queue-runner: release machine reservation while copying outputs

This allows for better builder usage when the queue runner is busy. To
avoid running into uncontrollable imbalances between builder/queue
runner, we only release the machine reservation after the local
throttler has found a slot to start copying the outputs for that build.

As opposed to asserting uniqueness to understand resource utilization,
we just switch to using `std::unique_ptr`.
This commit is contained in:
Pierre Bourdon
2024-04-21 01:46:41 +02:00
committed by John Ericson
parent cc4b206d85
commit 143a07bff0
4 changed files with 24 additions and 17 deletions

View File

@@ -16,7 +16,7 @@ void setThreadName(const std::string & name)
}
void State::builder(MachineReservation::ptr reservation)
void State::builder(std::unique_ptr<MachineReservation> reservation)
{
setThreadName("bld~" + std::string(reservation->step->drvPath.to_string()));
@@ -35,22 +35,20 @@ void State::builder(MachineReservation::ptr reservation)
activeSteps_.lock()->erase(activeStep);
});
std::string machine = reservation->machine->storeUri.render();
try {
auto destStore = getDestStore();
res = doBuildStep(destStore, reservation, activeStep);
// Might release the reservation.
res = doBuildStep(destStore, std::move(reservation), activeStep);
} catch (std::exception & e) {
printMsg(lvlError, "uncaught exception building %s on %s: %s",
localStore->printStorePath(reservation->step->drvPath),
reservation->machine->storeUri.render(),
localStore->printStorePath(activeStep->step->drvPath),
machine,
e.what());
}
}
/* Release the machine and wake up the dispatcher. */
assert(reservation.unique());
reservation = 0;
wakeDispatcher();
/* If there was a temporary failure, retry the step after an
exponentially increasing interval. */
Step::ptr step = wstep.lock();
@@ -72,11 +70,11 @@ void State::builder(MachineReservation::ptr reservation)
State::StepResult State::doBuildStep(nix::ref<Store> destStore,
MachineReservation::ptr reservation,
std::unique_ptr<MachineReservation> reservation,
std::shared_ptr<ActiveStep> activeStep)
{
auto & step(reservation->step);
auto & machine(reservation->machine);
auto step(reservation->step);
auto machine(reservation->machine);
{
auto step_(step->state.lock());
@@ -211,7 +209,7 @@ State::StepResult State::doBuildStep(nix::ref<Store> destStore,
try {
/* FIXME: referring builds may have conflicting timeouts. */
buildRemote(destStore, machine, step, buildOptions, result, activeStep, updateStep, narMembers);
buildRemote(destStore, std::move(reservation), machine, step, buildOptions, result, activeStep, updateStep, narMembers);
} catch (Error & e) {
if (activeStep->state_.lock()->cancelled) {
printInfo("marking step %d of build %d as cancelled", stepNr, buildId);