Fix NPE of cumulative ack mode and incorrect unack message count #14021
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
@congbobo184:Thanks for your contribution. For this PR, do we need to update docs? |
congbobo184
added
release/2.10.0
doc-not-needed
|
@congbobo184:Thanks for providing doc info! |
|
@congbobo184:Thanks for providing doc info! |
mattisonchao
approved these changes
Jan 29, 2022
|
@Technoboy- PTAL :) |
codelipenghui
approved these changes
Jan 29, 2022
|
Unack messages for Failover and Exclusive subscription is meaningless. |
gaoran10
reviewed
Jan 29, 2022
...oker/src/test/java/org/apache/pulsar/broker/service/BatchMessageWithBatchIndexLevelTest.java
Outdated
Show resolved
Hide resolved
codelipenghui
changed the title
added 2 commits
January 29, 2022 19:10
…rease_broker_consumer_unackMessageCount_throw_npe
…unt_throw_npe' of https://github.com/congbobo184/pulsar into congbobo184_fix_decrease_broker_consumer_unackMessageCount_throw_npe
added 2 commits
January 29, 2022 19:21
eolivelli
approved these changes
Jan 29, 2022
Technoboy-
reviewed
Jan 29, 2022
pulsar-broker/src/main/java/org/apache/pulsar/broker/service/Consumer.java
Outdated
Show resolved
Hide resolved
Technoboy-
reviewed
Jan 29, 2022
pulsar-broker/src/main/java/org/apache/pulsar/broker/service/Consumer.java
Outdated
Show resolved
Hide resolved
codelipenghui
pushed a commit
that referenced
this pull request
Jan 30, 2022
) link #13383 ## Motivation #13383 has fixed the batch message ack does not decrease the unacked-msg count, but in cumulative ack mode also decrease, it will use pendingAcks, but in cumulative ack, this will not init.   If ack the batch index one by one, the last ack of a batch will decrease unack message with `batchSize` ``` ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 9 ================ batch size -> 10 ``` ### Modifications add judge `Subscription.isIndividualAckMode(subType)` when get ackCount. If the ack from consumer don't have ackset, we should treat it as empty ackset to calculate the ack count with the currently ackset. (cherry picked from commit 618f17c)
zymap
pushed a commit
that referenced
this pull request
Jan 30, 2022
) link #13383 ## Motivation #13383 has fixed the batch message ack does not decrease the unacked-msg count, but in cumulative ack mode also decrease, it will use pendingAcks, but in cumulative ack, this will not init.   If ack the batch index one by one, the last ack of a batch will decrease unack message with `batchSize` ``` ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 9 ================ batch size -> 10 ``` ### Modifications add judge `Subscription.isIndividualAckMode(subType)` when get ackCount. If the ack from consumer don't have ackset, we should treat it as empty ackset to calculate the ack count with the currently ackset. (cherry picked from commit 618f17c)
congbobo184
deleted the
congbobo184_fix_decrease_broker_consumer_unackMessageCount_throw_npe
branch
Nicklee007
pushed a commit
to Nicklee007/pulsar
that referenced
this pull request
Apr 20, 2022
…che#14021) link apache#13383 ## Motivation apache#13383 has fixed the batch message ack does not decrease the unacked-msg count, but in cumulative ack mode also decrease, it will use pendingAcks, but in cumulative ack, this will not init.   If ack the batch index one by one, the last ack of a batch will decrease unack message with `batchSize` ``` ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 1 ================ batch size -> 10 ================ message id -> 3:1 ================ acked count -> 9 ================ batch size -> 10 ``` ### Modifications add judge `Subscription.isIndividualAckMode(subType)` when get ackCount. If the ack from consumer don't have ackset, we should treat it as empty ackset to calculate the ack count with the currently ackset.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
link #13383
Motivation
#13383 has fixed the batch message ack does not decrease the unacked-msg count, but in cumulative ack mode also decrease, it will use pendingAcks, but in cumulative ack, this will not init.
If ack the batch index one by one, the last ack of a batch will decrease unack message with
batchSizeModifications
add judge
Subscription.isIndividualAckMode(subType)when get ackCount.If the ack from consumer don't have ackset, we should treat it as empty ackset to calculate the ack count with the currently ackset.
Verifying this change
Add the tests for it
Does this pull request potentially affect one of the following parts:
If yes was chosen, please highlight the changes
Dependencies (does it add or upgrade a dependency): (no)
The public API: (no)
The schema: (no)
The default values of configurations: (no)
The wire protocol: (no)
The rest endpoints: (no)
The admin cli options: (no)
Anything that affects deployment: (no)