Chris Lu c2668fbc64 fix(volume): make tier-down crash-safe and serve from local (Rust) (#10113) ...

* fix(volume): fsync .vif and downloaded tier .dat (Rust)

save_volume_info wrote the .vif with a plain write and no fsync, and the
tier download never synced the .dat it wrote. Either could be lost on a
crash before the tier-down path acts on them. fsync both, matching the Go
volume server's util.WriteFile and DownloadFile.

* fix(volume): swap to local before deleting remote on tier-down (Rust)

The tier-down path deleted the shared remote object before trimming the
.vif, so a crash in between left the volume's .vif pointing at a deleted
object. It also dropped the remote backend only on the delete path and
never opened the downloaded local .dat, so reads broke until reload and a
keep-remote download kept serving from the slow remote object.

Trim the .vif and swap to the local .dat on both paths, bracketed by
directory fsyncs, before removing the remote object; gate only the object
removal on keep_remote_dat_file. Matches the Go volume server's crash-safe
ordering.

2026-06-25 12:29:21 -07:00

..

endpoint_guard.rs

volume: validate remote S3 endpoints in FetchAndWriteNeedle (Rust) (#10001)

2026-06-17 00:56:13 -07:00

mod.rs

volume: validate remote S3 endpoints in FetchAndWriteNeedle (Rust) (#10001)

2026-06-17 00:56:13 -07:00

s3_tier.rs

fix(volume): make tier-down crash-safe and serve from local (Rust) (#10113)

2026-06-25 12:29:21 -07:00

s3.rs

Rust volume server implementation with CI (#8539)

2026-03-26 17:24:35 -07:00