reset flight recorder trace #283
Closed
Conversation
This file contains hidden or 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
|
@tushar00jain has exported this pull request. If you are a Meta employee, you can view the originating Diff in D84260745. |
meta-cla
bot
added
the
CLA Signed
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 16, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
2 times, most recently
from
1d99280 to
d048341
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 16, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Differential Revision: D84260745
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 16, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
d048341 to
19ec38e
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 16, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
19ec38e to
06ae4f8
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 17, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
06ae4f8 to
2132ea5
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 17, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
2132ea5 to
eaf4f9e
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 17, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
eaf4f9e to
229e4fa
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 21, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
229e4fa to
85840e1
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 21, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
2 times, most recently
from
ec50ef7 to
575d7c0
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 21, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 21, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
575d7c0 to
9397313
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 21, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
9397313 to
d52d777
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 21, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
d52d777 to
931fa7a
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Oct 24, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
dc0194c to
3698465
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Nov 3, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
3698465 to
2f87b09
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Nov 4, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
2f87b09 to
20ed64e
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Nov 6, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
2 times, most recently
from
e603706 to
59b647a
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Nov 6, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum - return replica id's in quorum response so we can determine global ranks in the pg - this is used to set the metadata on the pg for flight recorder to work Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Nov 6, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum - return replica id's in quorum response so we can determine global ranks in the pg - this is used to set the metadata on the pg for flight recorder to work Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
59b647a to
b83be4c
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Nov 6, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum - return replica id's in quorum response so we can determine global ranks in the pg - this is used to set the metadata on the pg for flight recorder to work Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
2 times, most recently
from
189dbad to
c5a3407
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Nov 6, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum - return replica id's in quorum response so we can determine global ranks in the pg - this is used to set the metadata on the pg for flight recorder to work Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Nov 6, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum - return replica id's in quorum response so we can determine global ranks in the pg - this is used to set the metadata on the pg for flight recorder to work Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
c5a3407 to
661a91d
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Nov 6, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum - return replica id's in quorum response so we can determine global ranks in the pg - this is used to set the metadata on the pg for flight recorder to work Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
from
661a91d to
7d73b88
Compare
Summary: Remove device mesh since we don't really use it. Device mesh is undergoing a lot of changes and using private api's makes the subclass difficult to maintain. We will revisit device mesh integration with public api's. Differential Revision: D86466239
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum - return replica id's in quorum response so we can determine global ranks in the pg - this is used to set the metadata on the pg for flight recorder to work Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Nov 7, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum - return replica id's in quorum response so we can determine global ranks in the pg - this is used to set the metadata on the pg for flight recorder to work Reviewed By: d4l3k Differential Revision: D84260745
tushar00jain
force-pushed
the
export-D84260745
branch
2 times, most recently
from
c28793a to
831683f
Compare
tushar00jain
added a commit
to tushar00jain/torchft
that referenced
this pull request
Nov 7, 2025
Summary: - call FR api to reset the trace after every quorum - we reset so that after every quorum, we start a fresh FR trace since the pg's could have changed and we already dumped FR trace from previous errors - change the env var that's used to determine the file after every quorum - return replica id's in quorum response so we can determine global ranks in the pg - this is used to set the metadata on the pg for flight recorder to work Reviewed By: d4l3k Differential Revision: D84260745
|
This pull request has been merged in 854fb2d. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Summary:
Differential Revision: D84260745