[improve][client] Make seek wait for reconnect

[nodece]

Motivation

When doing seek operation, the client needs to seed a seek command to the broker. When one seeks is successful, the broker disconnects the client, and then the client reconnects and resubscribes to the broker based on the seek position.

Right now, the consumer responds to the caller after sending a seek command succeeds, this should be incorrect behavior. The right behavior should be to wait for reconnect and resubscribe, then respond to the caller.

Modifications

Based on the global CompletableFuture and seekMessageId check the seek operation to wait for reconnect and resubscribe.

When reconnecting is done, make the CompletableFuture to complete, when resubscribing is done, cleanup the seekMessageId.

Documentation

• [ ] doc
• [ ] doc-required
• [x] doc-not-needed
• [ ] doc-complete

Matching PR in forked repository

PR in forked repository: https://github.com/nodece/pulsar/pull/9

[codecov-commenter]

Codecov Report

Merging #18202 (68ba899) into master (6c65ca0) will increase coverage by 5.44%. The diff coverage is 35.90%.

@@             Coverage Diff              @@
##             master   #18202      +/-   ##
============================================
+ Coverage     34.91%   40.35%   +5.44%     
- Complexity     5707     8685    +2978     
============================================
  Files           607      687      +80     
  Lines         53396    67396   +14000     
  Branches       5712     7219    +1507     
============================================
+ Hits          18644    27200    +8556     
- Misses        32119    37161    +5042     
- Partials       2633     3035     +402     

Flag	Coverage Δ
unittests	40.35% <35.90%> (+5.44%)	:arrow_up:

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...org/apache/bookkeeper/mledger/LedgerOffloader.java	0.00% <ø> (ø)
...che/bookkeeper/mledger/LedgerOffloaderFactory.java	0.00% <ø> (ø)
...pache/bookkeeper/mledger/LedgerOffloaderStats.java	0.00% <ø> (ø)
...ookkeeper/mledger/LedgerOffloaderStatsDisable.java	0.00% <ø> (ø)
...a/org/apache/bookkeeper/mledger/ManagedCursor.java	85.71% <ø> (ø)
...che/bookkeeper/mledger/ManagedLedgerException.java	41.17% <ø> (ø)
...bookkeeper/mledger/ManagedLedgerFactoryConfig.java	86.66% <ø> (ø)
...g/apache/bookkeeper/mledger/ManagedLedgerInfo.java	22.22% <ø> (ø)
...he/bookkeeper/mledger/OffloadedLedgerMetadata.java	0.00% <ø> (ø)
...ava/org/apache/bookkeeper/mledger/ScanOutcome.java	0.00% <ø> (ø)
... and 741 more

[lhotari]

@nodece Do you think that this PR addresses #10671 ?

[nodece]

@nodece Do you think that this PR addresses #10671 ?

I think this PR can fix race condition of seeking.

[lhotari]

is there any relationship to #16757?

[nodece]

is there any relationship to #16757?

Right, but this PR fixes some issues on the client side.

[nodece]

I think we should reconsider seek. This PR only solves some problems, but it does not solve all seek and reconnect concurrency problems.

Do you have any suggestions?

[congbobo184]

I have no idea about that. [PIP-194] https://lists.apache.org/thread/97o9t4ltkds5pfq41l9xbbd31t41qm8w discuss how to solve this problem.

[nodece]

I have no idea about that. [PIP-194] https://lists.apache.org/thread/97o9t4ltkds5pfq41l9xbbd31t41qm8w discuss how to solve this problem.

I think this is a different issue, I just fix the concurrency problems of seeking on the client side.

[github-actions[bot]]

The pr had no activity for 30 days, mark with Stale label.

[nodece]

No reviewer, so close this PR.