form. Use this forum thread as a reference.
WolfR1der
Posts: 13
Joined: Tue Apr 24, 2018 8:30 pm

Tue Jun 30, 2020 7:54 pm

Here are the logs from both servers. Please let me know when I can delete them.

Server1
https://drive.google.com/file/d/1pgPOv2 ... sp=sharing

Server2
https://drive.google.com/file/d/1weLhD5 ... sp=sharing

I'm going to wait to force the synch until you've had a chance to look at these.
yaroslav (staff)
Staff
Posts: 2279
Joined: Mon Nov 18, 2019 11:11 am

Tue Jun 30, 2020 9:20 pm

Sorry for delayed response.

So, I found some misconfiguration in logs.
First, iSCSI is configured with the Default value for Initiator IP. It should be reconfigured according to this guide https://www.starwindsoftware.com/resour ... rver-2016/.
There are also no local witness connections in iSCSI initiator.

In other words, due to iSCSI initiator settings the storage is not highly available. Please see the guide above to learn how to reconfigure the storage.

HA devices on both nodes became not synchronized simultaneously. You can follow this article to recover the synchronization status.
https://knowledgebase.starwindsoftware. ... -blackout/.
You can check data consistency by reviewing VM system logs. If the system logs contain the entries when you expect the VM running, that means that data is fine. You can check each node one by one as it discussed in the article provided above.

Please let me know if there is anything else I can assist you with.

Please note that if there is any production, neither StarWind Trial nor NFR is intended for production.
WolfR1der
Posts: 13
Joined: Tue Apr 24, 2018 8:30 pm

Wed Jul 01, 2020 4:17 am

Thanks for all the help. Looks like we are up and running.
We are just weeks away from moving to a full hardware storage system so we won't needs Starwinds but you guys have been a great help!

As to that first link not sure what I'm missing. When I set it up I recall setting it up according to the SW documentation. Maybe I missed something. I will have to peruse the document further tomorrow to see what I've improperly set.
yaroslav (staff)
Staff
Posts: 2279
Joined: Mon Nov 18, 2019 11:11 am

Wed Jul 01, 2020 7:55 am

Hi WolfR1der,

Happy to know that the cluster is up and running!
In iSCSI Initiator, you need to select local IP for iSCSI as initiator IP, and partner iSCSI IP as the target. Also, Microsoft iSCSI Initiator is to be set as the Local adapter. Find out more from the Provisioning StarWind HA Storage to Windows Server Hosts section of the guide I shared with you.
Post Reply