Conversation
The 120 s ResponseHeaderTimeout on upstream transports turned queued or cold-loading inference requests into 502s with no way to change it. Add --response-header-timeout (Go duration) to both proxies, falling back to NVPAIR_PROXY_RESPONSE_HEADER_TIMEOUT and then the 120 s default; the broker inherits the env var into its child proxies. Invalid values log a warning and keep the default. Signed-off-by: mkalkere <14184493+mkalkere@users.noreply.github.com>
Signed-off-by: mkalkere <14184493+mkalkere@users.noreply.github.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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Fixes #57: the 120 s upstream response-header timeout (
proxyResponseTimeout) turned any queued or cold model load longer than 120 s into a 502, with no way to raise it and no documentation of the ceiling.Both proxies (
services/ollama-proxy,services/lmstudio-proxy) gain a--response-header-timeoutflag. Precedence follows the repo's existing convention: flag >NVPAIR_PROXY_RESPONSE_HEADER_TIMEOUTenv > 120 s default. Invalid or non-positive values log a warning and keep the 120 s default (fail-safe). The broker spawns proxies as child processes, so the env var is inherited with zero broker changes. The effective value is still logged at startup on the existingresponse_header_timeoutline.Scope
Included: both proxies, flag-table rows in both proxy READMEs,
docs/proxy-response-header-timeout.mdx, a new troubleshooting section for the 120 s 502s. Excluded: broker changes (none needed) and engine launch-option changes (mentioned by the maintainer separately).Validation
go test -race ./...in both proxy modules: pass.go vetclean,gofmtclean,node scripts/spdx-headers.mjsreports 0 missing headers on every branch.Risk
None for existing setups: the default is unchanged at 120 s, so this is purely additive. Minor version bumps (0.26.2 -> 0.27.0, 0.16.2 -> 0.17.0) per VERSIONING.md -- additive, user-visible feature.
Checklist
git commit -s), certifying the Developer Certificate of Origin.services/versions.json, and described user-visible changes above so they reach the release notes.