Skip to content

Instantly share code, notes, and snippets.

@verdimrc
verdimrc / wsl2-scratchpad.md
Last active July 11, 2024 06:29
wsl2 journey

Journey to the center of WSL2 Ubuntu

ssh matters

Host OS: ensure file permissions comply with ssh requirements.

$ find /mnt/c/Users/vmarch/Documents/.ssh -printf "%M %p\n"  
drwx------ /mnt/c/Users/$USER/Documents/.ssh
-rw------- /mnt/c/Users/$USER/Documents/.ssh/authorized_keys
@verdimrc
verdimrc / ec2-from-sys.md
Created April 19, 2024 07:43
Probe ec2 instances without instance metadata
$ cat /sys/devices/virtual/dmi/id/product_name
trn1.32xlarge

$ cat /sys/devices/virtual/dmi/id/board_asset_tag
i-0000000000example
@verdimrc
verdimrc / mount-s3.sh
Last active April 11, 2024 02:12
mountpoint-for-s3
#!/bin/bash
################################################################################
# NOTE for Slurm users: when Slurm is configured to enable cgroup, upon job
# completion Slurm will kill the mount-s3 process. This causes on-access error
# "transport not connected".
#
# [20240404] In the practical sense, running this script under srun will:
# - not work on pcluster-3.9.0 (ProctrackType=proctrack/cgroup)
# - probably work on SageMaker HyperPod (ProctrackType=proctrack/linuxproc)
@verdimrc
verdimrc / aws-cli-mfa-osx.md
Created March 4, 2024 12:38
aws-cli-mfa-osx.md
# Install awsume https://awsu.me/
$ brew install awsume

# Edit ~/.aws/config, and create a profile:
$ vi ~/.aws/config
[hyperpod]
output = json
region = us-west-2
mfa_serial = arn:aws:iam:::mfa/
@verdimrc
verdimrc / hf-env-vars.sh
Created February 14, 2024 07:47
HF env vars
# https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables
export HF_HOME=/fsx/marcverd/hf_home
export HF_HUB_DISABLE_TELEMETRY=1
@verdimrc
verdimrc / pcluster-notes.md
Created January 18, 2024 02:52
pcluster-notes.md

Various notes on pcluster

PCluster does not work with vanilla DLAMI.

The error message must be pulled from head node's /var/log/cloud-init-output.log.

$ tail /var/log/cloud-init-output.log
...
@verdimrc
verdimrc / py-to-learn.txt
Created January 5, 2024 02:53
Python: selected new features to learn
NOTE: also check additional changes in typing across major python versions.
- 312 pep669: new debugging/profiling API
- 312 pep684: C-API for per-interpreger GIL -> low-level features.
* Py-API expected for 313.
- 312 @override decorator for methods
- 312 pep701: more flexible f-string parsing
- 311 typing: Self type
- 311 exception groups, except*
- 311 pep678: enrich exceptions with notes
@verdimrc
verdimrc / docker-gpu.md
Last active February 14, 2024 07:39
docker-gpu.md
@verdimrc
verdimrc / cdk-BucketDeployment-checkov.ts
Last active April 11, 2023 04:18
Add Checkov metadata to CDK BucketDeployment
interface CheckovRule {
id: string,
comment: string,
}
function silence_checkov(construct: Construct, rules: CheckovRule[]) {
let metadata = (construct.node.defaultChild as cdk.CfnResource).cfnOptions.metadata;
metadata = { checkov: { skip: rules }, ...metadata };
(construct.node.defaultChild as cdk.CfnResource).cfnOptions.metadata = metadata
}
@verdimrc
verdimrc / trainer-logging.sh
Last active February 15, 2023 12:45
Huggingface
#!/bin/bash
#set -aex
echo "PWD = $(pwd)"
: "${SM_NUM_GPUS:=4}"
: "${MODEL_NAME:=gpt2}"
: "${OUTPUT_ROOT:=/mnt/scratch}"
: "${TRAINING_JOB_NAME:=haha}"