rapidsai / rapidsai/deployment

EC2-MNMG instructions do not connect to workers within NVIDIA AWS instances

Open
#308 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Jupyter Notebook
Stars
15
Forks
41
Avg merge
1d 8h
Merged PRs (30d)
6

Description

Deploying an ec2-mnmg cluster setup does not properly connect the workers that were created (and can be shut down) properly.

https://docs.rapids.ai/deployment/nightly/cloud/aws/ec2-multi/#cluster-setup

This may be a network issue with NVIDIA security groups or an issue with dask-cloud provider, which is reportedly needs more resources. Found during 23.12 deployment testing. cc @aravenel @jacobtomlinson

Proposed solutions:

  1. Test with a more permission security group, closer to our customer's or average user
  2. If the issue persists after testing one, allocate resources to fix any bugs in dask cloud provider.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the EC2-MNMG cluster setup instructions at the linked deployment page and reproduce the worker connection failure in an NVIDIA AWS instance. Test the proposed security-group change first, then determine whether the dask-cloud-provider resource issue remains; done means workers connect correctly and can be shut down as described.

Written by the indexing model from the issue text.

Assessment

Tech stack
aws
Domain
cloud, infrastructure
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.