dunglas / mercure / 24782398641

Builds Branch Commit Type Ran Committer Via Coverage
24782398641 main chore: prepare release 0.23.2 push dunglas github
83.89
24781496320 main fix(bolt): stop leaking post-subscribe events into Last-Event-ID header (#1215) dispatchHistory used to update responseLastEventID with each key the cursor visited BEFORE checking whether that key was already past the subscriber's sequence snapsh... push web-flow github
83.89
24781484449 main fix(chart): multiple fixes + SSE-friendly rolling update defaults (#1213) feat(chart): SSE-friendly rolling update defaults for the hub Deployment Hardcode three fields on the Deployment that make a rolling update non-disruptive for hubs carryin... push web-flow github
84.1
24781128534 fix/chart-issues feat(chart): SSE-friendly rolling update defaults for the hub Deployment Hardcode three fields on the Deployment that make a rolling update non-disruptive for hubs carrying live SSE subscribers: - strategy.rollingUpdate.maxSurge=1, maxUnavailabl... Pull #1213 dunglas github
84.1
24780847762 fix/chart-issues feat(chart): SSE-friendly rolling update defaults for the hub Deployment Hardcode three fields on the Deployment that make a rolling update non-disruptive for hubs carrying live SSE subscribers: - strategy.rollingUpdate.maxSurge=1, maxUnavailabl... Pull #1213 dunglas github
84.1
24780844452 main fix(subscribe): drain SSE subscribers naturally on hub shutdown (#1212) On hub shutdown (Caddy "stopping" event, pod receiving SIGTERM, graceful config reload), the subscriber select loop no longer observes h.ctx.Done() when writeTimeout is enabl... push web-flow github
84.1
24779259645 fix/graceful-shutdown fix(subscribe): drain SSE subscribers naturally on hub shutdown On hub shutdown (Caddy "stopping" event, pod receiving SIGTERM, graceful config reload), the subscriber select loop no longer observes h.ctx.Done() when writeTimeout is enabled. Exis... Pull #1212 dunglas github
84.1
24779140259 fix/bolt-unknown-lasteventid fix(bolt): stop leaking post-subscribe events into Last-Event-ID header dispatchHistory used to update responseLastEventID with each key the cursor visited BEFORE checking whether that key was already past the subscriber's sequence snapshot. Unde... Pull #1215 dunglas github
83.76
24777317976 fix/chart-issues feat(chart): SSE-friendly rolling update defaults for the hub Deployment Hardcode three fields on the Deployment that make a rolling update non-disruptive for hubs carrying live SSE subscribers: - strategy.rollingUpdate.maxSurge=1, maxUnavailabl... Pull #1213 dunglas github
83.97
24777270645 docs/rolling-updates Update docs/hub/rolling-updates.md Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Kévin Dunglas <kevin@dunglas.fr> Pull #1216 web-flow github
83.97
24777136196 fix/bolt-unknown-lasteventid fix(bolt): stop leaking post-subscribe events into Last-Event-ID header dispatchHistory used to update responseLastEventID with each key the cursor visited BEFORE checking whether that key was already past the subscriber's sequence snapshot. Unde... Pull #1215 dunglas github
83.79
24777042251 docs/rolling-updates docs(hub): document rolling updates and graceful shutdown New hub/rolling-updates.md explains how the hub drains SSE subscribers via write_timeout on shutdown, why restarts don't produce reconnect storms, and how to configure Kubernetes (terminat... Pull #1216 dunglas github
83.97
24776834198 fix/graceful-shutdown fix(subscribe): drain SSE subscribers naturally on hub shutdown On hub shutdown (Caddy "stopping" event, pod receiving SIGTERM, graceful config reload), the subscriber select loop no longer observes h.ctx.Done() when writeTimeout is enabled. Exis... Pull #1212 dunglas github
84.1
24776707551 fix/bolt-unknown-lasteventid fix(bolt): stop leaking post-subscribe events into Last-Event-ID header dispatchHistory used to update responseLastEventID with each key the cursor visited BEFORE checking whether that key was already past the subscriber's sequence snapshot. Unde... Pull #1215 dunglas github
83.77
24776163301 fix/graceful-shutdown fix(subscribe): drain SSE subscribers naturally on hub shutdown On hub shutdown (Caddy "stopping" event, pod receiving SIGTERM, graceful config reload), the subscriber select loop no longer observes h.ctx.Done() when writeTimeout is enabled. Exis... Pull #1212 dunglas github
84.1