1446b66085
wip
2023-11-01 21:37:11 +01:00
2ec2f61904
WIP
2023-10-04 10:04:21 +02:00
d28cce3fba
chore(build): cleanup apt cache after update & install
2023-09-12 17:50:44 +02:00
8ea12591b6
style(bun.dockerfile): add new line at the end of the file
2023-09-12 17:49:55 +02:00
dcd90e4fff
chore(deps): bump bun to 1.0.1
2023-09-12 17:49:19 +02:00
b0affaa936
wip: use bun (1.0) as runtime
2023-09-09 00:16:53 +02:00
453e48c2b5
fix(nextauth): check if originLinkAccount is not undefined
2023-09-07 11:43:05 +02:00
7cdcf52149
feat(helm): set replicaCount to 2 for production
2023-09-04 11:26:40 +02:00
47919e48c2
fix(nextauth): drop custom Keycloak fields before saving in DB
...
Keycloak response has more fields than we expect, therefore we cannot save the response, this commit removes the extra fields and allowsus to save to DB without an issue or a schema change (which is widely proposed on the internet, but I did not want to implement that)
2023-09-04 08:51:42 +02:00
0f7fee21e0
fix(helm): difference between preferred... and required...
...
- preferred (soft) pod (anti) affinity has weights and podAffinityTerm, required does not
- required (hard) pod (anti) affinity has directly array with labelSelector (no weight and no podAffinityTerm)
2023-09-03 14:11:21 +02:00
aa0637e6fc
fix(helm): add missing weight param to podAntiAffinity item
2023-09-03 14:04:33 +02:00
d16eb3d7b9
fix(helm): add default pdb.minAvailable to values.yaml
2023-09-03 12:02:44 +02:00
779c5aae0e
ci: fix jobs always running when previous stage failed
...
- future note: do not add `when: always`, it runs literarly always - not a surprise, but actually is
2023-09-03 11:48:02 +02:00
bdb17ba397
fix(helm): .Values.replicaCount type for comparison
2023-09-03 11:24:26 +02:00
e54c69f216
fix(nextauth): eslint errors
2023-09-03 11:24:08 +02:00
60d455c938
feat(helm): add podAntiAffinity (soft), run instances across nodes
2023-09-03 10:53:25 +02:00
e66c493366
feat(helm): add PodDisruptionBudget and run staging in two replicas
2023-09-03 10:52:39 +02:00
377e9e9570
feat(nextauth): properly handle signout
...
When signing out, by default keycloak session is not terminated, this fixes it and sends a request to keycloak instance to terminate the keycloak session.
2023-09-03 10:50:40 +02:00
2a7b833b2e
refactor(nextauth): change provider from Google to Keycloak
...
- using self-hosted keycloak as SSO `https://sso.mareshq.com `
2023-09-02 17:37:45 +02:00
083fe4565d
ci: switch to github hosted toolkit image
2023-07-11 10:06:57 +02:00
f5ad99c4dd
ci: add missing SKIP_ENV_VALIDATION=1 to 'lint next.js' job (before env var for npm run lint)
2023-07-05 17:13:33 +02:00
e78c412637
ci: add lint stage to lint next.js app and helm chart
2023-07-05 17:12:28 +02:00
a976fcdce0
refactor(helm): increase job .spec.activeDeadlineSeconds to 5 minutes
2023-07-05 17:10:09 +02:00
b88aec0a3d
refactor(helm): x2 cpu resources for jobs (to 200mCPU)
2023-07-05 17:09:02 +02:00
4c663dfe52
refactor(helm): x2 job (migration, seed) resources (to 512Mi RAM)
2023-07-05 17:04:35 +02:00
b479b98ef3
refactor(helm): x2 job (migration, seed) resources
2023-07-05 17:01:18 +02:00
ccdc5fb9b6
feat(helm): add resource requests and limits to jobs, adjust resources for deployment
2023-07-05 16:56:05 +02:00
385fa9998b
ci: rename jobs to use spaces instead of colon
2023-07-04 21:20:55 +02:00
595fc73364
ci: add deploy dry run and deploy jobs to production
2023-07-04 21:20:46 +02:00
5e46503a3a
feat(helm): Deployment: specify upgrade strategy (RollingUpdate)
2023-07-04 21:07:58 +02:00
6428d1e346
fix(helm): remove Ingress cert-manager.io/issuer annotation (.../cluster-issuer is already set)
2023-07-04 21:05:12 +02:00
a509c09d75
fix(helm): add Deployment's missing .spec.template.metadata.labels
2023-07-04 20:48:08 +02:00
800f65a5e6
refactor(helm): do not specify Job .spec.selector (bast practice)
...
See: https://kubernetes.io/docs/concepts/workloads/controllers/job/#specifying-your-own-pod-selector
2023-07-04 20:44:25 +02:00
d09c18f654
feat(helm): add and unify labels
2023-07-04 20:40:54 +02:00
060e5a4f14
fix(helm): seed job esbuild output file path to /app/cache/seed.cjs
2023-07-04 19:31:28 +02:00
22ebadf393
refactor(helm): move seed job emptyDir mount to /app/cache (from /cache)
2023-07-04 19:28:34 +02:00
b73d9e5f48
refactor(helm): add emptyDir volume to seed job to write seed.cjs file to (/cache/seed.cjs)
2023-07-04 18:59:34 +02:00
9b2200afbd
fix(dockerfile): copy content/ to build image
...
- needed to seed the database
2023-07-04 18:49:52 +02:00
17b341ec7c
fix(dockerfile): copy src/content/training.ts to build image
...
- needed to seed the database
2023-07-04 18:48:34 +02:00
9b6bc1ba5d
chore: ignore seed.cjs
2023-07-04 18:48:12 +02:00
7ae10b4ecc
refactor: move lib/* to src/
2023-07-04 18:47:05 +02:00
2745bb32c7
refactor(package.json): replace tsx with esbuild for seeding database
2023-07-04 18:37:42 +02:00
f6558172b0
refactor(helm): run seed job with esbuild
...
See: https://github.com/prisma/prisma/issues/7053#issuecomment-851589268
2023-07-04 18:29:51 +02:00
29b5c1fcae
fix(helm): set NODE_ENV to 'production' for staging
2023-07-04 18:19:19 +02:00
94754910c2
fix(dockerfile): copy ./src/server/db.ts and ./src/env.mjs for 'npx prisma db seed'
2023-07-04 18:15:14 +02:00
6e74ef1810
fix(helm): typo
2023-07-04 13:17:26 +02:00
d914213c08
chore(build): copy tsconfig.json to build - required for db seeding
2023-07-04 13:14:02 +02:00
86d3320673
refactor(helm): job-db-seed run npx prisma db seed
2023-07-04 13:13:40 +02:00
ada27c7284
refactor(helm): job-db-seed: run as shell, install tsx, prisma, run directly npm tsx prisma/seed.ts
2023-07-04 13:07:17 +02:00
d45defffe5
revert: "refactor(deps): make tsx dependency (not devDependency anymore)"
...
This reverts commit 5ad56955da
2023-07-04 13:03:09 +02:00
