worker: process timeouts first, and exclusively

check goals for timeouts first, and their activity fds only if no
timeout has occurred. checking for timeouts *after* activity sets
us up for assertion failures by running multiple build completion
notifiers, the first of which will kill/reap the the goal process
and consuming the Pid instance. when the second notifier attempts
to do the same it will core dump with an assertion failure in Pid
and take down not only the single goal, but the entire daemon and
all goals it was building. luckily this is rare in practice since
it requires a build to both finish and time out at the same time.

writing a test for this is not feasible due to how much it relies
on scheduling to actually trigger the underlying bug, but on idle
machines it can usually be triggered by running multiple sleeping
builds with timeout set to the sleep duration and `--keep-going`:

    nix-build --timeout 10 --builders '' --keep-going -E '
      with import <nixpkgs> {};
      builtins.genList
        (i: runCommand "foo-${toString i}" {} "sleep 10")
        100
    '

Change-Id: I394d36b2e5ffb909cf8a19977d569bbdb71cb67b
This commit is contained in:
eldritch horrors 2024-05-26 17:45:36 +02:00
parent 742c62a6eb
commit ed6b3165ea

View file

@ -429,6 +429,28 @@ void Worker::waitForInput()
GoalPtr goal = j->goal.lock(); GoalPtr goal = j->goal.lock();
assert(goal); assert(goal);
if (goal->exitCode == Goal::ecBusy &&
0 != settings.maxSilentTime &&
j->respectTimeouts &&
after - j->lastOutput >= std::chrono::seconds(settings.maxSilentTime))
{
goal->timedOut(Error(
"%1% timed out after %2% seconds of silence",
goal->getName(), settings.maxSilentTime));
continue;
}
else if (goal->exitCode == Goal::ecBusy &&
0 != settings.buildTimeout &&
j->respectTimeouts &&
after - j->timeStarted >= std::chrono::seconds(settings.buildTimeout))
{
goal->timedOut(Error(
"%1% timed out after %2% seconds",
goal->getName(), settings.buildTimeout));
continue;
}
std::set<int> fds2(j->fds); std::set<int> fds2(j->fds);
std::vector<unsigned char> buffer(4096); std::vector<unsigned char> buffer(4096);
for (auto & k : fds2) { for (auto & k : fds2) {
@ -455,26 +477,6 @@ void Worker::waitForInput()
} }
} }
} }
if (goal->exitCode == Goal::ecBusy &&
0 != settings.maxSilentTime &&
j->respectTimeouts &&
after - j->lastOutput >= std::chrono::seconds(settings.maxSilentTime))
{
goal->timedOut(Error(
"%1% timed out after %2% seconds of silence",
goal->getName(), settings.maxSilentTime));
}
else if (goal->exitCode == Goal::ecBusy &&
0 != settings.buildTimeout &&
j->respectTimeouts &&
after - j->timeStarted >= std::chrono::seconds(settings.buildTimeout))
{
goal->timedOut(Error(
"%1% timed out after %2% seconds",
goal->getName(), settings.buildTimeout));
}
} }
if (!waitingForAWhile.empty() && lastWokenUp + std::chrono::seconds(settings.pollInterval) <= after) { if (!waitingForAWhile.empty() && lastWokenUp + std::chrono::seconds(settings.pollInterval) <= after) {