fix: distinguish TB4 from TB5 for Thunderbolt connections#1813
Open
dogukanveziroglu wants to merge 3 commits intoexo-explore:mainfrom
Open
fix: distinguish TB4 from TB5 for Thunderbolt connections#1813dogukanveziroglu wants to merge 3 commits intoexo-explore:mainfrom
dogukanveziroglu wants to merge 3 commits intoexo-explore:mainfrom
Conversation
Contributor
|
What do you mean when you say "breaking placement for TB4 clusters". |
Author
You're right Ring doesn't use RDMA, so the apply.py change was unnecessary. My bad, the RDMA warning on the dashboard confused me into thinking the connection type was the issue, but it was actually ping discovery failing in my environment (macOS LNP blocking check_reachable in SSH sessions). I'll remove the apply.py change and keep only the dashboard warning fix. |
TB4 devices (≤40 Gb/s) don't support RDMA, but macOS Tahoe creates rdma_enX devices for all Thunderbolt ports regardless. This caused Exo to register all Thunderbolt connections as RDMAConnection, breaking placement for TB4 clusters since the ring backend requires SocketConnection edges. Backend (apply.py): - Parse link_speed from ThunderboltIdentifier to detect TB4 vs TB5 - TB5 (>40 Gb/s): create RDMAConnection as before - TB4 (≤40 Gb/s): create SocketConnection via Thunderbolt Bridge IP Dashboard (+page.svelte): - Only show "RDMA NOT ENABLED" warning for actual TB5 hardware - Filter by linkSpeed > 40 Gb/s instead of any TB interface presence Fixes exo-explore#1636
Remove unnecessary backend changes — Ring doesn't use RDMA connections. Only the dashboard linkSpeed check is needed to distinguish TB4 from TB5.
dogukanveziroglu
force-pushed
the
fix/tb4-socket-connections
branch
from
def062d to
533df0c
Compare
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
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
rdma_enXdevices for all Thunderbolt ports. This caused all Thunderbolt connections to be registered asRDMAConnection, breaking placement for TB4 clusters.linkSpeedto distinguish TB4 from TB5: TB5 (>40 Gb/s) getsRDMAConnection, TB4 getsSocketConnectionvia Thunderbolt Bridge IPTest plan
Fixes #1636