Failed to start cluster

[gazzyjuruj]

Hi, I ran the cluster more than 5-6 times with it failing to start since this past morning (about 11-12 hours now) since i'm facing this problem.

Attaching screenshot below and also typing in case someone comes from the web to this thread in future.

Problem : -

It says "failed to start cluster" initially, then i click OK and refresh then i get this below:-

Waiting for cluster to start: Unexpected failure during launch. databricks_error_message: com.databricks.backend.manager.instance.InstanceHolder$PlacementLockTimeout: Could not acquire placement lock context = terminateInstance...

 

Hope somebody comes and fixes this issue or clears me about it, thank you have a nice day all.

[Hubert-Dudek]

Please check the CPU quota.

[gazzyjuruj]

Thank you, yes I have been experiencing databricks_error = user initiated cluster termination

databricks_error_message: timeout while placing nodes

and soetimes it stays stuck on "finding new nodes, instances if neccessary" for hours..

Hopefully it's fixed soon, thanks for all the hard work you do to keep such a great product and community alive.

Have a great day ahead, thanks again!

[Hubert-Dudek]

Yes, in the community edition, I had errors and saw in logs that from 3000 nodes, all are unavailable. I managed to start by deleting it and creating a new one in a different availability zone, but it was hard.

[gazzyjuruj · Accepted solution]

Thank you @Kaniz Fatma​  I can recall now since yesterday everything is working better than ever, fluid smooth! Thanks again, have a great day ahead, really appreciate all the hard work you people do at Databricks, hence Databricks rocks always !! ❤️

[Hubert-Dudek]

Yes, as @gazzyjuruj said, it rocks 🙂

[josstone]

On the node that you are checking, click Start, point to All Programs, click Accessories, right-click Command Prompt, and then click Run as administrator. 

[richardsmith]

Please check the CPU quota.

[worthpicker]

The cluster will be able to start and the nodes will automatically obtain the updated cluster configuration data.

[Richard_Pullman]

Thank you buddy finally my problem is solved.....