Starwind witness node corrupted

[Chinny]

Hi,

In our situation we have 3 servers and created a Starwind VSAN 2 node setup with a extra SW node. Node 1&2 contains the VSAN and the Node 3 acts as a witness node only.

However, due to unknown reasons, i cannot reach the management web page anymore of the witness server. We have spent alot of time trying to find out what the problem is on the witness server, but I figured it would be easier to recreate the witness server with the OVA deployment. Right now, the witness server is recreated with a empty Starwind configuration.

Node 1 & Node 2 are syncing toward each other and is synchronized and are online right now. The iSCSI is also available in VMWare. They are also communicating towards the witness server and i can see Node 1 and Node 2 are trying to reach the target of the witness server.

To minimize downtime, How can I recreate the witness node on the witness server with powershell without affecting node 1 & 2?

Any help would be greatly appreciated!

[yaroslav (staff)]

Hi,

Did you delete the old files from the storage (i.e., the OS disk and the headers)?

[Chinny]

Hi,

Did you delete the old files from the storage (i.e., the OS disk and the headers)?

I dont know what you mean, I still have the old witness server VM with the configuration, yes. I just disabled the ethernet adapters for now.
Then created a new witness server VM with a empty configuration, the management page works on this VM.
If I understood you wrong, What do you mean exactly or which paths do you mean?

[yaroslav (staff)]

Go into that Witness server and copy
1. StarWind.cfg
2. Headers from /opt/starwind/starwind-virtual-san/drive_c/starwind/headers
3. *.swdsk.bak files (only you know where you created those. might be some /dev/sdb1/ path)