Skip to content

Back off endpoint rebuilds during relay outages - #94

Merged
andrewchen5678 merged 3 commits into
mainfrom
relay-rebuild-backoff
Sep 4, 2026
Merged

andrewchen5678 merged 3 commits into
mainfrom
relay-rebuild-backoff

Conversation

@andrewchen5678

Copy link
Copy Markdown
Contributor

Implement a backoff strategy for endpoint rebuilds when relays are unreachable, preventing frequent drops of LAN clients. The watchdog now tracks relay connectivity and adjusts rebuild deadlines accordingly. Update version to 0.0.72 for flextunnel packages.

Andrew Chen and others added 3 commits September 4, 2026 07:07
A rebuild only helps when iroh's relay bookkeeping went stale. When the
relay itself is unreachable the fresh endpoint never registers either (the
rebuild factory tolerates a failed online wait), so the watchdog tripped
again 180s later and every LAN client was dropped every few minutes for as
long as the relay outage lasted.

The watchdog now returns a RelayOutage that says whether a home relay was
connected at any point of the watch, and takes its rebuild deadline as a
parameter. run_server doubles that deadline for each consecutive endpoint
that never registered (180s, 6m, 12m, 24m, capped at 30m) and resets it
once an endpoint does register. The 60s nudge and the 30s bind-failure
retry are unchanged. Same change as ezvpn's relay-watchdog branch; the
module stays identical between the two repos.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PFw1C9hzWKyoieoxWVSqbA
@andrewchen5678
andrewchen5678 merged commit 07ef94c into main Sep 4, 2026
4 checks passed
@andrewchen5678
andrewchen5678 deleted the relay-rebuild-backoff branch September 4, 2026 14:14
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant