fix port conflict in parallel dygraph mode #28497
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thanks for your contribution! |
sandyhouse
reviewed
Nov 11, 2020
| @@ -140,21 +140,6 @@ def _check_var_exists(var_name): | |||
| http_server.start() | |||
| wait_server_ready([ParallelEnv().trainer_endpoints[0]]) | |||
There was a problem hiding this comment.
This line can be moved to line #169 (just after the nccl initialization) to decrease the wait time.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR types
Bug fixes
PR changes
Others
Describe
when running dygraph parallel code in local machine, random failure will occur. It may be caused by the port conflict when gloo and nccl initialized at the same time.
rank0 log (send id, connect failed):
rank3 log (receive id, bind failed):
In this PR, we change the order of the gloo and nccl init like below:
HttpServer start -> NCCL init -> GLOO init
HttpServer is used for gloo initial, which will use rank0 port. After changing order, port conflict will be resolved.