frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

Debugging using with ssh for Github Actions - add table of contents, last verified date, page maintainers, note about windows

clee2000
2025-06-04 09:55:53 -07:00
parent 92f87e7061
commit 7cc7e55bc7
1 changed files with 25 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
Debugging-using-with-ssh-for-Github-Actions.md

@ -1,5 +1,18 @@
# SSH for Github Actions
- [SSH for Github Actions](#ssh-for-github-actions)
- [Platform availability:](#platform-availability)
- [Conditions and Limitations](#conditions-and-limitations)
- [Workflow for users](#workflow-for-users)
- [VSCode](#vscode)
- [Linux](#linux)
- [Windows](#windows)
- [RDP](#rdp)
- [(For META Employees) Debugging using AWS SSM](#for-meta-employees-debugging-using-aws-ssm)
- [Conditions and Limitations](#conditions-and-limitations-1)
- [Usage](#usage)
## Platform availability:
- [x] Linux (https://github.com/pytorch/pytorch/pull/62280)
@ -74,11 +87,12 @@ AWS SSM can be used to log into _any_ currently running EC2 instances whether or
This can be used to debug currently running jobs on PRs or trunk.
### Pre-requisites
* Network connection on META VPN
* Runner is from the AWS Meta account
* Cloud SSO Access (`bunnylol cloud fbossci`)
* [Session manager plugin](https://docs.aws.amazon.com/systems-manager/latest/userguide/session-manager-working-with-install-plugin.html) for AWS CLI (`aws ssm start-session` will fail silently without it)
### Conditions and Limitations
* You must be connected to the META VPN
* The runner must be from the AWS Meta account (see above about how to confirm this)
* You must have access to the AWS account (`bunnylol cloud fbossci`)
* Download the [session manager plugin](https://docs.aws.amazon.com/systems-manager/latest/userguide/session-manager-working-with-install-plugin.html) for AWS CLI (`aws ssm start-session` will fail silently without it)
* Jobs will not wait for your session to finish before concluding and starting a new job, and nodes will still be reaped on their regular schedule. There is no extra hold on the machine, and the 2 hour timeout does not apply here, so sessions will most likely end immediately at the next reap cycle.
### Usage
```bash
@ -91,7 +105,9 @@ Example:
aws ssm start-session --target i-0099d5a07d34e8904 --region us-east-1
```
### Caveats
* Nodes will still be reaped on their regular schedule
* Jobs will not wait for your session to finish before concluding
* There is no extra hold on the machine and the 2 hour timeout does not apply here, sessions will most likely end immediately at the next reap cycle
---
Page maintainers: @pytorch/pytorch-dev-infra
<br>
Last verified: 2025-06-04
<br>
Everything except the windows section should be correct and up to date