linkedin / linkedin/dynamometer

Blockgen job fails to clean up failed reduce attempts

Open
#56 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Java
Stars
134
Forks
34
PR merge metrics
No merged PRs in 30d

Description

The block generation job has custom output logic to allow each reducer to output to multiple block files.

When speculative execution is enabled, this can result in two copies of the same block file being generated (one of which may be incomplete). This can be worked around by setting mapreduce.reduce.speculative = false.

When a reducer attempt fails, the partial output files will not be cleaned up. I'm not aware of an easy workaround for this beyond manually cleaning up the files after the job completes.

We should have each reducer use a staging directory and only move the output files when it completes.

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 by locating the block generation job's custom reducer output logic and how failed reducer attempts are handled. Verify that outputs are staged until successful completion and that failed attempts leave no partial block files, including when speculative execution is enabled.

Written by the indexing model from the issue text.

Assessment

Tech stack
java
Domain
distributed-systems, performance
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.