fix(electrum): handle shutdown race in acceptor thread#209
Open
EddieHouston wants to merge 1 commit intoBlockstream:new-indexfrom
Open
fix(electrum): handle shutdown race in acceptor thread#209EddieHouston wants to merge 1 commit intoBlockstream:new-indexfrom
EddieHouston wants to merge 1 commit intoBlockstream:new-indexfrom
Conversation
Collaborator
|
is there a way to reliably reproduce this locally/in test? |
shesek
approved these changes
Apr 21, 2026
Collaborator
shesek
left a comment
shesek
left a comment
There was a problem hiding this comment.
The changes makes sense and seem correct. Agreed that it would be great to have a test/reproduction, but I see that it can be kind of tricky.
Collaborator
|
LGTM |
The 'acceptor' thread panicked with 'send failed' when a TCP connection
arrived at the exact moment the server was shutting down. The 'rpc'
thread drops the channel receiver upon receiving the Exit notification,
leaving the 'acceptor' thread's next send() with no receiver.
- Break cleanly from the acceptor loop when the receiver is dropped
- Replace unwrap() with a warn! log in the notification thread for the
symmetric case where the acceptor has already exited before Exit fires
EddieHouston
force-pushed
the
fix/electrum-acceptor-shutdown-race
branch
from
e8380f0 to
e4b15a4
Compare
Collaborator
Author
Thought about that... it is tricky as it is such a narrow window between rpc thread processing Exit and the acceptor thread's next send(). So "reliably test" is the key... not really. It's a fairly straight forward fix without a downside I can see. |
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
4 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Fixes a shutdown race we saw in the Electrum RPC server that caused the
acceptorthread to panic with"send failed"when a TCP connectionarrived at the exact moment the server was tearing down.
Root Cause
The
acceptorthread and therpcthread communicate via a channel:acceptorthread callslistener.accept()in a loop and sendseach new connection as
Some((stream, addr))to therpcthreadnotificationthread sendsNoneinto the channelrpcthread receivesNone, exits itswhile let Some(...)loop,and drops the channel receiver
acceptorthread accepts a new TCP connection at this momentand calls
send(Some(...)), the send fails because the receiver isalready gone →
.expect("send failed")→ panicThis is a normal production scenario: on a live Electrum server with
active wallet connections, a client connecting during shutdown is
expected.