davidradl commented on code in PR #27318:
URL: https://github.com/apache/flink/pull/27318#discussion_r2588389429
##########
flink-runtime/src/main/java/org/apache/flink/runtime/blob/BlobClient.java:
##########
@@ -95,7 +95,7 @@ public BlobClient(InetSocketAddress serverAddress,
Configuration clientConfig)
// the
// InetSocketAddress can cache a failure in hostname resolution
forever.
socket.connect(
- new InetSocketAddress(serverAddress.getHostName(),
serverAddress.getPort()),
+ new InetSocketAddress(serverAddress.getHostString(),
serverAddress.getPort()),
Review Comment:
The change looks good. nit: I wonder if we could add a comment to explain
why we do. not use getHostName. Maybe include some of the following reasons:
Addresses are often configured as IP addresses for reliability
DNS may be unavailable or slow in containerized/cloud environments
Every blob transfer would trigger unnecessary reverse DNS lookups
Connection failures could occur if DNS resolution fails
--
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]
