Sign up
View stargazersSign up

Issues

Milestone
MilestonesMilestonesNew IssueNew
Operator papercuts: startCommand/entrypoint, .internal DNS race, dead-server severity, flag delivery, volume admission, skew banner#3411Opened18d agobyDDvorinka•severity:low
server set: a deploy through another Entry or Cloud can still plan from old Machine settings#3301Opened19d agobyNNick
N
Volume data protection: snapshots, mirrors and offsite backups, with restore as a new Volume#3128Opened20d agobyNNick
Copies offers Remove mirror while the placement gate refuses every save#2901Opened21d agobyNNick•severity:low
Copies guesses the first unheard Server is the primary when no root answers#2817Opened21d agobyNNick•severity:medium
Volume Runs: a landed pull waits out a full backoff before the run finishes it#2815Opened21d agobyNNick•severity:low
Relay outage went undetected and couldn't be recovered: add /ping alerting and a second relay off Railway#2813Opened21d agobyNNick•severity:medium
Cloud management and enrollment depend on a single relay (relay.ployz.dev); a relay outage stops all of it#2812Opened21d agobyNNick•severity:high
Dashboard: surface host health on the projects/servers overview#2811Opened21d agobyDDvorinka•severity:low
Canvas: edges that explain themselves#2809Opened21d agobyDDvorinka•severity:low
Canvas: group frames for organizing services#2808Opened21d agobyDDvorinka•severity:low
Canvas: keyboard layer, minimap and interaction hint bar#2807Opened21d agobyDDvorinka•severity:low
Copies offers Make primary onto a Server without the copy floor, which the Deploy refuses#2802Opened21d agobyNNick•severity:low
Copies offers Make primary on a Volume no Service mounts, then the Deploy refuses#2798Opened21d agobyNNick•severity:low
Deploy settles a staged primary move as held when its target and the writer are both silent#2796Opened21d agobyNNick•severity:medium
After a partition heals, the CLI's ssh ControlMaster keeps a dead connection for ~1 minute#2795Opened21d agobyNNick•severity:medium
Replacing the Ingress Proxy container drops HTTPS on every Machine for about 1 s#2793Opened21d agobyNNick•severity:medium
Daemon records neither the Volume cut's stop nor a requested removal, so status reads both as a copy not serving#2792Opened21d agobyNNick•severity:low
Status hints drop --context and --connect, so a hint after `ployz status --context prod` reaches the default context#2791Opened21d agobyNNick•severity:medium
Status doesn't flag a route to a Service's removed last copy#2782Opened21d agobyNNick•severity:low
Tidy the Caddy watcher's ingress-role read, local-ingress filter and apply-result handling#2774Opened21d agobyNNick•severity:low
An open Watch or stalled management client makes every daemon stop wait the full 5 s and drop the endpoint ungracefully#2773Opened21d agobyNNick•severity:low
A start-first redeploy's rotation guards read Corrosion N² times#2770Opened21d agobyNNick•bugseverity:medium
Exact convergence after an R≥2 failover: surplus copy stays until a restart or the next Deploy#2767Opened21d agobyNNick•severity:low
Boot pass judges with specs fetched before a concurrent in-place Deploy#2764Opened21d agobyNNick•severity:medium
25 of 169 Issues