2
0
mirror of https://github.com/xcat2/xcat-core.git synced 2026-11-21 18:02:18 +00:00

fix(xcat-core): a worker that dies before release ends the build with SIGPIPE

start_build_worker writes one byte to the gate pipe to release a worker
after the build lock records its group. When the worker dies between the
record and that write, the pipe has no reader and the write raises
SIGPIPE. The default disposition ends the builder at once, so its
cleanup never runs: the Debian package tree is not restored and the lock
is not released.

Ignore SIGPIPE for that one write. The write then fails with EPIPE and
start_build_worker reports the worker as not released, which is the
existing failure path. build_worker_gate_sigpipe.t passes with this
change.

Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
This commit is contained in:
Daniel Hilst
2026-11-06 16:24:46 -03:00
parent 2edfa6bfe6
commit 17b84cfd3a
+6 -1
View File
@@ -797,7 +797,12 @@ sub start_build_worker {
# lock forever, and the gated child has not run, so record nothing.
my $ok = (POSIX::setpgid($pid, $pid) || $!{EACCES}) ? 1 : 0;
$ok = attach_build_workers($pid) if $ok;
$ok = 0 if $ok && (syswrite($gate_w, 'g') // 0) != 1;
if ($ok) {
# A worker that died after the record leaves no reader. SIGPIPE would end the builder
# and skip its cleanup, so let the write fail with EPIPE instead.
local $SIG{PIPE} = 'IGNORE';
$ok = 0 if (syswrite($gate_w, 'g') // 0) != 1;
}
close($gate_w);
# Scalar context gives the pid, so `start_build_worker(...) and next` keeps the meaning of
# `$pm->start and next` whether or not the record succeeded.