aztk

aztk copied to clipboard

paselem

Job UI not always available

2

Job UI only runs on the node where the Driver is executed. If this isn't the master node, then when you ssh and try to go to localhost:4040, you get...

jafreck

jafreck

jafreck

Currently, job submission mode doesn't have ability to specify file shares: https://github.com/Azure/aztk/blob/17755e0ffadfca78908c9a4b4c0e7a6e78c2dda3/aztk/spark/client.py#L170

jafreck

For plugins that run in the Spark container, some client-side validation of toolkit (docker-image) and plugin compatibility would prevent a lot of confusion.

jafreck

By placing the correct driver, we should be able to read/write data from/to MS SQL server. [https://blogs.msdn.microsoft.com/bigdatasupport/2015/10/22/how-to-allow-spark-to-access-microsoft-sql-server/](url)

praneet22

There are many errors that could happen when deploying a new cluster that just get silently ignored. We should make sure the node gets into `StartTaskFailed` state if any of...

timotheeguerin

Feature: Add --apps option to get job command to get all job info in one go

5

After submitting a job, I often find myself juggling several commands to get all the info about it (`get`, `list-apps`, `get-app`, `get-app-logs`), so I put together one command that gets...

emlyn

Since #534, any node can be ssh'ed into. We should allow specifying the node_id in the ssh command. `aztk spark cluster ssh --id {cluster-id} --node-id {node-id}` By default, we still...

jafreck