return (System.currentTimeMillis() >> 10) + _investigateRetryInterval;
}
long vmId = work.getInstanceId();
VMInstanceVO vm = _itMgr.findByIdAndType(work.getType(), work.getInstanceId());
if (vm == null) {
s_logger.info("Unable to find vm: " + vmId);
return null;
}
s_logger.info("HA on " + vm);
if (vm.getState() != work.getPreviousState() || vm.getUpdated() != work.getUpdateTime()) {
s_logger.info("VM " + vm + " has been changed. Current State = " + vm.getState() + " Previous State = " + work.getPreviousState() + " last updated = " + vm.getUpdated()
+ " previous updated = " + work.getUpdateTime());
return null;
}
short alertType = AlertManager.ALERT_TYPE_USERVM;
if (VirtualMachine.Type.DomainRouter.equals(vm.getType())) {
alertType = AlertManager.ALERT_TYPE_DOMAIN_ROUTER;
} else if (VirtualMachine.Type.ConsoleProxy.equals(vm.getType())) {
alertType = AlertManager.ALERT_TYPE_CONSOLE_PROXY;
} else if (VirtualMachine.Type.SecondaryStorageVm.equals(vm.getType())) {
alertType = AlertManager.ALERT_TYPE_SSVM;
}
HostVO host = _hostDao.findById(work.getHostId());
boolean isHostRemoved = false;
if (host == null) {
host = _hostDao.findByIdIncludingRemoved(work.getHostId());
if (host != null) {
s_logger.debug("VM " + vm.toString() + " is now no longer on host " + work.getHostId() + " as the host is removed");
isHostRemoved = true;
}
}
DataCenterVO dcVO = _dcDao.findById(host.getDataCenterId());
HostPodVO podVO = _podDao.findById(host.getPodId());
String hostDesc = "name: " + host.getName() + "(id:" + host.getId() + "), availability zone: " + dcVO.getName() + ", pod: " + podVO.getName();
Boolean alive = null;
if (work.getStep() == Step.Investigating) {
if (!isHostRemoved) {
if (vm.getHostId() == null || vm.getHostId() != work.getHostId()) {
s_logger.info("VM " + vm.toString() + " is now no longer on host " + work.getHostId());
return null;
}
Investigator investigator = null;
for(Investigator it : _investigators) {
investigator = it;
alive = investigator.isVmAlive(vm, host);
s_logger.info(investigator.getName() + " found " + vm + "to be alive? " + alive);
if (alive != null) {
break;
}
}
boolean fenced = false;
if (alive == null) {
s_logger.debug("Fencing off VM that we don't know the state of");
for(FenceBuilder fb : _fenceBuilders) {
Boolean result = fb.fenceOff(vm, host);
s_logger.info("Fencer " + fb.getName() + " returned " + result);
if (result != null && result) {
fenced = true;
break;
}
}
} else if (!alive) {
fenced = true;
} else {
s_logger.debug("VM " + vm.getHostName() + " is found to be alive by " + investigator.getName());
if (host.getStatus() == Status.Up) {
s_logger.info(vm + " is alive and host is up. No need to restart it.");
return null;
} else {
s_logger.debug("Rescheduling because the host is not up but the vm is alive");
return (System.currentTimeMillis() >> 10) + _investigateRetryInterval;
}
}
if (!fenced) {
s_logger.debug("We were unable to fence off the VM " + vm);
_alertMgr.sendAlert(alertType, vm.getDataCenterId(), vm.getPodIdToDeployIn(), "Unable to restart " + vm.getHostName() + " which was running on host " + hostDesc,
"Insufficient capacity to restart VM, name: " + vm.getHostName() + ", id: " + vmId + " which was running on host " + hostDesc);
return (System.currentTimeMillis() >> 10) + _restartRetryInterval;
}
try {
_itMgr.advanceStop(vm, true, _accountMgr.getSystemUser(), _accountMgr.getSystemAccount());
} catch (ResourceUnavailableException e) {
assert false : "How do we hit this when force is true?";
throw new CloudRuntimeException("Caught exception even though it should be handled.", e);
} catch (OperationTimedoutException e) {
assert false : "How do we hit this when force is true?";
throw new CloudRuntimeException("Caught exception even though it should be handled.", e);
} catch (ConcurrentOperationException e) {
assert false : "How do we hit this when force is true?";
throw new CloudRuntimeException("Caught exception even though it should be handled.", e);
}
work.setStep(Step.Scheduled);
_haDao.update(work.getId(), work);
} else {
s_logger.debug("How come that HA step is Investigating and the host is removed? Calling forced Stop on Vm anyways");
try {
_itMgr.advanceStop(vm, true, _accountMgr.getSystemUser(), _accountMgr.getSystemAccount());
} catch (ResourceUnavailableException e) {
assert false : "How do we hit this when force is true?";
throw new CloudRuntimeException("Caught exception even though it should be handled.", e);
} catch (OperationTimedoutException e) {
assert false : "How do we hit this when force is true?";
throw new CloudRuntimeException("Caught exception even though it should be handled.", e);
} catch (ConcurrentOperationException e) {
assert false : "How do we hit this when force is true?";
throw new CloudRuntimeException("Caught exception even though it should be handled.", e);
}
}
}
vm = _itMgr.findByIdAndType(vm.getType(), vm.getId());
if (!_forceHA && !vm.isHaEnabled()) {
if (s_logger.isDebugEnabled()) {
s_logger.debug("VM is not HA enabled so we're done.");
}
return null; // VM doesn't require HA
}
if (!this.volumeMgr.canVmRestartOnAnotherServer(vm.getId())) {
if (s_logger.isDebugEnabled()) {
s_logger.debug("VM can not restart on another server.");
}
return null;
}
if (work.getTimesTried() > _maxRetries) {
s_logger.warn("Retried to max times so deleting: " + vmId);
return null;
}
try {
HashMap<VirtualMachineProfile.Param, Object> params = new HashMap<VirtualMachineProfile.Param, Object>();
if (_haTag != null) {
params.put(VirtualMachineProfile.Param.HaTag, _haTag);
}
VMInstanceVO started = _itMgr.advanceStart(vm, params, _accountMgr.getSystemUser(), _accountMgr.getSystemAccount());
if (started != null) {
s_logger.info("VM is now restarted: " + vmId + " on " + started.getHostId());
return null;
}
if (s_logger.isDebugEnabled()) {
s_logger.debug("Rescheduling VM " + vm.toString() + " to try again in " + _restartRetryInterval);