Skip to content

KAFKA-2614; No more clients can connect after TooManyConnectionsException threshold (max.connections.per.ip) is reached#288

Closed
ijuma wants to merge 3 commits into
apache:trunkfrom
ijuma:kafka-2614-connection-count-not-updated
Closed

KAFKA-2614; No more clients can connect after TooManyConnectionsException threshold (max.connections.per.ip) is reached#288
ijuma wants to merge 3 commits into
apache:trunkfrom
ijuma:kafka-2614-connection-count-not-updated

Conversation

@ijuma

@ijuma ijuma commented Oct 8, 2015

Copy link
Copy Markdown
Member
  • Call ConnectionQuotas.decr when calling Selector.close and when disconnections happen.
  • Expand SocketServerTest to test for this and to close sockets.
  • Refactor and clean-up SocketServer and Acceptor to make the code easier to understand.

…isconnections happen

Also:
* Expand `SocketServerTest` to test for this and to close sockets.
* Refactor and clean-up `SocketServer` and `Acceptor` to make the code easier to understand.
@ijuma

ijuma commented Oct 8, 2015

Copy link
Copy Markdown
Member Author

Tests passed locally.

@ijuma

ijuma commented Oct 8, 2015

Copy link
Copy Markdown
Member Author

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Does this work for ipv6? We probably should use the parser in BrokerEndPoint.

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good point, fixed as you suggested.

ijuma added 2 commits October 9, 2015 08:03
…n-count-not-updated

* apache/trunk:
  KAFKA-2596: reject commits from unknown groups with positive generations
  MINOR: typing ProcessorDef
…ion string

The previous code did not handle IPv6 properly as pointed out by Jun.

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

As far as I can see socket can never be null here.

@ijuma

ijuma commented Oct 9, 2015

Copy link
Copy Markdown
Member Author

@junrao Fixed the issue you mentioned, tests passing locally. I also moved some comments I had mistakenly left on the commit into the "Files changed" tab to make the review easier.

@junrao

junrao commented Oct 10, 2015

Copy link
Copy Markdown
Contributor

Thanks for the patch. LGTM

@asfgit asfgit closed this in 1265d7c Oct 10, 2015
@ijuma ijuma deleted the kafka-2614-connection-count-not-updated branch March 1, 2016 22:47
wyuka pushed a commit to wyuka/kafka that referenced this pull request Mar 4, 2022
…= KAFKA_2_3_IV2 (apache#288)

TICKET = KAFKA-7186
LI_DESCERIPTION =
The original PR for this change is apache#75.
Before this patch, the controller needs to allocate an individual UpdateMetadataRequest.Builder
for each broker. This may incur significant memory overhead if the controller has just started,
and is trying to send the cluster's full metadata to all brokers.

This PR tries to reduce the memory footprint by reusing the same UpdateMetadataRequest.Builder.
This is only achievable when all of the UMR have the same body payload, which means
the common UMR should use the maxBrokerEpoch field instead of individual broker epochs.

EXIT_CRITERIA = KAFKA-7186
wyuka pushed a commit to wyuka/kafka that referenced this pull request Mar 28, 2022
…= KAFKA_2_3_IV2 (apache#288)

TICKET = KAFKA-7186
LI_DESCERIPTION =
The original PR for this change is apache#75.
Before this patch, the controller needs to allocate an individual UpdateMetadataRequest.Builder
for each broker. This may incur significant memory overhead if the controller has just started,
and is trying to send the cluster's full metadata to all brokers.

This PR tries to reduce the memory footprint by reusing the same UpdateMetadataRequest.Builder.
This is only achievable when all of the UMR have the same body payload, which means
the common UMR should use the maxBrokerEpoch field instead of individual broker epochs.

EXIT_CRITERIA = KAFKA-7186
wyuka pushed a commit to wyuka/kafka that referenced this pull request Jun 16, 2022
…= KAFKA_2_3_IV2 (apache#288)

TICKET = KAFKA-7186
LI_DESCERIPTION =
The original PR for this change is apache#75.
Before this patch, the controller needs to allocate an individual UpdateMetadataRequest.Builder
for each broker. This may incur significant memory overhead if the controller has just started,
and is trying to send the cluster's full metadata to all brokers.

This PR tries to reduce the memory footprint by reusing the same UpdateMetadataRequest.Builder.
This is only achievable when all of the UMR have the same body payload, which means
the common UMR should use the maxBrokerEpoch field instead of individual broker epochs.

EXIT_CRITERIA = KAFKA-7186
davide-armand pushed a commit to aiven/kafka that referenced this pull request Dec 1, 2025
jeqo added a commit to aiven/kafka that referenced this pull request Jan 16, 2026
traceyyoshima added a commit to traceyyoshima/kafka that referenced this pull request May 22, 2026
Single bulk apply of the Language Engine's IntelliJ-style format
profile across the kafka source tree. Pairs with the IntelliJ-real
control branch `intellij-formatting` for side-by-side comparison.

Engine state at apply time includes the following format fixes
landed against the language-engine repo:

  Pre-PR #11:
  - PR apache#275 multi-line // comment indent group
  - PR apache#276 BLANK_LINES_AROUND_CLASS + sibling no-op repair
  - PR apache#277 string-concat chain anchor preservation
  - PR apache#279 forward style through apply routes
  - PR apache#281 Result-tab line-number alignment
  - PR apache#282 spaces after return / throw / yield / instanceof
  - PR apache#283 chain-dot postfix preservation
  - PR apache#284 BlankLines sibling minimum.* conversion
  - PR apache#285 chain-dot single-anchor (N=1) preservation
  - PR apache#286 string-concat chain partial-cascade fix
  - PR apache#287 method-decl param re-align for misaligned source
  - PR apache#288 annotation-in-array-init indent

  Post-original-PR #11 (new in this re-apply):
  - PR apache#289 partial-cascade post-rewrite-anchor (4 shape fixes:
    chain-dot postfix follow-on, MI/NewClass close-paren cascade,
    ternary continuation, lambda body continuation)

Stats: 3101 files written, 0 failures, 2678 already-idempotent
files skipped (no_change). Corpus byte delta vs trunk: -510 bytes
(106275 insertions / 106785 deletions).
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants