[magma-beef]

Summary

Feature — git-mirror cleanup runs a best-effort fstrim on the mirror sticky disk right before unmounting a disk it is about to commit, for at most the number of seconds the VM agent sends. Nothing changes until a host is configured: the agent sends 0 today, and the hypervisor does not negotiate discard yet.

Start here: src/blacksmith-cache.ts (fstrimTimeoutSecsFrom, the trim in cleanup); src/state-helper.ts, src/git-source-provider.ts, src/main.ts only carry the budget from mount to cleanup. dist/ is generated.

Why

The git-mirror sticky disk is an RBD image; packs that git gc/maintenance deletes stay allocated in Ceph until the guest discards them. Same change as useblacksmith/stickydisk#82 for the bare action and useblacksmith/setup-docker-builder#143 for buildkit.

What

• On cleanup, when the run will commit the mirror (shouldCommit) and the host enabled the trim, run timeout <secs> sudo fstrim -v -- <mount> after sync/maintenance and before umount. Any exit code or exception only logs ([git-mirror] fstrim ...: <output> | timed out after Ns | failed: ...).
• The budget comes from the VM agent: GetStickyDiskResponse.fstrim_timeout_secs (FastActions/fa#6600, live config guest_fstrim_timeout_secs), saved in action state at mount time. 0/absent = no trim, so this release does nothing until the agent sends a budget — the rollout and its tuning are agent-side.
• fstrimTimeoutSecsFrom() reads the typed field when the bundled proto package has it, else unknown field 8 via listUnknownFields() (protobuf-es keeps unknown binary fields; the transport is gRPC), so this version keeps working once the agent starts sending the field without a package bump.

Evidence

Check	Result
npm run lint, npm run format-check, npm run build	clean; dist/ committed
npm test	199 pass; new tests for parseFstrimTimeoutSecs, fstrimTimeoutSecsFrom, decodeVarint
CI	build, check-dist, test (ubuntu/macos/windows), git-mirror e2e legs, CodeQL green

Risks

Verdict: Ship. With the agent sending 0 (current default) cleanup only reads one state value; the trim path is reachable only after a host is configured.
Worry about: once a host is configured and discard is negotiated, the first trim of a large never-trimmed mirror runs until the budget expires and cleanup is longer by up to that budget; later trims are incremental.
Checked, not worried: every trim failure mode (unsupported discard, missing binary, timeout) logs and continues; the trim happens after maintenance/sync so it describes the committed state; non-commit runs skip it.
If it goes wrong: cleanup logs [git-mirror] fstrim ...: failed|timed out; set guest_fstrim_timeout_secs back to 0 on the host — no action release needed.

Related: useblacksmith/stickydisk#82, useblacksmith/setup-docker-builder#143, FastActions/fa#6600, useblacksmith/web#18629.

Link to Devin session: https://app.devin.ai/sessions/2e17815c306b4c3cb7af4f6bce008a66
Open in Devin Desktop: https://app.devin.ai/desktop/session/2e17815c306b4c3cb7af4f6bce008a66?variant=devin
Requested by: @magma-beef

---

---

^{Need help on this PR? Tag @codesmith-bot with what you need.}

---

^{Need help on this PR? Tag @codesmith-bot with what you need. (Staging)}

[devin-ai-integration]

I'll fix CI failures and address comments from users with write access. I'll skip comments containing "(aside)".

• Disable automatic comment, CI, and merge conflict monitoring

[blacksmith-staging]

Post step trims the mirror before unmount

After, the post step trims the mounted mirror between maintenance and unmount, and a deleted 512 MiB pack's blocks are released from the backing image; before, they stay allocated through commit.

_Merge-base

_{This PR f36ea7dd}

Also checked · 3 more claims, all matched observations

Budget read from agent field 8
No budget or no commit: no trim
fstrim failure only logs

_{Proof on f36ea7dd}

View full Proof report