hpc / hpc/mpifileutils

Is dsync walking all target items when updating timestamps?

Open
#425 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C
Stars
200
Forks
85
Avg merge
3d 21h
Merged PRs (30d)
2

Description

Looking at how long the "Updating timestamps on newly copied files" phase takes it looks to me as if dsync is walking ALL items on the target when doing that, instead of, as the message says, only the "newly copied files".

[2020-11-13T13:29:39] Syncing directory updates to disk.
[2020-11-13T13:29:39] Sync completed in 0.018284 seconds.
[2020-11-13T13:29:39] Started: Nov-13-2020,13:28:07
[2020-11-13T13:29:39] Completed: Nov-13-2020,13:29:39
[2020-11-13T13:29:39] Seconds: 92.411
[2020-11-13T13:29:39] Items: 30039
[2020-11-13T13:29:39]   Directories: 13
[2020-11-13T13:29:39]   Files: 30026
[2020-11-13T13:29:39]   Links: 0
[2020-11-13T13:29:39] Data: 1.206 TB (1325900635063 bytes)
[2020-11-13T13:29:39] Rate: 312.349 MB/s (30266497975 bytes in 92.411 seconds)
[2020-11-13T13:29:39] Updating timestamps on newly copied files

It's now 14:31 and looking at one of the mpi processes it's processing a lot more than 30026 files...

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

No source files or tests are named. Start by tracing dsync's "Updating timestamps on newly copied files" phase and compare the target items it visits with the newly copied file count in the report; done when the traversal scope is established and matches the documented behavior.

Written by the indexing model from the issue text.

Assessment

Tech stack
c
Domain
cli, performance
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.