Skip to content

fix: enable upstream connect retry by default #114

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
secxena merged 2 commits into from
Jul 4, 2026
Merged

fix: enable upstream connect retry by default #114

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
51c6d4c
fix: enable upstream connect retry by default
secxena Jul 4, 2026
bc61cbd
docs: surface upstream_retry keys in soth.example.yaml
secxena Jul 4, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 7 additions & 1 deletion crates/soth-cli/src/cli_config.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -98,7 +98,13 @@ impl Default for ForwardProxyConfig {
// response. 120s gives streaming LLM responses the headroom
// they need without hiding genuinely-stuck upstreams.
upstream_timeout: DurationSetting::millis(120_000),
upstream_retry_on_failure: false,
// On by default: a single retry (after upstream_retry_delay, bounded
// by the connect deadline) for transient connect failures — refused
// / unreachable / reset — which are routine during hotspot handoffs
// and gateway blips. Without it a single blip surfaces to the user
// as a 502. The retry only fires on transient errors, never on
// timeouts, so it can't mask a genuinely-stuck upstream.
upstream_retry_on_failure: true,
upstream_retry_delay: DurationSetting::millis(200),
capture_max_body_bytes: 64 * 1024 * 1024,
buffer_request_bodies: true,
Expand Down
5 changes: 4 additions & 1 deletion crates/soth-proxy/src/config.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -264,7 +264,10 @@ impl Default for MitmRuntimeConfig {
h2_body_idle_timeout_ms: 60_000,
h2_response_overflow_mode: H2ResponseOverflowModeConfig::TruncateContinue,
upstream_connect_timeout_ms: 5_000,
upstream_retry_on_failure: false,
// On by default (matches the CLI-generated config): one retry for
// transient connect failures during hotspot/gateway blips; never
// retries timeouts.
upstream_retry_on_failure: true,
upstream_retry_delay_ms: 200,
verify_upstream_tls: true,
max_connections_per_host: 64,
Expand Down
8 changes: 8 additions & 0 deletions soth.example.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -121,6 +121,14 @@ forward_proxy:
cache_ttl: "1h" # Certificate cache TTL
cache_max: 10000 # Maximum cached certificates

# Upstream connect retry. On by default: a single retry (after
# upstream_retry_delay, bounded by the connect deadline) for transient
# connect failures — refused / unreachable / reset — which are routine
# during hotspot handoffs and gateway blips. Timeouts are never retried,
# so it can't mask a genuinely-stuck upstream. Set to false to disable.
upstream_retry_on_failure: true
upstream_retry_delay: "200ms"

# Connection pool settings
pool:
max_connections_per_host: 10
Expand Down
Loading