deps: bump to the released fixes, for a v1.13.8 that actually carries them - #441
Merged
Conversation
… them
Six modules were released today with fixes for defects found in a
codebase-wide sweep, but this repo still pinned the versions from before
them. Tagging a release in that state would have shipped a binary carrying
none of the fixes — the versions would move and nothing would change.
handshake v0.2.6 -> v0.2.7 port 444 leaked a goroutine + 64 KiB per
accepted connection; Close appeared
nowhere in the file
beacon v0.2.8 -> v0.2.9 dstNodes grew unbounded from unauthenticated
datagrams and was in no sweep
rendezvous v0.2.7 -> v0.2.8 RateLimiter.Allow had ZERO call sites, so
the per-IP tier was dead code and
pilot_ratelimit_denied_total{kind=ip} was
structurally always zero
dataexchange beta -> v0.2.1 peer-chosen transfer ids opened an *os.File
each with no cap -> FD exhaustion
eventstream v0.2.2 -> v0.2.3 time.After per subscriber per event pinned
up to 500*S timers
runtime already v0.3.2 (bumped with #435; listed for completeness)
trustedagents, policy and webhook are deliberately NOT bumped: their newest
tags are betas and this repo tracks stable.
Verified the fixes are actually in the resolved modules rather than trusting
version strings — grepped the module cache for the close-on-exit defer,
maxConcurrentTransfers, the dstNodes sweep, the NewTimer replacement and the
rateLimiter.Allow call site. All present.
Built and tested with GOWORK=off so resolution matches CI rather than the
local go.work overlay: build clean, vet clean, pkg/daemon suite green.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
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
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.
Six modules were released today with fixes from a codebase-wide sweep — but this repo still pinned the versions from before them.
Tagging a release in that state would have shipped a binary carrying none of the fixes. The version number would move and nothing would actually change.
Closeappeared nowhere in the file)dstNodesgrew unbounded from unauthenticated datagrams, in no sweepRateLimiter.Allowhad zero call sites — per-IP tier was dead code*os.Fileeach, no cap → FD exhaustiontime.Afterper subscriber per event pinned up to500*Stimerstrustedagents,policyandwebhookare deliberately not bumped — their newest tags are betas and this repo tracks stable.Verified the fixes are really there
Rather than trusting version strings, I grepped the resolved module cache for the actual code:
Built and tested with
GOWORK=offso resolution matches CI rather than the localgo.workoverlay: build clean, vet clean,pkg/daemonsuite green.Once this lands, tagging
v1.13.8produces a release that genuinely incorporates the whole sweep.🤖 Generated with Claude Code