Explicitly calling out the need to close the client in both the produ… - #4
Merged
Conversation
|
Thank you for submitting this pull request, however I do not see a valid CLA on file for you. Before we can merge this request please visit https://yahoocla.herokuapp.com/ and agree to the terms. Thanks! 😄 |
|
CLA is valid! |
Contributor
|
👍 Thanks for submitting the fix! |
sijie
referenced
this pull request
in sijie/pulsar
Mar 4, 2018
- introduce a `fs` package for function state related classes - add `FunctionConfig` for user defined function configurations - add `CmdFunctions`: it takes a yaml file and also command parameters
Closed
hrsakai
pushed a commit
to hrsakai/pulsar
that referenced
this pull request
Dec 10, 2020
Signed-off-by: xiaolong.ran <ranxiaolong716@gmail.com>
hangc0276
referenced
this pull request
in hangc0276/pulsar
May 26, 2021
Master Issue: #4 This is the request produce implementation. Basic logic: - For passed in PartitionTopic, create related Pulsar persistent topic. - For passed in Kafka Records, write data to related pulsar persistent topic.
hangc0276
referenced
this pull request
in hangc0276/pulsar
May 26, 2021
Master Issue: #4 This is the request Fetch implementation. Basic logic: For each Fetch request from a partition, create and maintain a NonDurableCursor to read from backed PersistentTopic. ** changes ** - Add basic code implementation - Add unit tests.
hangc0276
referenced
this pull request
in hangc0276/pulsar
May 26, 2021
Master Issue: #4 *Motivation* Support Kafka group coordinator protocols.
hangc0276
referenced
this pull request
in hangc0276/pulsar
May 26, 2021
Master Issue: #4 *Motivation* Support storing and committing offsets in group metadata manager.
dlg99
referenced
this pull request
in dlg99/pulsar
Jul 20, 2021
… Downgrading debezium to the same version as apache. (#4)
1 task
Closed
2 tasks
2 tasks
lhotari
added a commit
to lhotari/pulsar
that referenced
this pull request
Jul 4, 2026
…SslContexts (PIP-478 FIX-6) Motivation: apply PIP-478's own anti-kitchen-sink philosophy (Motivation #2/apache#4) to the TLS util layer. The generic org.apache.pulsar.common.util.SecurityUtility grab-bag mixes three unrelated concerns (PEM parsing, JCA/JCE provider resolution, JDK SSLContext assembly) plus dead legacy Netty/rotation helpers. SecurityUtility carries no @InterfaceStability and has no client-api importer, so it is internal and safe to decompose. Modifications: introduce three single-concern containers in a new package org.apache.pulsar.common.util.tls (kept out of the dependency-light pulsar-common-api; these pull BouncyCastle/Netty), each holding exactly one concern lifted verbatim from SecurityUtility: - PemReader — loadCertificatesFromPemFile/Stream + loadPrivateKeyFromPemFile/Stream. - JcaProviders — getProvider/isBCFIPS + BC/Conscrypt provider constants and the Conscrypt hostname-verifier workaround; resolveProvider and the TrustManager processing are package-private (only JdkSslContexts uses them). - JdkSslContexts — createSslContext(...) JDK SSLContext assembly, composing PemReader, JcaProviders and the existing KeyStoreHolder. This commit is purely additive; SecurityUtility is untouched and still compiles. Callers migrate in the follow-up commit, and SecurityUtility (plus the dead helpers) is removed after that. New package uses org.apache.pulsar.common.util.tls (not ...common.tls) to avoid a split package with pulsar-common-api's org.apache.pulsar.common.tls. Assisted-by: Claude Code (Opus 4.8)
lhotari
added a commit
to lhotari/pulsar
that referenced
this pull request
Jul 19, 2026
…vider, not a JCA crypto provider jsseProvider selects the provider that builds the TLS SSLContext (SslContextBuilder.sslContextProvider) — a JSSE role. The previous wording mis-framed it as a "JCA crypto provider" and used BCFIPS/PKCS#11 as example values, but a crypto-only provider such as BCFIPS exposes no SSLContext.TLS and cannot be named as the jsseProvider directly. Reframe the javadoc, @ApiModelProperty/@FieldContext descriptions, wiring comments, and pip-478.md (Motivation apache#4, Goal #5, out-of-scope, the SPI/design and Configuration sections, and the v4 sslProvider->jsseProvider migration): - jsseProvider names the JSSE (SSLContext) java.security.Provider that builds the SSLContext on the JDK engine (SslProvider.JDK + sslContextProvider), taking precedence over the factory engine choice. - FIPS example is now jsseProvider=BCJSSE (the BouncyCastle JSSE provider from bctls-fips), with the FIPS-140-validated BC-FIPS provider registered separately as the crypto provider BCJSSE delegates to. - Drop BCFIPS/PKCS#11 as jsseProvider VALUE examples; the v4-migration example values are the JSSE providers Conscrypt/SunJSSE. Also update BCFIPS test tokens to BCJSSE in the propagation/mapping tests (they assert the config is carried, never resolve a provider), keeping the eager PulsarAdmin test on the resolvable SunJSSE. The JcaProviders utility keeps its name and generic JCA/JCE role. Assisted-by: Claude Code (Opus 4.8)
lhotari
added a commit
to lhotari/pulsar
that referenced
this pull request
Jul 19, 2026
…on-default value" Two documentation reconciliations in pip/pip-478.md: - Break summary (apache#4): the PIP-337-removed BREAKING CHANGE bullet listed only the v4 builder-method source break and framed ClusterData as a metadata-model removal. Add the ClusterData brokerClientSslFactoryPlugin/...Params API accessors + builder methods (another public-API source break) and the pulsar-admin clusters --tls-factory-plugin / --tls-factory-plugin-params CmdClusters CLI options that wrote them, so the migration guidance is not understated. (The detailed removal-impact table already inventoried these.) - "Non-default value" (#6): give the removed-key validation a precise definition matching the implementation — a *Plugin key is tolerated (treated as unset) when blank or equal to the old default FQCN org.apache.pulsar.common.util.DefaultPulsarSslFactory, and rejected otherwise; a *PluginParams key has no default, so any non-blank value is rejected. Make the config-file removed-key wording consistently "stale, non-default" (the separate ClusterData metadata lenient-drop case stays "stale value", since the removed-key validation does not run on a metadata read). Assisted-by: Claude Code (Opus 4.8)
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
Motivation
I'm making this change for clarification in the Getting Started guide which is the gateway for new users to Pulsar.
Modifications
Updated documentation.
Result
Just the documentation.