silent push handling was deployed on 2026-09-01 to messenger-relay-1
(flamenet-relay:prod, serving relay.flamenet.io) via the staging gate.
The federation fleet — relays-relay1-1, relays-relay2-1, relays-relay3-1, all on flamenet-relay:fnid and up 7 days — is a separate compose project and was not
touched. Those still send an alert push for every envelope, so anything pointed at relay1/2/3.flamenet.io keeps the old behaviour: a burst of "New message" banners for
call signalling, delivery keys and log gossip.
This matters for testing as much as for users — Scripts/lifecycle-gate.sh defaults to relay1.flamenet.io, so the gate exercises the old push path.
Done looks like: the fleet rebuilt from the same commit and deployed, or a decision
recorded that it stays behind and why.
`silent` push handling was deployed on 2026-09-01 to `messenger-relay-1`
(`flamenet-relay:prod`, serving `relay.flamenet.io`) via the staging gate.
The federation fleet — `relays-relay1-1`, `relays-relay2-1`, `relays-relay3-1`, all on
`flamenet-relay:fnid` and up 7 days — is a **separate compose project** and was not
touched. Those still send an alert push for every envelope, so anything pointed at
`relay1/2/3.flamenet.io` keeps the old behaviour: a burst of "New message" banners for
call signalling, delivery keys and log gossip.
This matters for testing as much as for users — `Scripts/lifecycle-gate.sh` defaults to
`relay1.flamenet.io`, so the gate exercises the old push path.
**Done looks like:** the fleet rebuilt from the same commit and deployed, or a decision
recorded that it stays behind and why.
Done 2026-09-04. All three fleet relays were recreated on flamenet-relay:prod — the same image production now runs — after prod was deployed through the staging gate. Verified from off-box: relay1/2/3 all answer 200 on /health and /status.
⚠️ Worth recording why it mattered beyond tidiness: the fleet was 11 days stale, so the live suite would have been testing the Android client against an OLD relay and reporting green about a build it never touched.
Done 2026-09-04. All three fleet relays were recreated on `flamenet-relay:prod` — the same image production now runs — after prod was deployed through the staging gate. Verified from off-box: relay1/2/3 all answer 200 on `/health` and `/status`.
⚠️ Worth recording why it mattered beyond tidiness: the fleet was 11 days stale, so the live suite would have been testing the Android client against an OLD relay and reporting green about a build it never touched.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
silentpush handling was deployed on 2026-09-01 tomessenger-relay-1(
flamenet-relay:prod, servingrelay.flamenet.io) via the staging gate.The federation fleet —
relays-relay1-1,relays-relay2-1,relays-relay3-1, all onflamenet-relay:fnidand up 7 days — is a separate compose project and was nottouched. Those still send an alert push for every envelope, so anything pointed at
relay1/2/3.flamenet.iokeeps the old behaviour: a burst of "New message" banners forcall signalling, delivery keys and log gossip.
This matters for testing as much as for users —
Scripts/lifecycle-gate.shdefaults torelay1.flamenet.io, so the gate exercises the old push path.Done looks like: the fleet rebuilt from the same commit and deployed, or a decision
recorded that it stays behind and why.
Done 2026-09-04. All three fleet relays were recreated on
flamenet-relay:prod— the same image production now runs — after prod was deployed through the staging gate. Verified from off-box: relay1/2/3 all answer 200 on/healthand/status.⚠️ Worth recording why it mattered beyond tidiness: the fleet was 11 days stale, so the live suite would have been testing the Android client against an OLD relay and reporting green about a build it never touched.