topic Re: Quantum Spark cluster locally managed is stuck at "step 3" of primary member in Spark Firewall (SMB)

Quantum Spark cluster locally managed is stuck at "step 3" of primary member

Wei_Soon_Heng — Wed, 17 Jan 2024 12:21:58 GMT

Hi All,

Did anyone experiences the issue stuck at step 3 "Applying settings(This may take a few minutes)" at primary member when configuring cluster(locally-managed)?

Please refer to screenshot below:

Re: Quantum Spark cluster locally managed is stuck at "step 3" of primary member

Chris_Atkinson — Wed, 17 Jan 2024 13:51:40 GMT

Please confirm the version used and attach the screenshot?

Re: Quantum Spark cluster locally managed is stuck at "step 3" of primary member

G_W_Albrecht — Thu, 18 Jan 2024 09:17:05 GMT

I would suggest to try a firmware install from CLI:

- Perform the following Expert Mode command in order to allow SCP protocol connection:

bashUser on

- Please download the latest GA version offirmware for your appliance
- Enter the appliance via WinSCP connection of SCP protocol and upload to the /storage directory the latest GA version image.
- Before uploading image to /storage directory, all the files in this directory have to be deleted.
- Perform the following Expert Mode command for a manual upgrade via CLI:

upgrade_revert_image.sh /storage/%FILENAME% upgrade safe

Re: Quantum Spark cluster locally managed is stuck at "step 3" of primary member

Wei_Soon_Heng — Mon, 22 Jan 2024 05:52:30 GMT

The firmware version is R81_10_08 with build number 996001608. I tried to upgrade to latest build number 996001683 but issue still persists. I tried refresh browser after waiting for one hour, and it shows "loading data" in page of High Availability. The output of command "configload_status" showing cluster service "Queued to Run" which cause the secondary member having error establishing trust when joining HA to primary member.

Re: Quantum Spark cluster locally managed is stuck at "step 3" of primary member

Pedro_Espindola — Wed, 24 Jan 2024 17:25:19 GMT

Try disabling out of state protections in advanced settings of both members, create the cluster and when everything is fine, then enable it again.

Also, I don't know if this is your case, but I had this kind of issue when creating a cluster connected to Spark Management in Infinity Portal, which has lots of limitations. There will be no connection sync and the standby member will not be able to complete TCP connections or ping, unless you use the cluster fold option, which forces routing of outbound connections through the active member using VIP address.