Run everything via `ssh` + `docker exec` as opposed to SSH Proxy.

run-house / runhouse

Dispatch and distribute your ML training to "serverless" clusters in Python, like PyTorch for ML infra. Iterable, debuggable, multi-cloud/on-prem, identical across research and production.

https://run.house

Apache License 2.0

977 stars 37 forks source link

Run everything via `ssh` + `docker exec` as opposed to SSH Proxy. #1235

Closed rohinb2 closed 2 months ago

sentry-io[bot] commented 2 months ago

🔍 Existing Issues For Review

Your pull request is modifying functions with the following pre-existing issues:

📄 File: runhouse/resources/hardware/cluster.py

Function	Unhandled Issue
`_sync_runhouse_to_cluster`	ValueError: Error installing runhouse on cluster node <34.229.108.213> ... `Event Count:` 1

_{Did you find this useful? React with a 👍 or 👎}

rohinb2 commented 2 months ago

#1235 : 2 dependent PRs (#1252 , #1253 ) 👈
#1251
#1239
#1238
#1237
main

This stack of pull requests is managed by Graphite. Learn more about stacking.

Join @rohinb2 and the rest of your teammates on Graphite

rohinb2 commented 2 months ago

Merge activity

Sep 11, 5:04 PM EDT: @rohinb2 started a stack merge that includes this pull request via Graphite.
Sep 11, 5:09 PM EDT: Graphite rebased this pull request as part of a merge.
Sep 11, 5:10 PM EDT: @rohinb2 merged this pull request with Graphite.