fe4439b95f
fix(ssh): switch CI SSH to permissive mode (no known_hosts)
...
Root cause: ssh-keygen -R prints 'Host X not found in F:\CI\State\known_hosts'
to stderr when the IP has no prior entry. In PS5.1 with ErrorActionPreference=Stop,
native stderr is converted to ErrorRecords before 2>$null can discard them,
making the command a terminating error caught by the job's catch block.
This caused all 4 concurrent burn-in jobs to fail after ~15s with:
Error: Host 192.168.79.XXX not found in F:\CI\State\known_hosts
Additionally, 4 concurrent SSH connections writing to the same known_hosts file
with no advisory locking risk file corruption under concurrent load.
Fix: set SshKnownHostsFile default to '' in Invoke-CIJob, Invoke-RemoteBuild,
and Get-BuildArtifacts. When KnownHostsFile='' _Transport.psm1 uses
StrictHostKeyChecking=no + UserKnownHostsFile=NUL (permissive mode), consistent
with Wait-VMReady and Prepare-LinuxBuild2404.ps1. Ephemeral CI VMs on a
local NAT network have no meaningful MITM risk.
Remove the ssh-keygen purge block from Invoke-CIJob.ps1 (no longer needed).
AGENTS.md: document PS5.1 native stderr / 2>$null interaction (error #12 ).
2026-05-12 22:51:20 +02:00
9de86ac289
feat(sprint4): operational hardening, diagnostics, VMDK SHA256, runbook
...
Task 4.1: Invoke-CIJob.ps1 pre-clone disk gate (needs 10 GB free), JobId/ExtraGuestEnv
pattern validation, redacted command logger (already included via _Common.psm1).
Task 4.2: Get-GuestDiagnostics helper invoked from catch block (best-effort, no throw).
Task 4.3: UseSharedCache wired through composite action and orchestrator.
Task 4.4: SSH known_hosts in F:\CI\State\known_hosts for CI jobs; accept-new for templates.
Task 4.5: Deploy-LinuxBuild2404.ps1 verifies Ubuntu VMDK SHA256 before import.
Task 4.6: RUNBOOK.md consolidated template-refresh section: stop runner, backup, boot,
KMS reactivation, Prepare-*, shutdown, snapshot, Validate-DeployState, smoke,
update config, retain prior snapshot 7 days, rollback procedure.
Validate-SetupState.ps1: minor lint fixes and docblock updates.
Wait-VMReady.ps1: Get-BuildArtifacts.ps1: hardening alignment with Sprint 4 objectives.
2026-05-12 21:12:51 +02:00
fee5963a65
feat(linux): update orchestration scripts for dual-OS (Windows/Linux) support
...
Wait-VMReady.ps1:
- Phase 1: vmrun list + path matching instead of getGuestIPAddress (avoids hang)
- Phase 2/3: WinRM (Windows) or TCP:22 + ssh echo (Linux) based on -Transport param
- -Transport SSH auto-selected by Invoke-CIJob when guestOS=ubuntu-64
Invoke-CIJob.ps1:
- Auto-detect GuestOS from clone VMX guestOS field
- Pass -GuestOS Linux to Wait-VMReady, Invoke-RemoteBuild, Get-BuildArtifacts
- Git clone output streamed line-by-line (fix buffered stdout)
Invoke-RemoteBuild.ps1:
- -GuestOS Linux path: SSH+SCP via _Transport.psm1
- Always uses in-VM git clone for Linux builds
Get-BuildArtifacts.ps1:
- -GuestOS Linux path: scp -r ci_build@<ip>:/opt/ci/output/ to host artifact dir
2026-05-11 18:34:16 +02:00
d1f267b67c
fix(artifacts): Retry WinRM connect in Get-BuildArtifacts (wsmprovhost delay)
...
After Invoke-RemoteBuild closes its session, wsmprovhost.exe may take a moment
to release resources. Subsequent New-PSSession in Get-BuildArtifacts fails with
'WSMan service could not launch a host process'.
Add retry loop: 3 attempts, 10s backoff. Enough for WinRM to recover between
the build session and artifact collection session.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-10 22:23:11 +02:00
552eb628b7
feat(sprint4-6): quality, reliability, perf baseline
...
Sprint 4 — qualità codice:
- scripts/_Common.psm1: New-CISessionOption, Resolve-VmrunPath, Invoke-Vmrun
- PSScriptAnalyzerSettings.psd1: project-wide PSSA rules (security + formatting)
- Remove-BuildVM.ps1: SupportsShouldProcess (-WhatIf support)
- Invoke-RemoteBuild.ps1, Get-BuildArtifacts.ps1: use New-CISessionOption from module
- New-BuildVM.ps1: use Resolve-VmrunPath + Invoke-Vmrun from module
Sprint 5 — affidabilità:
- Backup-CITemplate.ps1: stop runner, timestamped VMDK backup, prune old, restart
- Watch-RunnerHealth.ps1: auto-restart act_runner, cooldown 3/h, EventLog + webhook
- Register-CIScheduledTasks.ps1: add CI-RunnerHealth task (every 15 min)
Sprint 6 — perf baseline:
- Set-TemplateSharedFolders.ps1: idempotent VMX editor for NuGet/pip shared folders
- Invoke-RemoteBuild.ps1: -UseSharedCache injects NUGET_PACKAGES + PIP_CACHE_DIR in guest
- Measure-CIBenchmark.ps1: times clone/start/IP/WinRM/destroy, appends to benchmark.jsonl
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-10 20:08:19 +02:00
68cde01c9d
security(sprint1): §1.1/1.3/1.4 — WinRM HTTPS/5986, SHA256 pinning infra, IP regex per-octet
...
§1.4 — Replace loose IP ValidatePattern with per-octet regex in 4 scripts
(Invoke-CIJob, Invoke-RemoteBuild, Get-BuildArtifacts, Wait-VMReady)
§1.1 — Migrate all WinRM connections from HTTP/5985 to HTTPS/5986
- Deploy post-install.ps1: remove AllowUnencrypted=true; self-signed cert + HTTPS listener
already present; firewall rule restricted to 192.168.79.0/24
- Setup-WinBuild2025.ps1: assert AllowUnencrypted=false
- Prepare-WinBuild2025.ps1: preflight uses TCP/5986; Test-WSMan -UseSSL -Port 5986;
New-PSSession -UseSSL -Port 5986; host AllowUnencrypted save/restore removed (not needed for HTTPS)
- Invoke-RemoteBuild, Get-BuildArtifacts: New-PSSession -UseSSL -Port 5986 -Authentication Basic
- Wait-VMReady: New-WSManSessionOption + Test-WSMan -Port 5986 -UseSSL
- Validate-DeployState, Validate-SetupState: Invoke-Command -UseSSL -Port 5986,
AllowUnencrypted check → false; AllowUnencrypted host-side removed from both
- Docs: PLAN, README, ARCHITECTURE, BEST-PRACTICES, HOST-SETUP, WINDOWS-TEMPLATE-SETUP,
LINUX-TEMPLATE-SETUP, TODO updated
§1.3 — SHA256 hash pinning infrastructure
- Assert-Hash function + $script:Hashes hashtable added to Setup-WinBuild2025.ps1
- Assert-Hash called after dotnet-install.ps1, python installer, vs_buildtools.exe downloads
- Hashes initialized to '' (warn-skip); must be filled before next snapshot refresh
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-10 14:46:41 +02:00
f373c0c24b
fix: apply all FIX-TODO items (P0/P1/P2)
...
P0 bugs:
- Invoke-CIJob.ps1: skip --depth 1 when $Commit specified (shallow clone
+ checkout specific commit was silently failing)
- Get-BuildArtifacts.ps1: check .Length -eq 0 not rounded sizeKB (false
fail on artifacts smaller than 50 bytes)
- build-nsis.yml: upgrade upload-artifact v3 -> v4 (v3 deprecated)
P0 security:
- Remove hardcoded default password CIBuild!ChangeMe2026 from all scripts,
README, TODO; prompt Read-Host -AsSecureString at runtime instead
- Setup-TemplateVM.ps1: replace `net user $u $p /add` (password in argv /
audit log 4688) with New-LocalUser + local ConvertTo-SecureString
- Prepare-TemplateSetup.ps1: save and restore WSMan AllowUnencrypted +
TrustedHosts in finally block (was permanently mutating host WinRM config)
- Setup-TemplateVM.ps1: remove duplicate MaxMemoryPerShellMB (winrm set
+ Set-Item both setting same value)
P1 doc (stale Host-Only / VMnet11 era):
- Setup-TemplateVM.ps1: rewrite NETWORK REQUIREMENT block + final steps
(system uses VMnet8 NAT permanently, not VMnet11 Host-Only)
- Invoke-CIJob.ps1 Phase 1 comment: correct network model
- ARCHITECTURE.md: remove Git from toolchain list (not installed); clarify
zip-transfer rationale (no PAT in VM, not network isolation)
- BEST-PRACTICES.md: rewrite Step 8 as NAT topology verification
P1 minor:
- Invoke-RemoteBuild.ps1: remove wasted first New-Item (created then
immediately removed and recreated)
- Wait-VMReady.ps1: remove unused $elapsed; simplify try/catch around
native command (exit codes don't throw)
- Install-Runner.ps1: add deprecation notice pointing to Setup-Host.ps1
P2 operational:
- Add scripts/Cleanup-OrphanedBuildVMs.ps1 with -WhatIf support
- Get-BuildArtifacts.ps1 -IncludeLogs: add -Recurse (msbuild logs in subdir)
- Add gitea/workflows/lint.yml (PSScriptAnalyzer on .ps1 push/PR)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-09 00:39:52 +02:00
dd238121b3
chore: initial commit local CI/CD system (e2e verified, production-ready)
...
Sistema CI locale basato su VMware Workstation + Gitea Actions + act_runner.
Testato e2e con nsis-plugin-nsinnounp (MSBuild + Python, 4 configurazioni parallele).
- scripts/: Invoke-CIJob, Invoke-RemoteBuild, New/Remove-BuildVM, Wait-VMReady, Get-BuildArtifacts
- runner/: act_runner config (windows-build label, capacity 4)
- gitea/workflows/: build-nsis.yml (template per progetti MSBuild/Python)
- template/: script di provisioning template VM (VS BuildTools 2026, .NET SDK 10, Python 3.13)
- docs/: ARCHITECTURE, CI-FLOW, OPTIMIZATION, BEST-PRACTICES, Setup-GiteaSSH
Verificato: e2e-009 SUCCESS in 02:09, cleanup automatico VM confermato.
2026-05-08 23:25:50 +02:00
Simone