immich-app/yucca
1
0
Fork 0
mirror of https://github.com/immich-app/yucca.git synced 2026-09-30 13:33:00 +08:00
Code Issues Packages Projects Releases Wiki Activity
553 Commits 66 Branches 63 Tags
79c914c3757018545909a330865e0cee2e8dc772
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Antoine Lecompte 79c914c375 chore: remove the unused restic-api reference implementation (#684) 2026-09-15 13:08:31 +00:00
Antoine Lecompte 29a0b653e9 fix(ci): de-flake the e2e job (loop-device image, fail-fast wait, diagnostics, hook timeout) (#621) 2026-09-14 23:36:27 +01:00
Antoine Lecompteandizzy 0bfca45a5b fix(yucca sdk): surface the state database path and guard master key generation (#653) 
Co-authored-by: izzy <me@insrt.uk>
 2026-09-14 16:41:42 +01:00
Antoine Lecompte 0b15733005 fix(michael): drain in-flight restic requests before closing the listener on rollout (#660) 2026-09-14 08:30:44 -04:00
Antoine Lecompteandizzy a744273730 fix(yucca sdk): report interrupted backup runs to the backend on restart (#651) 
Co-authored-by: izzy <me@insrt.uk>
 2026-09-14 13:05:36 +01:00
Antoine Lecompte dcf3065fee fix(michael): answer GET of a missing blob or config with 404, not 500 (#650) 2026-09-11 08:49:06 -04:00
Antoine Lecompte 3e213f0891 fix(o11y): show traffic rate in bits per second on all dashboards (#639) 2026-09-10 10:50:05 -04:00
Antoine Lecompte 85c936abc6 feat(bot): add dashboard (#622) 2026-09-02 14:34:46 +00:00
Antoine Lecompte 56a1b18061 feat(cnpg): enable quorum commit on the real database clusters (#613) 2026-09-01 09:57:18 -04:00
Antoine Lecompte ec696ed762 fix(flux): make the database health check actually apply (#614) 2026-09-01 09:57:06 -04:00
Antoine Lecompte 2044b5095e fix(flux): gate the database kustomization on a serving primary (#610) 2026-09-01 09:26:14 -04:00
Antoine Lecompte f1640326b9 feat(infra): activate the futo-backups-bot prod creds (#601) 2026-09-01 09:00:41 -04:00
Antoine Lecompte 7750b71620 fix(futo-backups-bot): ignore interactions from other guilds (#602) 2026-08-31 15:42:57 -07:00
Antoine Lecompte bb0cd74bee feat(columbo): give the agent fixed fleet-health probes and a bigger tool budget (#600) 2026-08-31 16:10:02 -04:00
Antoine Lecompte 8e9cd64c16 feat(columbo): retry model calls with per-attempt deadlines (#598) 2026-08-31 11:28:12 -07:00
Antoine Lecompte ce32472f89 fix(michael): log the resolved route plus op and blob_type on access lines (#597) 2026-08-31 08:38:17 -04:00
Antoine Lecompte 500107b5ce feat(columbo): triage-refusal staff notes + per-user telemetry catalog (#596) 2026-08-31 08:38:05 -04:00
Antoine Lecompte b3aa57dc5d feat(columbo): per-investigation audit log, tool-error recovery, pipe-aware log scoping (#586) 2026-08-28 13:57:28 -04:00
Antoine Lecompte 88267dc1f7 feat(columbo): add columbo (#585) 2026-08-28 13:26:11 -04:00
Antoine Lecompte 7988ac3487 fix(futo-backups-bot): fall back to a generic agent name when the profile read is forbidden (#584) 2026-08-27 12:51:41 -07:00
Antoine Lecompte 001da241f7 chore(tf): consume the freshdesk-stack webhook credentials in the talos stacks (#582) 2026-08-27 15:33:58 -04:00
Antoine Lecompte 73006bd23b chore(tf): bump slop-place/freshdesk to 0.1.2 for the automation-action passthrough fix (#581) 2026-08-27 15:21:10 -04:00
Antoine Lecompte fa9cfd3891 refactor(tf): move freshdesk ownership into partition-level global/freshdesk stacks (#580) 2026-08-27 14:35:04 -04:00
Antoine Lecompte 64371fbaa9 feat(infra): wire the freshdesk webhook route, network policy and secrets (#578) 2026-08-27 13:09:21 -04:00
Antoine Lecompte 4e56080e0b feat(futo-backups-bot): bidirectional freshdesk sync for support tickets (#577) 2026-08-27 13:09:21 -04:00
Antoine Lecompte 121987ff79 feat(yucca-api): discord ticket to freshdesk mapping table and internal endpoints (#576) 2026-08-27 13:09:20 -04:00
Antoine Lecompte b726229b14 feat(yucca-admin-api): manage and cancel discord beta-invite drops via yuctl (#575) 2026-08-26 20:15:05 +01:00
Antoine Lecompte cca37e0200 refactor(futo-backups-bot): centralize user-facing copy in messages.ts (#574) 2026-08-26 14:18:07 -04:00
Antoine Lecompte 003cb71435 fix(futo-backups-bot): mention the everyone role as a literal @everyone (#573) 2026-08-26 13:54:37 -04:00
Antoine Lecompte 9abf625652 feat(futo-backups-bot): /beta-invite with personal links, claim batches, and optional role mention (#571) 2026-08-26 13:41:04 -04:00
Antoine Lecompte 6a37d2674a feat(web): discord beta-invite redemption on the invite page (#570) 2026-08-26 13:41:04 -04:00
Antoine Lecompte 85f3a59b2a feat(yucca-api): discord closed-beta invite claims and nonced redemption (#569) 2026-08-26 13:41:03 -04:00
Antoine Lecompte 1fbed25c53 feat(infra): activate the claim-flow ids for staging (#568) 2026-08-26 14:10:20 +00:00
Antoine Lecompte f7e60f0d41 feat(futo-backups-bot): /claim-backups-role grants the customer role (#567) 2026-08-26 09:58:48 -04:00
Antoine Lecompte 769572dea8 fix(staging): fix ingress locality in staging (#564) 2026-08-26 08:55:46 -04:00
Antoine Lecompte 2172679a36 feat(futo-backups-bot): /staff-notes links the paired staff thread (#566) 2026-08-26 08:55:33 -04:00
Antoine Lecompte 6f892c8ba6 fix(futo-backups-bot): unbreak the staff-thread spec after the suffix change (#565) 2026-08-26 08:54:57 -04:00
Antoine Lecompte 4442740ad9 feat(futo-backups-bot): sync the stored discord username from interactions (#563) 2026-08-26 08:54:35 -04:00
Antoine Lecompte 3167a48583 feat(yucca-admin-api): manage discord links via admin-api and yuctl (#562) 2026-08-26 08:54:22 -04:00
Antoine Lecompte 290e3d7a65 feat(futo-backups-bot): allow up to three open tickets per user (#561) 2026-08-26 12:31:52 +00:00
Antoine Lecompte 5eaf0f349d feat(futo-backups-bot): grafana per-user link in staff notes (#560) 2026-08-26 12:26:07 +00:00
Antoine Lecompte eb89c13118 fix(admin): wire-in the admin-api like prod (#559) 2026-08-26 12:20:09 +00:00
Antoine Lecompte e61dabe7e3 fix(futo-backups-bot): avoid the privileged thread-member listing (#557) 2026-08-26 08:13:36 -04:00
Antoine Lecompte 717ab17e18 fix(futo-backups-bot): pull staff into the staff-notes thread (#558) 2026-08-26 08:06:06 -04:00
Antoine Lecompte bed87ba085 fix(futo-backups-bot): dedupe the support sticky and survive pin failures (#551) 2026-08-25 19:35:29 +00:00
Antoine Lecompte f1bd9cf33e fix(futo-backups-bot): survive missing applications.commands scope (#550) 2026-08-25 18:55:21 +00:00
Antoine Lecompte 28834042cb feat(infra): activate futo-backups-bot staging creds (#549) 2026-08-25 14:51:36 -04:00
Antoine Lecompte a757e14ed0 feat(infra): force_apply dispatch input for prod/staging (#548) 2026-08-25 11:42:07 -07:00
Antoine Lecompte 5ad08e27d7 feat(infra): deploy futo-backups-bot (#545) 2026-08-25 14:24:51 -04:00
Antoine Lecompte 474879f399 feat(futo-backups-bot): discord support bot (#544) 2026-08-25 14:24:50 -04:00
Antoine Lecompte 93d90e3ecf feat(web): discord link confirmation page (#543) 2026-08-25 14:24:50 -04:00
Antoine Lecompte 7f61914cd8 feat(yucca-api): discord account linking (#542) 2026-08-25 14:24:50 -04:00
Antoine Lecompte 8392f4f379 feat(ceph): move user creation inside terraform (#541) 2026-08-25 06:32:50 -07:00
Antoine Lecompte c78ddee105 fix(all): various fixes (#537) 2026-08-24 13:51:25 +00:00
Antoine Lecompte dd12594761 fix(migrations): re-date again (#535) 2026-08-24 09:25:52 -04:00
Antoine Lecompte 4d3990e7f3 feat(infra): plumb postmark config into staging and prod (#493) 2026-08-24 06:45:00 -04:00
Antoine Lecompte 30166e0733 feat(yucca-admin-api): send invite emails from the allowlist flow (#492) 2026-08-24 06:44:59 -04:00
Antoine Lecompte 1d92b4b0c1 feat(emails): mailpit + mock-postmark local dev stack (#491) 2026-08-24 06:44:59 -04:00
Antoine Lecompte 2fc89b105b feat(common): postmark-backed email repository (#490) 2026-08-24 06:44:59 -04:00
Antoine Lecompte 54c880229b feat(emails): svelte invite email template rendered with the web theme (#489) 2026-08-24 06:44:58 -04:00
Antoine Lecompte 03ea70ea7c feat(o11y): michael resilience panels and alerts (#529) 2026-08-21 12:17:57 -04:00
Antoine Lecompte 6c8092ace8 feat(michael): per-backend circuit breakers with fast-fail load shedding (#528) 2026-08-21 12:17:57 -04:00
Antoine Lecompte bb28f016a3 feat(michael): budgeted cross-backend retries in the S3 pool (#527) 2026-08-21 12:17:56 -04:00
Antoine Lecompte f435601871 fix(michael): converge retried WORM puts when the existing blob matches (#526) 2026-08-21 12:17:56 -04:00
Antoine Lecompte 55488d6818 fix(go): add golangci-lint (#525) 2026-08-21 13:55:25 +00:00
Antoine Lecompte 359a227645 fix(alerts): make junos-exporter ignore one of em (#524) 2026-08-21 13:35:47 +00:00
Antoine Lecompte bbcfa7c295 fix(cnpg): fix backup alerts (#522) 2026-08-21 13:17:57 +00:00
Antoine Lecompte d4554a7a6f feat(o11y): staging observability to o11y staging mesh vmauth over netbird (#517) 2026-08-20 11:54:16 -04:00
Antoine Lecompte 57c790d660 fix(k8s): staging gw via SNI passthrough — mergeGateways is unsupported under GatewayNamespace mode (#516) 2026-08-20 11:51:21 -04:00
Antoine Lecompte a5de2a6f89 chore(netbird): remove the retired k8s operator's service user, token, and secret bootstrap (#515) 2026-08-20 15:47:29 +00:00
Antoine Lecompte 0c2298dc95 refactor(k8s): converge staging and prod onto shared infra/platform components (#514) 2026-08-20 11:21:27 -04:00
Antoine Lecompte a3b42946d5 fix(meta): log x-forwarded for IP address (#512) 2026-08-20 14:34:26 +00:00
Antoine Lecompte 153bf220e2 fix(o11y): field selector on log panels, variable-safe request tracing (#509) 
* fix(o11y): field selector on log panels, variable-safe request tracing

* docs: one-line PR descriptions and commit messages
 2026-08-20 13:59:32 +00:00
Antoine Lecompte 816ce71bd3 fix(migrations): re-date (#510) 2026-08-20 13:48:54 +00:00
Antoine Lecompte d760fb1993 feat(o11y): scope alerts to project=yucca, CNPG dashboard, title convention (#508) 2026-08-20 09:19:45 -04:00
Antoine Lecompte 4eaaaa6e6f fix(net): bgp (#507) 2026-08-20 13:07:35 +00:00
Antoine Lecompte 09f10fcbf5 feat(o11y): per-service log dashboards (#504) 
Six VictoriaLogs dashboards (yucca-api, admin-api, michael, web,
metrics-worker, meta): by-level volume histograms via the hits endpoint,
error streams with top-message tables, per-service request breakdowns
(status, handlers/routes, slowest requests, top users and source networks on
michael, cron heartbeat on metrics-worker, regex status classes on meta),
and full live streams with LogsQL filter and request-id textboxes. Query
types are the plugin's real enum (instant/statsRange/hits); every query was
validated against the live VictoriaLogs. No prometheus datasource variable
on these boards, which keeps the linter's PromQL rule away from LogsQL.
 2026-08-20 12:25:05 +00:00
Antoine Lecompte 045491d437 feat(cnpg): make better (#496) 2026-08-19 15:27:09 -04:00
Antoine Lecompte 569c17ff63 fix(o11y): point alert rules and imported dashboards at the fleet datasource (#502) 
* fix(o11y): pin 5m staleness lookback on alert rule queries

VictoriaMetrics uses an instant query's step as the staleness lookback and
Grafana derives step from intervalMs, clamped to ~15s. Metrics scraped less
often than the step resolve to no-data and noDataState: OK parks the rule at
Normal while its condition is true: the junos-based rules (60s NETCONF
scrape) never fired, with the Colt transit session down. intervalMs: 300000
restores Prometheus-standard staleness for every cadence in the fleet.

* fix(o11y): make slow-cadence alert queries staleness-proof

Grafana sends the rule group's evaluation interval as the instant query's
step and VictoriaMetrics uses step as the staleness lookback; intervalMs on
the query model does not change it. Sources at or above the step's cadence
(junos at 60s, the 5m metering cron) lose that race often enough that a
multi-minute for can never sustain, so the rules sat Normal via
noDataState: OK through a real transit outage. Wrap every such selector in
last_over_time so evaluation is cadence-independent at any step.

* fix(o11y): point alert rules and imported dashboards at the fleet datasource

The default VictoriaMetrics datasource fronts vmauth-self-select, which
serves only the o11y cluster's own series: every rule over yucca, fabric or
ceph data evaluated to NoData and sat Normal through the live Colt transit
outage (confirmed via the Grafana API — all instances Normal (NoData) except
the cert rule, whose only firing labels were cluster=o11y). Pin
datasourceUid: VictoriaMetricsFleet (vmselect, whole fleet) on every rule
query and give the imported dashboards' $datasource variable the house
/^VictoriaMetrics Fleet$/ regex so they stop defaulting to the self-scoped
datasource.
 2026-08-19 19:26:16 +00:00
Antoine Lecompte 552735a527 fix(o11y): make slow-cadence alert queries staleness-proof (#501) 
* fix(o11y): pin 5m staleness lookback on alert rule queries

VictoriaMetrics uses an instant query's step as the staleness lookback and
Grafana derives step from intervalMs, clamped to ~15s. Metrics scraped less
often than the step resolve to no-data and noDataState: OK parks the rule at
Normal while its condition is true: the junos-based rules (60s NETCONF
scrape) never fired, with the Colt transit session down. intervalMs: 300000
restores Prometheus-standard staleness for every cadence in the fleet.

* fix(o11y): make slow-cadence alert queries staleness-proof

Grafana sends the rule group's evaluation interval as the instant query's
step and VictoriaMetrics uses step as the staleness lookback; intervalMs on
the query model does not change it. Sources at or above the step's cadence
(junos at 60s, the 5m metering cron) lose that race often enough that a
multi-minute for can never sustain, so the rules sat Normal via
noDataState: OK through a real transit outage. Wrap every such selector in
last_over_time so evaluation is cadence-independent at any step.
 2026-08-19 18:59:13 +00:00
Antoine Lecompte 0fe43daaf0 fix(o11y): pin 5m staleness lookback on alert rule queries (#499) 
VictoriaMetrics uses an instant query's step as the staleness lookback and
Grafana derives step from intervalMs, clamped to ~15s. Metrics scraped less
often than the step resolve to no-data and noDataState: OK parks the rule at
Normal while its condition is true: the junos-based rules (60s NETCONF
scrape) never fired, with the Colt transit session down. intervalMs: 300000
restores Prometheus-standard staleness for every cadence in the fleet.
 2026-08-19 18:17:46 +00:00
Antoine Lecompte 756e6e080e fix(alerts): fix bad alerts (#497) 2026-08-19 17:43:13 +00:00
Antoine Lecompte 8c6e9e325d chore(o11y): add more dashboard and alerts (#495) 2026-08-19 13:14:25 -04:00
Antoine Lecompte ee43e55a2b feat(net): add colt (#488) 2026-08-19 13:04:17 -04:00
Antoine Lecompte b7bbe2e80e chore(yuctl): restructure and make pretty (#485) 2026-08-18 09:53:59 -04:00
Antoine Lecompte 0e3f47a769 perf(ci): stop the integration and e2e jobs waiting on rook-ceph (#435) 2026-08-07 13:40:51 +01:00
Antoine Lecompte 51472c37eb fix(infra): monitor all the things (#445) 2026-08-05 20:04:26 +00:00
Antoine Lecompteandgreptile-apps[bot] 75c77ecdfa feat(infra): add spegel (#436) 
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
 2026-08-05 20:03:59 +00:00
Antoine Lecompte 02efb46f53 chore(claude): make claude comply (#432) 2026-08-05 12:54:55 -04:00
Antoine Lecompte bf69aee578 fix(yucca-sdk): repair the orchestration-api integration suites (#429) 2026-08-05 12:45:59 -04:00
Antoine Lecompte a4d422709b feat(michael): per-client parallelism and latency metrics (#427) 2026-08-05 12:09:29 -04:00
Antoine Lecompte 72b343ceba feat(michael): attribute traffic to the client's source asn (#428) 2026-08-05 15:13:04 +00:00
Antoine Lecompte a4c4b0c337 fix(yucca-api): format failure (#426) 2026-08-05 13:34:40 +00:00
Antoine Lecompte 8feda20065 feat(web): connections ui (#415) 2026-08-03 18:30:57 +00:00
Antoine Lecompte b427a7045b feat(yucca): per-connection billing rollup (#391) 2026-08-03 14:09:23 -04:00
Antoine Lecompte 55ece2fbaf feat(all): connections (#390) 2026-08-03 14:09:23 -04:00
Antoine Lecompte 94a5df9667 feat(yucca): per-user feature flags (#389) 2026-08-03 14:09:22 -04:00
Antoine Lecompte c606637450 fix(k8s): drop deadlocking topology readyExpr (#401) 2026-07-30 13:01:06 +00:00
Antoine Lecompte 1039337b13 feat(orchestrator): meta discovery and placement (#385) 2026-07-30 08:29:30 -04:00
Antoine Lecompte 7911c3bf11 feat(yuctl): config commands (#384) 2026-07-30 08:29:30 -04:00