mirror of
https://github.com/kernelkit/infix.git
synced 2026-07-22 01:13:00 +02:00
Joachim Wiberg
e4203b0c9e
Backport two fixes addressing a critical failure reported by a customer:
an unclean dbus-daemon exit leaves a lingering /run/messagebus.pid, the
daemon then refuse to start, and Finit's restart loop gives up.
0001 service: clean stale pidfile after unclean daemon exit
Drop a daemon-owned (pid:!) pidfile when it still names the
just-reaped PID and that PID is no longer alive.
0002 service: log signal name and core dumps in death message
"by signal: 9" -> "killed by SIGKILL", with ", core dumped"
when applicable. Stronger breadcrumb for sudden deaths.
Signed-off-by: Joachim Wiberg <troglobit@gmail.com>
87 lines
2.6 KiB
Diff
87 lines
2.6 KiB
Diff
From 4a53f610cd05c2aba3da770384460f7e66488ff5 Mon Sep 17 00:00:00 2001
|
|
From: Joachim Wiberg <troglobit@gmail.com>
|
|
Date: Mon, 11 May 2026 13:55:11 +0200
|
|
Subject: [PATCH 1/2] service: clean stale pidfile after unclean daemon exit
|
|
Organization: Wires
|
|
|
|
With `pid:!/path` Finit does not manage the file -- the daemon
|
|
creates it on start and removes it on graceful exit. If the daemon
|
|
dies before cleanup (SIGKILL, OOM, segfault, exit during startup)
|
|
the file lingers and can block the next instance from starting,
|
|
e.g. dbus-daemon refuses with EEXIST and the restart loop fails.
|
|
|
|
Remove the file when it still names the just-reaped PID and that
|
|
PID is no longer alive (the liveness check guards against reuse).
|
|
Called from service_cleanup(), and from service_monitor()'s
|
|
forking+starting branch where cleanup was previously skipped.
|
|
|
|
Signed-off-by: Joachim Wiberg <troglobit@gmail.com>
|
|
---
|
|
src/service.c | 36 +++++++++++++++++++++++++++++++++++-
|
|
1 file changed, 35 insertions(+), 1 deletion(-)
|
|
|
|
diff --git a/src/service.c b/src/service.c
|
|
index 7ed4fceb..e930c4fd 100644
|
|
--- a/src/service.c
|
|
+++ b/src/service.c
|
|
@@ -1120,6 +1120,35 @@ static void service_notify_stop(svc_t *svc)
|
|
}
|
|
}
|
|
|
|
+/*
|
|
+ * Drop a daemon-owned (pid:!) pidfile if it still names the just-reaped
|
|
+ * PID and that PID is gone. The liveness check guards against reuse.
|
|
+ */
|
|
+static void service_clean_pidfile(svc_t *svc, pid_t reaped)
|
|
+{
|
|
+ pid_t pid;
|
|
+ char *fn;
|
|
+
|
|
+ if (reaped <= 1)
|
|
+ return;
|
|
+
|
|
+ fn = pid_file(svc);
|
|
+ if (!fn)
|
|
+ return;
|
|
+
|
|
+ pid = pid_file_read(fn);
|
|
+ if (pid != reaped || pid_alive(pid))
|
|
+ return;
|
|
+
|
|
+ if (remove(fn) && errno != ENOENT) {
|
|
+ logit(LOG_CRIT, "Failed removing stale service %s pidfile %s",
|
|
+ svc_ident(svc, NULL, 0), fn);
|
|
+ return;
|
|
+ }
|
|
+
|
|
+ dbg("Removed stale service %s pidfile %s", svc_ident(svc, NULL, 0), fn);
|
|
+}
|
|
+
|
|
/*
|
|
* Clean up any lingering state from dead/killed services
|
|
*/
|
|
@@ -1137,6 +1166,8 @@ static void service_cleanup(svc_t *svc)
|
|
if (remove(fn) && errno != ENOENT)
|
|
logit(LOG_CRIT, "Failed removing service %s pidfile %s",
|
|
svc_ident(svc, NULL, 0), fn);
|
|
+ } else if (svc->pidfile[0] == '!') {
|
|
+ service_clean_pidfile(svc, svc->pid);
|
|
}
|
|
|
|
/*
|
|
@@ -2405,7 +2436,10 @@ void service_monitor(pid_t lost, int status)
|
|
if (svc_is_forking(svc)) {
|
|
/* Likely start script exiting */
|
|
if (svc_is_starting(svc)) {
|
|
- svc->pid = 0; /* Expect no more activity from this one */
|
|
+ /* Daemon died before clearing 'starting'; drop any stale pidfile. */
|
|
+ service_clean_pidfile(svc, lost);
|
|
+ svc->oldpid = lost; /* So service_retry() logs the real PID */
|
|
+ svc->pid = 0; /* Expect no more activity from this one */
|
|
goto cont;
|
|
}
|
|
|
|
--
|
|
2.43.0
|
|
|