/**
* This attempts to create an acquired state znode for the procedure (snapshot name).
*
* <p>It then looks for the reached znode to trigger in-barrier execution. If not present we have
* a watcher, if present then trigger the in-barrier action.
*/
@Override
public void sendMemberAcquired(Subprocedure sub) throws IOException {
String procName = sub.getName();
try {
LOG.debug(
"Member: '"
+ memberName
+ "' joining acquired barrier for procedure ("
+ procName
+ ") in zk");
String acquiredZNode =
ZKUtil.joinZNode(
ZKProcedureUtil.getAcquireBarrierNode(zkController, procName), memberName);
ZKUtil.createAndFailSilent(zkController.getWatcher(), acquiredZNode);
// watch for the complete node for this snapshot
String reachedBarrier = zkController.getReachedBarrierNode(procName);
LOG.debug("Watch for global barrier reached:" + reachedBarrier);
if (ZKUtil.watchAndCheckExists(zkController.getWatcher(), reachedBarrier)) {
receivedReachedGlobalBarrier(reachedBarrier);
}
} catch (KeeperException e) {
member.controllerConnectionFailure(
"Failed to acquire barrier for procedure: " + procName + " and member: " + memberName,
e,
procName);
}
}
/**
* Kick off a new sub-procedure on the listener with the data stored in the passed znode.
*
* <p>Will attempt to create the same procedure multiple times if an procedure znode with the same
* name is created. It is left up the coordinator to ensure this doesn't occur.
*
* @param path full path to the znode for the procedure to start
*/
private synchronized void startNewSubprocedure(String path) {
LOG.debug("Found procedure znode: " + path);
String opName = ZKUtil.getNodeName(path);
// start watching for an abort notification for the procedure
String abortZNode = zkController.getAbortZNode(opName);
try {
if (ZKUtil.watchAndCheckExists(zkController.getWatcher(), abortZNode)) {
LOG.debug("Not starting:" + opName + " because we already have an abort notification.");
return;
}
} catch (KeeperException e) {
member.controllerConnectionFailure(
"Failed to get the abort znode (" + abortZNode + ") for procedure :" + opName, e, opName);
return;
}
// get the data for the procedure
Subprocedure subproc = null;
try {
byte[] data = ZKUtil.getData(zkController.getWatcher(), path);
if (!ProtobufUtil.isPBMagicPrefix(data)) {
String msg =
"Data in for starting procuedure "
+ opName
+ " is illegally formatted (no pb magic). "
+ "Killing the procedure: "
+ Bytes.toString(data);
LOG.error(msg);
throw new IllegalArgumentException(msg);
}
LOG.debug("start proc data length is " + data.length);
data = Arrays.copyOfRange(data, ProtobufUtil.lengthOfPBMagic(), data.length);
LOG.debug("Found data for znode:" + path);
subproc = member.createSubprocedure(opName, data);
member.submitSubprocedure(subproc);
} catch (IllegalArgumentException iae) {
LOG.error("Illegal argument exception", iae);
sendMemberAborted(subproc, new ForeignException(getMemberName(), iae));
} catch (IllegalStateException ise) {
LOG.error("Illegal state exception ", ise);
sendMemberAborted(subproc, new ForeignException(getMemberName(), ise));
} catch (KeeperException e) {
member.controllerConnectionFailure(
"Failed to get data for new procedure:" + opName, e, opName);
} catch (InterruptedException e) {
member.controllerConnectionFailure(
"Failed to get data for new procedure:" + opName, e, opName);
Thread.currentThread().interrupt();
}
}
private String submitTaskAndWait(TaskBatch batch, String name)
throws KeeperException, InterruptedException {
String tasknode = ZKSplitLog.getEncodedNodeName(zkw, name);
NodeCreationListener listener = new NodeCreationListener(zkw, tasknode);
zkw.registerListener(listener);
ZKUtil.watchAndCheckExists(zkw, tasknode);
slm.installTask(name, batch);
assertEquals(1, batch.installed);
assertTrue(slm.findOrCreateOrphanTask(tasknode).batch == batch);
assertEquals(1L, tot_mgr_node_create_queued.get());
LOG.debug("waiting for task node creation");
listener.waitForCreation();
LOG.debug("task created");
return tasknode;
}