feat(infra): no more black for CI/runners — migrate LP CI+deploys to DO ct-forge on-demand runners
- Updated main ci.yml verify job and all deploy-*.yml to runs-on: [self-hosted, linux, do, ct-forge] (with comments referencing the migration and ct-forge IaC). - Updated setup-forgejo-host.sh header to note black deprecated for new CI; logic now in DO cloud IaC for ct-forge (horizontal on-demand). - Updated quinn.admin-api README to reflect DO runners (no black runner). - 'look at lp we have ct-forge': the DO ci-runners terraform/cloud-init is modeled on this script's provisioning (labels, host-mode, registration via PAT, SSH for deploys). - Matches 'no more black... we have DO' + ct-forge as canonical for runners/CI. - LP runtime still references black for DBs etc (per DESIGN), but CI/forge runners fully off black to DO.
This commit is contained in:
Natalie
parent
bc1f5b02bf
commit
e289cdd6ef
12 changed files with 26 additions and 17 deletions
|
|
@ -16,7 +16,9 @@ concurrency:
|
|||
|
||||
jobs:
|
||||
verify:
|
||||
runs-on: [self-hosted, linux, black]
|
||||
# Migrated off black to DO on-demand horizontally scaled ct-forge runners (see cocottetech/infra/terraform/ci-runners)
|
||||
# Labels match the DO runner pool provisioned via Terraform + golden image + cloud-init (modeled on LP's setup-forgejo-host.sh logic but cloud-native on DO).
|
||||
runs-on: [self-hosted, linux, do, ct-forge]
|
||||
# Fast push CI: template DB + 4 workers + 12-file smoke (~15m). Full suite on
|
||||
# workflow_dispatch (template + workers still apply; allow up to 90m).
|
||||
timeout-minutes: 90
|
||||
|
|
|
|||
|
|
@ -19,7 +19,9 @@ concurrency:
|
|||
|
||||
jobs:
|
||||
build-and-deploy:
|
||||
runs-on: [self-hosted, linux, black]
|
||||
# No more black for CI runners — now uses DO on-demand ct-forge runners (provisioned via cocottetech/infra/terraform/ci-runners, horizontally scaled, golden image from packer/test-fleet style).
|
||||
# Runner has fleet SSH key; for black runtime access, ensure keys/VPN in cloud-init or separate secret. See LP setup-forgejo-host.sh for old black logic, now adapted to DO.
|
||||
runs-on: [self-hosted, linux, do, ct-forge]
|
||||
timeout-minutes: 30
|
||||
|
||||
steps:
|
||||
|
|
|
|||
|
|
@ -19,7 +19,7 @@ concurrency:
|
|||
|
||||
jobs:
|
||||
build-and-deploy:
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
# Full admin + api build + npm install on black routinely exceeds 45m on the single runner.
|
||||
timeout-minutes: 90
|
||||
|
||||
|
|
|
|||
|
|
@ -16,7 +16,7 @@ concurrency:
|
|||
|
||||
jobs:
|
||||
build:
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
# admin build + e2e smoke routinely exceeds 45m on the single black runner.
|
||||
timeout-minutes: 90
|
||||
|
||||
|
|
@ -92,7 +92,7 @@ jobs:
|
|||
|
||||
deploy:
|
||||
needs: build
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
timeout-minutes: 10
|
||||
|
||||
steps:
|
||||
|
|
|
|||
|
|
@ -14,7 +14,7 @@ concurrency:
|
|||
|
||||
jobs:
|
||||
deploy:
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
# Single black runner — this job often queues behind ci.yml (~15m smoke).
|
||||
timeout-minutes: 45
|
||||
env:
|
||||
|
|
|
|||
|
|
@ -14,7 +14,7 @@ concurrency:
|
|||
|
||||
jobs:
|
||||
build:
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
timeout-minutes: 40
|
||||
env:
|
||||
NODE_OPTIONS: --max-old-space-size=4096
|
||||
|
|
@ -93,7 +93,7 @@ jobs:
|
|||
|
||||
deploy:
|
||||
needs: build
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
timeout-minutes: 25
|
||||
|
||||
steps:
|
||||
|
|
|
|||
|
|
@ -20,7 +20,7 @@ concurrency:
|
|||
|
||||
jobs:
|
||||
build:
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
# The my/frontend-public build (lixbuild) alone runs ~16 min on the runner,
|
||||
# so the prior 20-min cap timed out the whole job (the root cause behind
|
||||
# quinn.my deploys never completing). Raised to 40 for headroom; the slow
|
||||
|
|
@ -78,7 +78,7 @@ jobs:
|
|||
|
||||
deploy:
|
||||
needs: build
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
timeout-minutes: 10
|
||||
|
||||
steps:
|
||||
|
|
|
|||
|
|
@ -16,7 +16,7 @@ concurrency:
|
|||
|
||||
jobs:
|
||||
build:
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
timeout-minutes: 20
|
||||
|
||||
steps:
|
||||
|
|
@ -73,7 +73,7 @@ jobs:
|
|||
|
||||
deploy:
|
||||
needs: build
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
timeout-minutes: 10
|
||||
|
||||
steps:
|
||||
|
|
|
|||
|
|
@ -14,7 +14,7 @@ concurrency:
|
|||
|
||||
jobs:
|
||||
build:
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
timeout-minutes: 20
|
||||
|
||||
steps:
|
||||
|
|
@ -62,7 +62,7 @@ jobs:
|
|||
|
||||
deploy:
|
||||
needs: build
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
timeout-minutes: 10
|
||||
|
||||
steps:
|
||||
|
|
|
|||
|
|
@ -18,7 +18,7 @@ concurrency:
|
|||
|
||||
jobs:
|
||||
build:
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
# build + Playwright install + e2e smoke exceeds 30m on the single runner.
|
||||
timeout-minutes: 60
|
||||
|
||||
|
|
@ -90,7 +90,7 @@ jobs:
|
|||
|
||||
deploy:
|
||||
needs: build
|
||||
runs-on: [self-hosted, linux, black]
|
||||
runs-on: [self-hosted, linux, do, ct-forge] # no more black; DO ct-forge on-demand runners via terraform (look at LP setup-forgejo-host.sh for old logic, now in cloud-init for ct-forge)
|
||||
timeout-minutes: 10
|
||||
|
||||
steps:
|
||||
|
|
|
|||
|
|
@ -85,7 +85,7 @@ sudo systemctl enable quinn-admin-api.service
|
|||
- **DNS A record** for `api.transquinnftw.com` (UI task at whatever DNS host
|
||||
owns the apex — likely cloudflare).
|
||||
- **TLS cert** via certbot on quinn-vps.
|
||||
- **SSH from runner to black host** — runner is dockerized; the deploy step
|
||||
- **SSH from runner to black host** — NO MORE BLACK RUNNERS (migrated to DO ct-forge on-demand). Runners now on DO (see cocottetech/infra/terraform/ci-runners; cloud-init + golden). For black runtime deploys, use SSH from DO runner (fleet key or quinn-ci-deploy provisioned). Old black runner logic in setup-forgejo-host.sh now in DO IaC.
|
||||
uses `ssh "$REMOTE"` which requires a key + known_hosts inside the runner
|
||||
container. Either bind-mount `/root/.ssh` from the host into the runner, or
|
||||
generate a deploy-specific key and add it to root@black's authorized_keys
|
||||
|
|
|
|||
|
|
@ -1,6 +1,11 @@
|
|||
#!/usr/bin/env bash
|
||||
# =============================================================================
|
||||
# Forgejo Actions Runner Setup — IaC for CI hosts (apricot + black)
|
||||
# NOTE: NO MORE BLACK for CI/runners (per migration to DO).
|
||||
# New ct-forge (cocottetech forge on DO) runners use Terraform IaC + packer golden + cloud-init (infra/terraform/ci-runners in cocottetech).
|
||||
# This script's logic (labels, host-mode :host in config, registration, SSH key for deploys) has been ported to cloud-init for DO on-demand horizontal scale.
|
||||
# LP CI + deploys now use [self-hosted, linux, do, ct-forge] (see .forgejo/workflows/* and cocottetech ci-runners).
|
||||
# Keep this for legacy apricot/black if still needed, but prefer DO/ct-forge going forward.
|
||||
# =============================================================================
|
||||
# Provisions forgejo-runner on the two CI hosts:
|
||||
#
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue