| 1 | package host |
| 2 | |
| 3 | import ( |
| 4 | "os" |
| 5 | "path/filepath" |
| 6 | |
| 7 | "github.com/carlosframework/platform/internal/safefile" |
| 8 | ) |
| 9 | |
| 10 | // adoptPassLockFile is the lock every root adoption pass holds for its whole |
| 11 | // run. It sits in the data dir beside the restart marks, so the single |
| 12 | // filesystem the sandboxed agent and the root passes both write is the single |
| 13 | // filesystem their coordination lives on. |
| 14 | const adoptPassLockFile = "adopt-pass.lock" |
| 15 | |
| 16 | func adoptPassLockPath(dd string) string { return filepath.Join(dd, adoptPassLockFile) } |
| 17 | |
| 18 | // lockAdoptPass takes an exclusive flock for the whole of one RunAdopt, in |
| 19 | // EVERY mode. |
| 20 | // |
| 21 | // Four units exec `carlos adopt`: carlos-adopt.service (hourly), |
| 22 | // carlos-stage.service (the agent relaying a stage it is sandboxed out of), |
| 23 | // carlos-restart.service and carlos-config-sync.service — and systemd will |
| 24 | // start them in the same second without a thought. Until the staged-version |
| 25 | // sweep existed that was tolerable: every write a pass performs is a |
| 26 | // temp-file-then-rename or a symlink swap, so two passes doing the same work |
| 27 | // converge on the same answer. |
| 28 | // |
| 29 | // os.RemoveAll is the first operation in a pass that is not idempotent |
| 30 | // against a concurrent one. The window is small and real: the sweep reads the |
| 31 | // live links, a stage pass repoints one at a version the sweep has just |
| 32 | // decided is unpinned, and the sweep deletes the tree that link now names. |
| 33 | // Rename-to-a-trash-dir does not close it either, because PointLive stats the |
| 34 | // target and THEN symlinks — a rename between those two leaves a dangling |
| 35 | // link and a unit that will not start. Serialising the passes does close it, |
| 36 | // so that is what this is. |
| 37 | // |
| 38 | // The lock is a file of its own rather than a flock on something the pass |
| 39 | // also writes: a flock lives on an inode, and every state file a pass owns is |
| 40 | // replaced by rename, which would leave the winner holding a lock on a file |
| 41 | // that is no longer the state (lockRestartMarks makes the same choice for the |
| 42 | // same reason). |
| 43 | // |
| 44 | // Blocking, like lockRestartMarks: a relayed stage that waits out an hourly |
| 45 | // pass is doing the right thing, and a stage is already a request the agent |
| 46 | // re-raises on its next loop. |
| 47 | func lockAdoptPass(dd string) (unlock func(), err error) { |
| 48 | return lockExclusive(adoptPassLockPath(dd)) |
| 49 | } |
| 50 | |
| 51 | // lockExclusive opens (creating) path and takes an exclusive advisory lock on |
| 52 | // it, returning the release. Shared by lockAdoptPass and lockRestartMarks so |
| 53 | // the two locks cannot drift in how they are taken; the ORDER between them is |
| 54 | // fixed by their callers, adopt-pass first, and there is no third holder. |
| 55 | func lockExclusive(path string) (unlock func(), err error) { |
| 56 | f, err := safefile.OpenFile(path, os.O_CREATE|os.O_RDWR, 0o644) |
| 57 | if err != nil { |
| 58 | return nil, err |
| 59 | } |
| 60 | if err := lockFileEx(f); err != nil { |
| 61 | f.Close() |
| 62 | return nil, err |
| 63 | } |
| 64 | return func() { |
| 65 | unlockFile(f) |
| 66 | f.Close() |
| 67 | }, nil |
| 68 | } |
| 69 | |
amadan