[Zen2] Add HandshakingTransportAddressConnector #32643
Conversation
The `PeerFinder`, introduced in elastic#32246, needs to be able to identify, and connect to, a remote master node using only its `TransportAddress`. This can be done by opening a single-channel connection to the address, performing a handshake, and only then forming a full-blown connection to the node. This change implements this logic.
DaveCTurner
added
>enhancement
v7.0.0
:Distributed Coordination/Cluster Coordination
|
Pinging @elastic/es-distributed |
| @Override | ||
| protected void doRun() throws Exception { | ||
|
|
||
| // TODO if transportService is already connected to this address then skip the handshaking |
There was a problem hiding this comment.
@tbrooks8 @ywelsch I think you discussed this area recently. Arguably it might make sense for some (or more) of this functionality to move elsewhere. Perhaps within TransportService itself as it currently stands although I know that you've plans for this area so that might change in future.
| IOUtils.closeWhileHandlingException(connection); | ||
| } | ||
|
|
||
| // NOMERGE better exceptions for failure cases? |
There was a problem hiding this comment.
maybe use ConnectTransportException? We use that one e.g. when application-level handshake fails (see connectToNode).
| } | ||
|
|
||
| void assertFailure() throws InterruptedException { | ||
| assertTrue(completionLatch.await(1, TimeUnit.SECONDS)); |
There was a problem hiding this comment.
maybe this is a little optimistic for our slow CI machines. 30 secs?
Same comment for other places in this class.
| .put(NODE_NAME_SETTING.getKey(), "node") | ||
| .put(CLUSTER_NAME_SETTING.getKey(), "local-cluster") | ||
| .build(); | ||
| threadPool = new ThreadPool(settings); |
There was a problem hiding this comment.
use TestThreadPool instead?
| @After | ||
| public void stopServices() { | ||
| transportService.stop(); | ||
| threadPool.shutdown(); |
There was a problem hiding this comment.
use terminate(threadPool); (this method is in ESTestCase)
|
Failure is #32215 not this PR. @elasticmachine retest this please. |
The
PeerFinder, introduced in #32246, needs to be able to identify, andconnect to, a remote master node using only its
TransportAddress. This can bedone by opening a single-channel connection to the address, performing a
handshake, and only then forming a full-blown connection to the node. This
change implements this logic.