lfg2025/archy
1
0
Fork 0
Code Issues 7 Pull Requests Actions Packages Projects Releases 33 Wiki Activity

fix(bootstrap): don't nuke podman socket dir during runtime self-heal

Browse Source 
Observed live on .198: heal_podman_state was removing
$XDG_RUNTIME_DIR/podman/ alongside containers/ and libpod/. That dir
holds the systemd-bound podman.sock — the listener systemd creates for
socket-activated podman.service. Removing it broke every libpod HTTP
call from the orchestrator until `systemctl --user restart
podman.socket` ran. Far worse than any wedge it was trying to repair.

Drop podman/ from the cleanup list. The runtime state we actually want
to clean for FM6 (bolt_state.db drift) lives in containers/ and
libpod/ only.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
archipelago
2026-05-01 15:57:15 -04:00
parent 1c0df95f9a
commit bb42180373
1 changed files with 12 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
core/archipelago/src/bootstrap.rs
View File

@@ -94,11 +94,17 @@ enum PodmanHealOutcome {
/// Probe `podman info`. If it succeeds the daemon's runtime state is
/// fine and we return `Healthy` immediately. If it times out, fails to
/// spawn, or returns an "invalid internal status" / "database state"
/// error, the runtime state in `$XDG_RUNTIME_DIR/{containers,libpod,podman}`
/// is likely wedged. We delete it and re-probe — podman rebuilds the
/// runtime state from persistent state under
/// error, the runtime state under `$XDG_RUNTIME_DIR/{containers,libpod}`
/// is likely wedged. We delete those two dirs and re-probe — podman
/// rebuilds runtime state from persistent storage under
/// `$HOME/.local/share/containers/storage/`.
///
/// `$XDG_RUNTIME_DIR/podman/` is **deliberately not touched**: that's
/// where systemd's socket-activated `podman.sock` listener lives. If we
/// removed it, every libpod HTTP call from the orchestrator would fail
/// with "connection refused" until `systemctl --user restart
/// podman.socket` ran — far worse than the wedge we'd be trying to fix.
///
/// Why this is safe at startup:
/// - We run BEFORE the orchestrator starts its reconcile loop, so no
/// archipelago code is currently calling podman.
@@ -120,10 +126,11 @@ async fn heal_podman_state() -> Result<PodmanHealOutcome> {
if probe_podman_ok().await {
return Ok(PodmanHealOutcome::Healthy);
}
// Wedged. Clean runtime state and try again.
// Wedged. Clean runtime state and try again. Note: `podman/` is
// intentionally absent from this list — see fn docstring.
let xdg = std::env::var("XDG_RUNTIME_DIR")
.context("XDG_RUNTIME_DIR not set; can't locate podman runtime state to clean")?;
for sub in &["containers", "libpod", "podman"] {
for sub in &["containers", "libpod"] {
let path = PathBuf::from(&xdg).join(sub);
match fs::remove_dir_all(&path).await {
Ok(()) => debug!(path = %path.display(), "removed podman runtime state dir"),