DomGarguilo commented on code in PR #5398: URL: https://github.com/apache/accumulo/pull/5398#discussion_r1992096528
########## core/src/main/java/org/apache/accumulo/core/rpc/AccumuloProtocolFactory.java: ########## @@ -0,0 +1,156 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * https://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ +package org.apache.accumulo.core.rpc; + +import org.apache.accumulo.core.trace.TraceUtil; +import org.apache.thrift.TException; +import org.apache.thrift.protocol.TCompactProtocol; +import org.apache.thrift.protocol.TMessage; +import org.apache.thrift.protocol.TProtocol; +import org.apache.thrift.transport.TTransport; + +import io.opentelemetry.api.trace.Span; +import io.opentelemetry.context.Scope; + +public class AccumuloProtocolFactory extends TCompactProtocol.Factory { + + private static final long serialVersionUID = 1L; + + private final boolean isClient; + + public static class AccumuloProtocol extends TCompactProtocol { + + private static final int MAGIC_NUMBER = 0x41434355; // "ACCU" in ASCII + private static final byte PROTOCOL_VERSION = 1; + + private final boolean isClient; + + private Span span = null; + private Scope scope = null; + + public AccumuloProtocol(TTransport transport, boolean isClient) { + super(transport); + this.isClient = isClient; + } + + @Override + public void writeMessageBegin(TMessage message) throws TException { + if (this.isClient) { Review Comment: Yea I am kinda unsure about this point (not sure whats correct/better). If I understand things correctly here, the thrift messages flow in both directions between "client" and "server" over the same connection. So as long as we validate compatibility when the client sends a message to the server, than we can avoid doing the extra work of adding (and therefore validating) the headers when the server responds to the client. I think things would work just fine if we did not distinguish between client and server in this context, we would just be doing more work than necessary which might be unoptimal (which seems like it matters in this context). I'm also thinking though there might be the benefit of always writing and validating these header just so we are certain that there are no edge cases or scenarios that I'm not thinking about where a server responds to a client over a connection that has not been validated yet. I don't know if this would ever happen but it seems like a trade off between absolute certainty that we have valid comms both ways and optimizing the amount of work done (again might be worth the optimization in this case, not sure). Definitely would appreciate feedback on this line of thinking and corrections regarding any incorrect assumptions ive made above. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
