jolshan commented on a change in pull request #10282:
URL: https://github.com/apache/kafka/pull/10282#discussion_r593351308
##########
File path: core/src/main/scala/kafka/cluster/Partition.scala
##########
@@ -425,36 +425,35 @@ class Partition(val topicPartition: TopicPartition,
}
/**
- * Checks if the topic ID provided in the request is consistent with the
topic ID in the log.
+ * Checks if the topic ID received is consistent with the topic ID in the
log.
* If a valid topic ID is provided, and the log exists but has no ID set,
set the log ID to be the request ID.
*
- * @param requestTopicId the topic ID from the request
- * @return true if the request topic id is consistent, false otherwise
+ * @param receivedTopicId the topic ID from the LeaderAndIsr request or from
the metadata records
+ * @return true if the received topic id is consistent, false otherwise
*/
- def checkOrSetTopicId(requestTopicId: Uuid): Boolean = {
- // If the request had an invalid topic ID, then we assume that topic IDs
are not supported.
- // The topic ID was not inconsistent, so return true.
- // If the log is empty, then we can not say that topic ID is inconsistent,
so return true.
- if (requestTopicId == null || requestTopicId == Uuid.ZERO_UUID)
- true
- else {
+ def checkOrSetTopicId(receivedTopicId: Uuid, usingRaft: Boolean): Boolean = {
Review comment:
Hmm this is a little tricky. One idea I had would be to check or set the
ID in the partition object only (not the file or in Log memory) then when the
the partitions are passed in to `makeLeaders/Followers` and a new log is
created, we will have the topic ID easily accessible to create the
partition.metadata file. It kind of changes how we do things. We would either
store in partition **_instead of_** or **_in addition to_** Log. Not sure if
we also want to change this for the ZK code. Thoughts?
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
