2946899504
It now receives notifications about started/finished builds/steps via PostgreSQL. This gets rid of the (substantial) overhead of starting hydra-notify for every event. It also allows other programs (even on other machines) to listen to Hydra notifications.
891 lines
31 KiB
C++
891 lines
31 KiB
C++
#include <iostream>
|
||
#include <thread>
|
||
|
||
#include <sys/types.h>
|
||
#include <sys/stat.h>
|
||
#include <fcntl.h>
|
||
|
||
#include "state.hh"
|
||
#include "build-result.hh"
|
||
#include "store-api.hh"
|
||
#include "remote-store.hh"
|
||
|
||
#include "globals.hh"
|
||
#include "hydra-config.hh"
|
||
#include "json.hh"
|
||
#include "s3-binary-cache-store.hh"
|
||
#include "shared.hh"
|
||
|
||
using namespace nix;
|
||
|
||
|
||
namespace nix {
|
||
|
||
template<> void toJSON<std::atomic<long>>(std::ostream & str, const std::atomic<long> & n) { str << n; }
|
||
template<> void toJSON<std::atomic<unsigned long>>(std::ostream & str, const std::atomic<unsigned long> & n) { str << n; }
|
||
template<> void toJSON<double>(std::ostream & str, const double & n) { str << n; }
|
||
|
||
}
|
||
|
||
|
||
static uint64_t getMemSize()
|
||
{
|
||
auto pages = sysconf(_SC_PHYS_PAGES);
|
||
return pages >= 0 ? pages * sysconf(_SC_PAGESIZE) : 4ULL << 30;
|
||
}
|
||
|
||
|
||
std::string getEnvOrDie(const std::string & key)
|
||
{
|
||
char * value = getenv(key.c_str());
|
||
if (!value) throw Error("environment variable '%s' is not set", key);
|
||
return value;
|
||
}
|
||
|
||
|
||
State::State()
|
||
: config(std::make_unique<::Config>())
|
||
, dbPool(config->getIntOption("max_db_connections", 128))
|
||
, memoryTokens(config->getIntOption("nar_buffer_size", getMemSize() / 2))
|
||
, maxOutputSize(config->getIntOption("max_output_size", 2ULL << 30))
|
||
, maxLogSize(config->getIntOption("max_log_size", 64ULL << 20))
|
||
, uploadLogsToBinaryCache(config->getBoolOption("upload_logs_to_binary_cache", false))
|
||
, rootsDir(config->getStrOption("gc_roots_dir", fmt("%s/gcroots/per-user/%s/hydra-roots", settings.nixStateDir, getEnvOrDie("LOGNAME"))))
|
||
{
|
||
debug("using %d bytes for the NAR buffer", memoryTokens.capacity());
|
||
|
||
hydraData = getEnvOrDie("HYDRA_DATA");
|
||
|
||
logDir = canonPath(hydraData + "/build-logs");
|
||
|
||
/* handle deprecated store specification */
|
||
if (config->getStrOption("store_mode") != "")
|
||
throw Error("store_mode in hydra.conf is deprecated, please use store_uri");
|
||
if (config->getStrOption("binary_cache_dir") != "")
|
||
printMsg(lvlError, "hydra.conf: binary_cache_dir is deprecated and ignored. use store_uri=file:// instead");
|
||
if (config->getStrOption("binary_cache_s3_bucket") != "")
|
||
printMsg(lvlError, "hydra.conf: binary_cache_s3_bucket is deprecated and ignored. use store_uri=s3:// instead");
|
||
if (config->getStrOption("binary_cache_secret_key_file") != "")
|
||
printMsg(lvlError, "hydra.conf: binary_cache_secret_key_file is deprecated and ignored. use store_uri=...?secret-key= instead");
|
||
|
||
createDirs(rootsDir);
|
||
}
|
||
|
||
|
||
nix::MaintainCount<counter> State::startDbUpdate()
|
||
{
|
||
if (nrActiveDbUpdates > 6)
|
||
printError("warning: %d concurrent database updates; PostgreSQL may be stalled", nrActiveDbUpdates.load());
|
||
return MaintainCount<counter>(nrActiveDbUpdates);
|
||
}
|
||
|
||
|
||
ref<Store> State::getDestStore()
|
||
{
|
||
return ref<Store>(_destStore);
|
||
}
|
||
|
||
|
||
void State::parseMachines(const std::string & contents)
|
||
{
|
||
Machines newMachines, oldMachines;
|
||
{
|
||
auto machines_(machines.lock());
|
||
oldMachines = *machines_;
|
||
}
|
||
|
||
for (auto line : tokenizeString<Strings>(contents, "\n")) {
|
||
line = trim(string(line, 0, line.find('#')));
|
||
auto tokens = tokenizeString<std::vector<std::string>>(line);
|
||
if (tokens.size() < 3) continue;
|
||
tokens.resize(8);
|
||
|
||
auto machine = std::make_shared<Machine>();
|
||
machine->sshName = tokens[0];
|
||
machine->systemTypes = tokenizeString<StringSet>(tokens[1], ",");
|
||
machine->sshKey = tokens[2] == "-" ? string("") : tokens[2];
|
||
if (tokens[3] != "")
|
||
string2Int(tokens[3], machine->maxJobs);
|
||
else
|
||
machine->maxJobs = 1;
|
||
machine->speedFactor = atof(tokens[4].c_str());
|
||
if (tokens[5] == "-") tokens[5] = "";
|
||
machine->supportedFeatures = tokenizeString<StringSet>(tokens[5], ",");
|
||
if (tokens[6] == "-") tokens[6] = "";
|
||
machine->mandatoryFeatures = tokenizeString<StringSet>(tokens[6], ",");
|
||
for (auto & f : machine->mandatoryFeatures)
|
||
machine->supportedFeatures.insert(f);
|
||
if (tokens[7] != "" && tokens[7] != "-")
|
||
machine->sshPublicHostKey = base64Decode(tokens[7]);
|
||
|
||
/* Re-use the State object of the previous machine with the
|
||
same name. */
|
||
auto i = oldMachines.find(machine->sshName);
|
||
if (i == oldMachines.end())
|
||
printMsg(lvlChatty, format("adding new machine ‘%1%’") % machine->sshName);
|
||
else
|
||
printMsg(lvlChatty, format("updating machine ‘%1%’") % machine->sshName);
|
||
machine->state = i == oldMachines.end()
|
||
? std::make_shared<Machine::State>()
|
||
: i->second->state;
|
||
newMachines[machine->sshName] = machine;
|
||
}
|
||
|
||
for (auto & m : oldMachines)
|
||
if (newMachines.find(m.first) == newMachines.end()) {
|
||
if (m.second->enabled)
|
||
printMsg(lvlInfo, format("removing machine ‘%1%’") % m.first);
|
||
/* Add a disabled Machine object to make sure stats are
|
||
maintained. */
|
||
auto machine = std::make_shared<Machine>(*(m.second));
|
||
machine->enabled = false;
|
||
newMachines[m.first] = machine;
|
||
}
|
||
|
||
static bool warned = false;
|
||
if (newMachines.empty() && !warned) {
|
||
printError("warning: no build machines are defined");
|
||
warned = true;
|
||
}
|
||
|
||
auto machines_(machines.lock());
|
||
*machines_ = newMachines;
|
||
|
||
wakeDispatcher();
|
||
}
|
||
|
||
|
||
void State::monitorMachinesFile()
|
||
{
|
||
string defaultMachinesFile = "/etc/nix/machines";
|
||
auto machinesFiles = tokenizeString<std::vector<Path>>(
|
||
getEnv("NIX_REMOTE_SYSTEMS", pathExists(defaultMachinesFile) ? defaultMachinesFile : ""), ":");
|
||
|
||
if (machinesFiles.empty()) {
|
||
parseMachines("localhost " +
|
||
(settings.thisSystem == "x86_64-linux" ? "x86_64-linux,i686-linux" : settings.thisSystem.get())
|
||
+ " - " + std::to_string(settings.maxBuildJobs) + " 1");
|
||
return;
|
||
}
|
||
|
||
std::vector<struct stat> fileStats;
|
||
fileStats.resize(machinesFiles.size());
|
||
for (unsigned int n = 0; n < machinesFiles.size(); ++n) {
|
||
auto & st(fileStats[n]);
|
||
st.st_ino = st.st_mtime = 0;
|
||
}
|
||
|
||
auto readMachinesFiles = [&]() {
|
||
|
||
/* Check if any of the machines files changed. */
|
||
bool anyChanged = false;
|
||
for (unsigned int n = 0; n < machinesFiles.size(); ++n) {
|
||
Path machinesFile = machinesFiles[n];
|
||
struct stat st;
|
||
if (stat(machinesFile.c_str(), &st) != 0) {
|
||
if (errno != ENOENT)
|
||
throw SysError(format("getting stats about ‘%1%’") % machinesFile);
|
||
st.st_ino = st.st_mtime = 0;
|
||
}
|
||
auto & old(fileStats[n]);
|
||
if (old.st_ino != st.st_ino || old.st_mtime != st.st_mtime)
|
||
anyChanged = true;
|
||
old = st;
|
||
}
|
||
|
||
if (!anyChanged) return;
|
||
|
||
debug("reloading machines files");
|
||
|
||
string contents;
|
||
for (auto & machinesFile : machinesFiles) {
|
||
try {
|
||
contents += readFile(machinesFile);
|
||
contents += '\n';
|
||
} catch (SysError & e) {
|
||
if (e.errNo != ENOENT) throw;
|
||
}
|
||
}
|
||
|
||
parseMachines(contents);
|
||
};
|
||
|
||
while (true) {
|
||
try {
|
||
readMachinesFiles();
|
||
// FIXME: use inotify.
|
||
sleep(30);
|
||
} catch (std::exception & e) {
|
||
printMsg(lvlError, format("reloading machines file: %1%") % e.what());
|
||
}
|
||
}
|
||
}
|
||
|
||
|
||
void State::clearBusy(Connection & conn, time_t stopTime)
|
||
{
|
||
pqxx::work txn(conn);
|
||
txn.parameterized
|
||
("update BuildSteps set busy = 0, status = $1, stopTime = $2 where busy != 0")
|
||
((int) bsAborted)
|
||
(stopTime, stopTime != 0).exec();
|
||
txn.commit();
|
||
}
|
||
|
||
|
||
unsigned int State::allocBuildStep(pqxx::work & txn, BuildID buildId)
|
||
{
|
||
auto res = txn.parameterized("select max(stepnr) from BuildSteps where build = $1")(buildId).exec();
|
||
return res[0][0].is_null() ? 1 : res[0][0].as<int>() + 1;
|
||
}
|
||
|
||
|
||
unsigned int State::createBuildStep(pqxx::work & txn, time_t startTime, BuildID buildId, Step::ptr step,
|
||
const std::string & machine, BuildStatus status, const std::string & errorMsg, BuildID propagatedFrom)
|
||
{
|
||
restart:
|
||
auto stepNr = allocBuildStep(txn, buildId);
|
||
|
||
auto r = txn.parameterized
|
||
("insert into BuildSteps (build, stepnr, type, drvPath, busy, startTime, system, status, propagatedFrom, errorMsg, stopTime, machine) values ($1, $2, $3, $4, $5, $6, $7, $8, $9, $10, $11, $12) on conflict do nothing")
|
||
(buildId)
|
||
(stepNr)
|
||
(0) // == build
|
||
(step->drvPath)
|
||
(status == bsBusy ? 1 : 0)
|
||
(startTime, startTime != 0)
|
||
(step->drv.platform)
|
||
((int) status, status != bsBusy)
|
||
(propagatedFrom, propagatedFrom != 0)
|
||
(errorMsg, errorMsg != "")
|
||
(startTime, startTime != 0 && status != bsBusy)
|
||
(machine).exec();
|
||
|
||
if (r.affected_rows() == 0) goto restart;
|
||
|
||
for (auto & output : step->drv.outputs)
|
||
txn.parameterized
|
||
("insert into BuildStepOutputs (build, stepnr, name, path) values ($1, $2, $3, $4)")
|
||
(buildId)(stepNr)(output.first)(output.second.path).exec();
|
||
|
||
return stepNr;
|
||
}
|
||
|
||
|
||
void State::updateBuildStep(pqxx::work & txn, BuildID buildId, unsigned int stepNr, StepState stepState)
|
||
{
|
||
if (txn.parameterized
|
||
("update BuildSteps set busy = $1 where build = $2 and stepnr = $3 and busy != 0 and status is null")
|
||
((int) stepState)
|
||
(buildId)
|
||
(stepNr)
|
||
.exec().affected_rows() != 1)
|
||
throw Error("step %d of build %d is in an unexpected state", stepNr, buildId);
|
||
}
|
||
|
||
|
||
void State::finishBuildStep(pqxx::work & txn, const RemoteResult & result,
|
||
BuildID buildId, unsigned int stepNr, const std::string & machine)
|
||
{
|
||
assert(result.startTime);
|
||
assert(result.stopTime);
|
||
txn.parameterized
|
||
("update BuildSteps set busy = 0, status = $1, errorMsg = $4, startTime = $5, stopTime = $6, machine = $7, overhead = $8, timesBuilt = $9, isNonDeterministic = $10 where build = $2 and stepnr = $3")
|
||
((int) result.stepStatus)(buildId)(stepNr)
|
||
(result.errorMsg, result.errorMsg != "")
|
||
(result.startTime)(result.stopTime)
|
||
(machine, machine != "")
|
||
(result.overhead, result.overhead != 0)
|
||
(result.timesBuilt, result.timesBuilt > 0)
|
||
(result.isNonDeterministic, result.timesBuilt > 1)
|
||
.exec();
|
||
assert(result.logFile.find('\'') == std::string::npos);
|
||
txn.exec(fmt("notify step_finished, '%d %d %s'", buildId, stepNr,
|
||
result.logFile.empty() ? "-" : result.logFile));
|
||
}
|
||
|
||
|
||
int State::createSubstitutionStep(pqxx::work & txn, time_t startTime, time_t stopTime,
|
||
Build::ptr build, const Path & drvPath, const string & outputName, const Path & storePath)
|
||
{
|
||
restart:
|
||
auto stepNr = allocBuildStep(txn, build->id);
|
||
|
||
auto r = txn.parameterized
|
||
("insert into BuildSteps (build, stepnr, type, drvPath, busy, status, startTime, stopTime) values ($1, $2, $3, $4, $5, $6, $7, $8) on conflict do nothing")
|
||
(build->id)
|
||
(stepNr)
|
||
(1) // == substitution
|
||
(drvPath)
|
||
(0)
|
||
(0)
|
||
(startTime)
|
||
(stopTime).exec();
|
||
|
||
if (r.affected_rows() == 0) goto restart;
|
||
|
||
txn.parameterized
|
||
("insert into BuildStepOutputs (build, stepnr, name, path) values ($1, $2, $3, $4)")
|
||
(build->id)(stepNr)(outputName)(storePath).exec();
|
||
|
||
return stepNr;
|
||
}
|
||
|
||
|
||
/* Get the steps and unfinished builds that depend on the given step. */
|
||
void getDependents(Step::ptr step, std::set<Build::ptr> & builds, std::set<Step::ptr> & steps)
|
||
{
|
||
std::function<void(Step::ptr)> visit;
|
||
|
||
visit = [&](Step::ptr step) {
|
||
if (steps.count(step)) return;
|
||
steps.insert(step);
|
||
|
||
std::vector<Step::wptr> rdeps;
|
||
|
||
{
|
||
auto step_(step->state.lock());
|
||
|
||
for (auto & build : step_->builds) {
|
||
auto build_ = build.lock();
|
||
if (build_ && !build_->finishedInDB) builds.insert(build_);
|
||
}
|
||
|
||
/* Make a copy of rdeps so that we don't hold the lock for
|
||
very long. */
|
||
rdeps = step_->rdeps;
|
||
}
|
||
|
||
for (auto & rdep : rdeps) {
|
||
auto rdep_ = rdep.lock();
|
||
if (rdep_) visit(rdep_);
|
||
}
|
||
};
|
||
|
||
visit(step);
|
||
}
|
||
|
||
|
||
void visitDependencies(std::function<void(Step::ptr)> visitor, Step::ptr start)
|
||
{
|
||
std::set<Step::ptr> queued;
|
||
std::queue<Step::ptr> todo;
|
||
todo.push(start);
|
||
|
||
while (!todo.empty()) {
|
||
auto step = todo.front();
|
||
todo.pop();
|
||
|
||
visitor(step);
|
||
|
||
auto state(step->state.lock());
|
||
for (auto & dep : state->deps)
|
||
if (queued.find(dep) == queued.end()) {
|
||
queued.insert(dep);
|
||
todo.push(dep);
|
||
}
|
||
}
|
||
}
|
||
|
||
|
||
void State::markSucceededBuild(pqxx::work & txn, Build::ptr build,
|
||
const BuildOutput & res, bool isCachedBuild, time_t startTime, time_t stopTime)
|
||
{
|
||
if (build->finishedInDB) return;
|
||
|
||
if (txn.parameterized("select 1 from Builds where id = $1 and finished = 0")(build->id).exec().empty()) return;
|
||
|
||
txn.parameterized
|
||
("update Builds set finished = 1, buildStatus = $2, startTime = $3, stopTime = $4, size = $5, closureSize = $6, releaseName = $7, isCachedBuild = $8, notificationPendingSince = $4 where id = $1")
|
||
(build->id)
|
||
((int) (res.failed ? bsFailedWithOutput : bsSuccess))
|
||
(startTime)
|
||
(stopTime)
|
||
(res.size)
|
||
(res.closureSize)
|
||
(res.releaseName, res.releaseName != "")
|
||
(isCachedBuild ? 1 : 0).exec();
|
||
|
||
txn.parameterized("delete from BuildProducts where build = $1")(build->id).exec();
|
||
|
||
unsigned int productNr = 1;
|
||
for (auto & product : res.products) {
|
||
txn.parameterized
|
||
("insert into BuildProducts (build, productnr, type, subtype, fileSize, sha1hash, sha256hash, path, name, defaultPath) values ($1, $2, $3, $4, $5, $6, $7, $8, $9, $10)")
|
||
(build->id)
|
||
(productNr++)
|
||
(product.type)
|
||
(product.subtype)
|
||
(product.fileSize, product.isRegular)
|
||
(product.sha1hash.to_string(Base16, false), product.isRegular)
|
||
(product.sha256hash.to_string(Base16, false), product.isRegular)
|
||
(product.path)
|
||
(product.name)
|
||
(product.defaultPath).exec();
|
||
}
|
||
|
||
txn.parameterized("delete from BuildMetrics where build = $1")(build->id).exec();
|
||
|
||
for (auto & metric : res.metrics) {
|
||
txn.parameterized
|
||
("insert into BuildMetrics (build, name, unit, value, project, jobset, job, timestamp) values ($1, $2, $3, $4, $5, $6, $7, $8)")
|
||
(build->id)
|
||
(metric.second.name)
|
||
(metric.second.unit, metric.second.unit != "")
|
||
(metric.second.value)
|
||
(build->projectName)
|
||
(build->jobsetName)
|
||
(build->jobName)
|
||
(build->timestamp).exec();
|
||
}
|
||
|
||
nrBuildsDone++;
|
||
}
|
||
|
||
|
||
bool State::checkCachedFailure(Step::ptr step, Connection & conn)
|
||
{
|
||
pqxx::work txn(conn);
|
||
for (auto & path : step->drv.outputPaths())
|
||
if (!txn.parameterized("select 1 from FailedPaths where path = $1")(path).exec().empty())
|
||
return true;
|
||
return false;
|
||
}
|
||
|
||
|
||
void State::notifyBuildStarted(pqxx::work & txn, BuildID buildId)
|
||
{
|
||
txn.exec(fmt("notify build_started, '%s'", buildId));
|
||
}
|
||
|
||
|
||
void State::notifyBuildFinished(pqxx::work & txn, BuildID buildId,
|
||
const std::vector<BuildID> & dependentIds)
|
||
{
|
||
auto payload = fmt("%d ", buildId);
|
||
for (auto & d : dependentIds)
|
||
payload += fmt("%d ", d);
|
||
// FIXME: apparently parameterized() doesn't support NOTIFY.
|
||
txn.exec(fmt("notify build_finished, '%s'", payload));
|
||
}
|
||
|
||
|
||
std::shared_ptr<PathLocks> State::acquireGlobalLock()
|
||
{
|
||
Path lockPath = hydraData + "/queue-runner/lock";
|
||
|
||
createDirs(dirOf(lockPath));
|
||
|
||
auto lock = std::make_shared<PathLocks>();
|
||
if (!lock->lockPaths(PathSet({lockPath}), "", false)) return 0;
|
||
|
||
return lock;
|
||
}
|
||
|
||
|
||
void State::dumpStatus(Connection & conn, bool log)
|
||
{
|
||
std::ostringstream out;
|
||
|
||
{
|
||
JSONObject root(out);
|
||
time_t now = time(0);
|
||
root.attr("status", "up");
|
||
root.attr("time", time(0));
|
||
root.attr("uptime", now - startedAt);
|
||
root.attr("pid", getpid());
|
||
{
|
||
auto builds_(builds.lock());
|
||
root.attr("nrQueuedBuilds", builds_->size());
|
||
}
|
||
{
|
||
auto steps_(steps.lock());
|
||
for (auto i = steps_->begin(); i != steps_->end(); )
|
||
if (i->second.lock()) ++i; else i = steps_->erase(i);
|
||
root.attr("nrUnfinishedSteps", steps_->size());
|
||
}
|
||
{
|
||
auto runnable_(runnable.lock());
|
||
for (auto i = runnable_->begin(); i != runnable_->end(); )
|
||
if (i->lock()) ++i; else i = runnable_->erase(i);
|
||
root.attr("nrRunnableSteps", runnable_->size());
|
||
}
|
||
root.attr("nrActiveSteps", activeSteps_.lock()->size());
|
||
root.attr("nrStepsBuilding", nrStepsBuilding);
|
||
root.attr("nrStepsCopyingTo", nrStepsCopyingTo);
|
||
root.attr("nrStepsCopyingFrom", nrStepsCopyingFrom);
|
||
root.attr("nrStepsWaiting", nrStepsWaiting);
|
||
root.attr("bytesSent", bytesSent);
|
||
root.attr("bytesReceived", bytesReceived);
|
||
root.attr("nrBuildsRead", nrBuildsRead);
|
||
root.attr("buildReadTimeMs", buildReadTimeMs);
|
||
root.attr("buildReadTimeAvgMs", nrBuildsRead == 0 ? 0.0 : (float) buildReadTimeMs / nrBuildsRead);
|
||
root.attr("nrBuildsDone", nrBuildsDone);
|
||
root.attr("nrStepsStarted", nrStepsStarted);
|
||
root.attr("nrStepsDone", nrStepsDone);
|
||
root.attr("nrRetries", nrRetries);
|
||
root.attr("maxNrRetries", maxNrRetries);
|
||
if (nrStepsDone) {
|
||
root.attr("totalStepTime", totalStepTime);
|
||
root.attr("totalStepBuildTime", totalStepBuildTime);
|
||
root.attr("avgStepTime", (float) totalStepTime / nrStepsDone);
|
||
root.attr("avgStepBuildTime", (float) totalStepBuildTime / nrStepsDone);
|
||
}
|
||
root.attr("nrQueueWakeups", nrQueueWakeups);
|
||
root.attr("nrDispatcherWakeups", nrDispatcherWakeups);
|
||
root.attr("dispatchTimeMs", dispatchTimeMs);
|
||
root.attr("dispatchTimeAvgMs", nrDispatcherWakeups == 0 ? 0.0 : (float) dispatchTimeMs / nrDispatcherWakeups);
|
||
root.attr("nrDbConnections", dbPool.count());
|
||
root.attr("nrActiveDbUpdates", nrActiveDbUpdates);
|
||
root.attr("memoryTokensInUse", memoryTokens.currentUse());
|
||
|
||
{
|
||
auto nested = root.object("machines");
|
||
auto machines_(machines.lock());
|
||
for (auto & i : *machines_) {
|
||
auto & m(i.second);
|
||
auto & s(m->state);
|
||
auto nested2 = nested.object(m->sshName);
|
||
nested2.attr("enabled", m->enabled);
|
||
|
||
{
|
||
auto list = nested2.list("systemTypes");
|
||
for (auto & s : m->systemTypes)
|
||
list.elem(s);
|
||
}
|
||
|
||
{
|
||
auto list = nested2.list("supportedFeatures");
|
||
for (auto & s : m->supportedFeatures)
|
||
list.elem(s);
|
||
}
|
||
|
||
{
|
||
auto list = nested2.list("mandatoryFeatures");
|
||
for (auto & s : m->mandatoryFeatures)
|
||
list.elem(s);
|
||
}
|
||
|
||
nested2.attr("currentJobs", s->currentJobs);
|
||
if (s->currentJobs == 0)
|
||
nested2.attr("idleSince", s->idleSince);
|
||
nested2.attr("nrStepsDone", s->nrStepsDone);
|
||
if (m->state->nrStepsDone) {
|
||
nested2.attr("totalStepTime", s->totalStepTime);
|
||
nested2.attr("totalStepBuildTime", s->totalStepBuildTime);
|
||
nested2.attr("avgStepTime", (float) s->totalStepTime / s->nrStepsDone);
|
||
nested2.attr("avgStepBuildTime", (float) s->totalStepBuildTime / s->nrStepsDone);
|
||
}
|
||
|
||
auto info(m->state->connectInfo.lock());
|
||
nested2.attr("disabledUntil", std::chrono::system_clock::to_time_t(info->disabledUntil));
|
||
nested2.attr("lastFailure", std::chrono::system_clock::to_time_t(info->lastFailure));
|
||
nested2.attr("consecutiveFailures", info->consecutiveFailures);
|
||
|
||
}
|
||
}
|
||
|
||
{
|
||
auto nested = root.object("jobsets");
|
||
auto jobsets_(jobsets.lock());
|
||
for (auto & jobset : *jobsets_) {
|
||
auto nested2 = nested.object(jobset.first.first + ":" + jobset.first.second);
|
||
nested2.attr("shareUsed", jobset.second->shareUsed());
|
||
nested2.attr("seconds", jobset.second->getSeconds());
|
||
}
|
||
}
|
||
|
||
{
|
||
auto nested = root.object("machineTypes");
|
||
auto machineTypes_(machineTypes.lock());
|
||
for (auto & i : *machineTypes_) {
|
||
auto nested2 = nested.object(i.first);
|
||
nested2.attr("runnable", i.second.runnable);
|
||
nested2.attr("running", i.second.running);
|
||
if (i.second.runnable > 0)
|
||
nested2.attr("waitTime", i.second.waitTime.count() +
|
||
i.second.runnable * (time(0) - lastDispatcherCheck));
|
||
if (i.second.running == 0)
|
||
nested2.attr("lastActive", std::chrono::system_clock::to_time_t(i.second.lastActive));
|
||
}
|
||
}
|
||
|
||
auto store = getDestStore();
|
||
|
||
auto nested = root.object("store");
|
||
|
||
auto & stats = store->getStats();
|
||
nested.attr("narInfoRead", stats.narInfoRead);
|
||
nested.attr("narInfoReadAverted", stats.narInfoReadAverted);
|
||
nested.attr("narInfoMissing", stats.narInfoMissing);
|
||
nested.attr("narInfoWrite", stats.narInfoWrite);
|
||
nested.attr("narInfoCacheSize", stats.pathInfoCacheSize);
|
||
nested.attr("narRead", stats.narRead);
|
||
nested.attr("narReadBytes", stats.narReadBytes);
|
||
nested.attr("narReadCompressedBytes", stats.narReadCompressedBytes);
|
||
nested.attr("narWrite", stats.narWrite);
|
||
nested.attr("narWriteAverted", stats.narWriteAverted);
|
||
nested.attr("narWriteBytes", stats.narWriteBytes);
|
||
nested.attr("narWriteCompressedBytes", stats.narWriteCompressedBytes);
|
||
nested.attr("narWriteCompressionTimeMs", stats.narWriteCompressionTimeMs);
|
||
nested.attr("narCompressionSavings",
|
||
stats.narWriteBytes
|
||
? 1.0 - (double) stats.narWriteCompressedBytes / stats.narWriteBytes
|
||
: 0.0);
|
||
nested.attr("narCompressionSpeed", // MiB/s
|
||
stats.narWriteCompressionTimeMs
|
||
? (double) stats.narWriteBytes / stats.narWriteCompressionTimeMs * 1000.0 / (1024.0 * 1024.0)
|
||
: 0.0);
|
||
|
||
auto s3Store = dynamic_cast<S3BinaryCacheStore *>(&*store);
|
||
if (s3Store) {
|
||
auto nested2 = nested.object("s3");
|
||
auto & s3Stats = s3Store->getS3Stats();
|
||
nested2.attr("put", s3Stats.put);
|
||
nested2.attr("putBytes", s3Stats.putBytes);
|
||
nested2.attr("putTimeMs", s3Stats.putTimeMs);
|
||
nested2.attr("putSpeed",
|
||
s3Stats.putTimeMs
|
||
? (double) s3Stats.putBytes / s3Stats.putTimeMs * 1000.0 / (1024.0 * 1024.0)
|
||
: 0.0);
|
||
nested2.attr("get", s3Stats.get);
|
||
nested2.attr("getBytes", s3Stats.getBytes);
|
||
nested2.attr("getTimeMs", s3Stats.getTimeMs);
|
||
nested2.attr("getSpeed",
|
||
s3Stats.getTimeMs
|
||
? (double) s3Stats.getBytes / s3Stats.getTimeMs * 1000.0 / (1024.0 * 1024.0)
|
||
: 0.0);
|
||
nested2.attr("head", s3Stats.head);
|
||
nested2.attr("costDollarApprox",
|
||
(s3Stats.get + s3Stats.head) / 10000.0 * 0.004
|
||
+ s3Stats.put / 1000.0 * 0.005 +
|
||
+ s3Stats.getBytes / (1024.0 * 1024.0 * 1024.0) * 0.09);
|
||
}
|
||
}
|
||
|
||
if (log && time(0) >= lastStatusLogged + statusLogInterval) {
|
||
printMsg(lvlInfo, format("status: %1%") % out.str());
|
||
lastStatusLogged = time(0);
|
||
}
|
||
|
||
{
|
||
auto mc = startDbUpdate();
|
||
pqxx::work txn(conn);
|
||
// FIXME: use PostgreSQL 9.5 upsert.
|
||
txn.exec("delete from SystemStatus where what = 'queue-runner'");
|
||
txn.parameterized("insert into SystemStatus values ('queue-runner', $1)")(out.str()).exec();
|
||
txn.exec("notify status_dumped");
|
||
txn.commit();
|
||
}
|
||
}
|
||
|
||
|
||
void State::showStatus()
|
||
{
|
||
auto conn(dbPool.get());
|
||
receiver statusDumped(*conn, "status_dumped");
|
||
|
||
string status;
|
||
bool barf = false;
|
||
|
||
/* Get the last JSON status dump from the database. */
|
||
{
|
||
pqxx::work txn(*conn);
|
||
auto res = txn.exec("select status from SystemStatus where what = 'queue-runner'");
|
||
if (res.size()) status = res[0][0].as<string>();
|
||
}
|
||
|
||
if (status != "") {
|
||
|
||
/* If the status is not empty, then the queue runner is
|
||
running. Ask it to update the status dump. */
|
||
{
|
||
pqxx::work txn(*conn);
|
||
txn.exec("notify dump_status");
|
||
txn.commit();
|
||
}
|
||
|
||
/* Wait until it has done so. */
|
||
barf = conn->await_notification(5, 0) == 0;
|
||
|
||
/* Get the new status. */
|
||
{
|
||
pqxx::work txn(*conn);
|
||
auto res = txn.exec("select status from SystemStatus where what = 'queue-runner'");
|
||
if (res.size()) status = res[0][0].as<string>();
|
||
}
|
||
|
||
}
|
||
|
||
if (status == "") status = R"({"status":"down"})";
|
||
|
||
std::cout << status << "\n";
|
||
|
||
if (barf)
|
||
throw Error("queue runner did not respond; status information may be wrong");
|
||
}
|
||
|
||
|
||
void State::unlock()
|
||
{
|
||
auto lock = acquireGlobalLock();
|
||
if (!lock)
|
||
throw Error("hydra-queue-runner is currently running");
|
||
|
||
auto conn(dbPool.get());
|
||
|
||
clearBusy(*conn, 0);
|
||
|
||
{
|
||
pqxx::work txn(*conn);
|
||
txn.exec("delete from SystemStatus where what = 'queue-runner'");
|
||
txn.commit();
|
||
}
|
||
}
|
||
|
||
|
||
void State::run(BuildID buildOne)
|
||
{
|
||
/* Can't be bothered to shut down cleanly. Goodbye! */
|
||
auto callback = createInterruptCallback([&]() { std::_Exit(0); });
|
||
|
||
startedAt = time(0);
|
||
this->buildOne = buildOne;
|
||
|
||
auto lock = acquireGlobalLock();
|
||
if (!lock)
|
||
throw Error("hydra-queue-runner is already running");
|
||
|
||
Store::Params localParams;
|
||
localParams["max-connections"] = "16";
|
||
localParams["max-connection-age"] = "600";
|
||
localStore = openStore(getEnv("NIX_REMOTE"), localParams);
|
||
|
||
auto storeUri = config->getStrOption("store_uri");
|
||
_destStore = storeUri == "" ? localStore : openStore(storeUri);
|
||
|
||
useSubstitutes = config->getBoolOption("use-substitutes", false);
|
||
|
||
// FIXME: hacky mechanism for configuring determinism checks.
|
||
for (auto & s : tokenizeString<Strings>(config->getStrOption("xxx-jobset-repeats"))) {
|
||
auto s2 = tokenizeString<std::vector<std::string>>(s, ":");
|
||
if (s2.size() != 3) throw Error("bad value in xxx-jobset-repeats");
|
||
jobsetRepeats.emplace(std::make_pair(s2[0], s2[1]), std::stoi(s2[2]));
|
||
}
|
||
|
||
{
|
||
auto conn(dbPool.get());
|
||
clearBusy(*conn, 0);
|
||
dumpStatus(*conn, false);
|
||
}
|
||
|
||
std::thread(&State::monitorMachinesFile, this).detach();
|
||
|
||
std::thread(&State::queueMonitor, this).detach();
|
||
|
||
std::thread(&State::dispatcher, this).detach();
|
||
|
||
/* Periodically clean up orphaned busy steps in the database. */
|
||
std::thread([&]() {
|
||
while (true) {
|
||
sleep(180);
|
||
|
||
std::set<std::pair<BuildID, int>> steps;
|
||
{
|
||
auto orphanedSteps_(orphanedSteps.lock());
|
||
if (orphanedSteps_->empty()) continue;
|
||
steps = *orphanedSteps_;
|
||
orphanedSteps_->clear();
|
||
}
|
||
|
||
try {
|
||
auto conn(dbPool.get());
|
||
pqxx::work txn(*conn);
|
||
for (auto & step : steps) {
|
||
printMsg(lvlError, format("cleaning orphaned step %d of build %d") % step.second % step.first);
|
||
txn.parameterized
|
||
("update BuildSteps set busy = 0, status = $1 where build = $2 and stepnr = $3 and busy != 0")
|
||
((int) bsAborted)
|
||
(step.first)
|
||
(step.second).exec();
|
||
}
|
||
txn.commit();
|
||
} catch (std::exception & e) {
|
||
printMsg(lvlError, format("cleanup thread: %1%") % e.what());
|
||
auto orphanedSteps_(orphanedSteps.lock());
|
||
orphanedSteps_->insert(steps.begin(), steps.end());
|
||
}
|
||
}
|
||
}).detach();
|
||
|
||
/* Make sure that old daemon connections are closed even when
|
||
we're not doing much. */
|
||
std::thread([&]() {
|
||
while (true) {
|
||
sleep(10);
|
||
try {
|
||
if (auto remoteStore = getDestStore().dynamic_pointer_cast<RemoteStore>())
|
||
remoteStore->flushBadConnections();
|
||
} catch (std::exception & e) {
|
||
printMsg(lvlError, format("connection flush thread: %1%") % e.what());
|
||
}
|
||
}
|
||
}).detach();
|
||
|
||
/* Monitor the database for status dump requests (e.g. from
|
||
‘hydra-queue-runner --status’). */
|
||
while (true) {
|
||
try {
|
||
auto conn(dbPool.get());
|
||
receiver dumpStatus_(*conn, "dump_status");
|
||
while (true) {
|
||
conn->await_notification(statusLogInterval / 2 + 1, 0);
|
||
dumpStatus(*conn, true);
|
||
}
|
||
} catch (std::exception & e) {
|
||
printMsg(lvlError, format("main thread: %1%") % e.what());
|
||
sleep(10); // probably a DB problem, so don't retry right away
|
||
}
|
||
}
|
||
}
|
||
|
||
|
||
int main(int argc, char * * argv)
|
||
{
|
||
return handleExceptions(argv[0], [&]() {
|
||
initNix();
|
||
|
||
signal(SIGINT, SIG_DFL);
|
||
signal(SIGTERM, SIG_DFL);
|
||
signal(SIGHUP, SIG_DFL);
|
||
|
||
bool unlock = false;
|
||
bool status = false;
|
||
BuildID buildOne = 0;
|
||
|
||
parseCmdLine(argc, argv, [&](Strings::iterator & arg, const Strings::iterator & end) {
|
||
if (*arg == "--unlock")
|
||
unlock = true;
|
||
else if (*arg == "--status")
|
||
status = true;
|
||
else if (*arg == "--build-one") {
|
||
if (!string2Int<BuildID>(getArg(*arg, arg, end), buildOne))
|
||
throw Error("‘--build-one’ requires a build ID");
|
||
} else
|
||
return false;
|
||
return true;
|
||
});
|
||
|
||
settings.verboseBuild = true;
|
||
settings.lockCPU = false;
|
||
|
||
State state;
|
||
if (status)
|
||
state.showStatus();
|
||
else if (unlock)
|
||
state.unlock();
|
||
else
|
||
state.run(buildOne);
|
||
});
|
||
}
|