Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Doc: Add json_lines known issue to release notes #16831

Draft
wants to merge 5 commits into
base: 8.17
Choose a base branch
from
Draft
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
95 changes: 88 additions & 7 deletions docs/static/releasenotes.asciidoc
Original file line number Diff line number Diff line change
Expand Up @@ -76,12 +76,39 @@ This section summarizes the changes in the following releases:
=== Logstash 8.17.0 Release Notes

[[known-issues-8-17-0]]
==== Known issue
==== Known issues

[[known-issue-8-17-0-jvm]]
===== JVM version changes needed when upgrading {ls} from 8.12.0 (or earlier)

If the `jvm.options` file was modified and not overwritten with the newest version, you may see a "deserialize invocation error" message, causing the pipeline to crash.
Users are affected if the Persistent Queue (PQ) is enabled, and the pipeline is processing messages larger than 20MB.

**Solution:** Apply the default change contained in the newer 'jvm.options' file, as seen in this https://github.com/elastic/logstash/blob/main/config/jvm.options#L66-L82[example].

[[known-issue-8-17-0-json_lines]]
===== "Input buffer full" error with {ls} 8.16.0, 8.16.1, or 8.17.0

If you are using `json_lines` codec 3.2.0 (or later) with {ls} 8.16.0, 8.16.1, or 8.17.0, you may see an error similar to this one, crashing the pipelines:
```
unable to process event. {:message=>"input buffer full", :class=>"Java::JavaLang::IllegalStateException", :backtrace=>["org.logstash.common.BufferedTokenizerExt.extract(BufferedTokenizerExt.java:83)", "usr.share.logstash.vendor.bundle.jruby.$3_dot_1_dot_0.gems.logstash_minus_codec_minus_json_lines_minus_3_dot_2_dot_2.lib.logstash.codecs.json_lines.RUBY$method$decode$0(/usr/share/logstash/vendor/bundle/jruby/3.1.0/gems/logstash-codec-json_lines-3.2.2/lib/logstash/codecs/json_lines.rb:69)", "org.jruby.internal.runtime.methods.CompiledIRMethod.call(CompiledIRMethod.java:165)", "org.jruby.internal.runtime.methods.MixedModeIRMethod.call(MixedModeIRMethod.java:185)",
```
The issue was fixed in https://github.com/elastic/logstash/pull/16760.

This problem is most likely to be seen when you are using the <<plugins-integrations-logstash,{ls} integration>> plugin to ship data between two {ls} instances, but may appear in other situations, too.

**Workaround for {ls}-to-{ls} communication**

When upgrading Logstash to 8.12.0 or earlier, if the `jvm.options` was previously modified and not overwritten with the newest version,
users may see a "deserialize invocation error" message causing the pipeline to crash. Users are affected if the Persistent Queue
is enabled and the pipeline is processing messages larger than 20MB.
The solution is to apply the default change contained in the newer 'jvm.options' file, as seen https://github.com/elastic/logstash/blob/main/config/jvm.options#L66-L82[here].
The {ls}-to-{ls} issue can be mitigated by:

* Downgrading the _receiving_ {ls} to `8.16.2`, or any {ls} in the `8.15` series, **_AND/OR_**
* Upgrading the {ls} integration filter of the _sending_ {ls} to version `1.0.4`.
Copy link
Contributor Author

@karenzone karenzone Dec 21, 2024

Choose a reason for hiding this comment

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

Suggested change
* Upgrading the {ls} integration filter of the _sending_ {ls} to version `1.0.4`.
* Upgrading the <<plugins-integrations-logstash,{ls} integration>> plugin of the _sending_ {ls} to version `1.0.4`.

This should be the Logstash integration plugin, right? The agent_integration filter doesn't make sense to me.

Note to self @karenzone: Make sure any change here gets propagated to 8.16.1 and 8.16.0 sections.


**Workaround for other `json_lines` codec situations**

Other `json_lines` codec issues can be mitigated by:

* Downgrading {ls} to `8.16.2`, or any {ls} in the `8.15` series.

[[notable-8-17-0]]
==== Notable fixes and improvements
Expand Down Expand Up @@ -119,6 +146,33 @@ The solution is to apply the default change contained in the newer 'jvm.options'
[[logstash-8-16-1]]
=== Logstash 8.16.1 Release Notes

[[known-issues-8-16-1]]
==== Known issue

[[known-issue-8-16-1-json_lines]]
===== "Input buffer full" error with {ls} 8.16.0, 8.16.1, or 8.17.0
Copy link
Contributor Author

Choose a reason for hiding this comment

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

It seems kinda weird to have 8.17.0 listed in the 8.16.1 and 8.16.0 release notes.
However, this information is complete and correct, and I'm inclined to leave it. LMKWYT.

Note to self @karenzone: Make sure any change here gets propagated to 8.16.1 and 8.16.0 sections.


If you are using `json_lines` codec 3.2.0 (or later) with {ls} 8.16.0, 8.16.1, or 8.17.0, you may see an error similar to this one, crashing the pipelines:
```
unable to process event. {:message=>"input buffer full", :class=>"Java::JavaLang::IllegalStateException", :backtrace=>["org.logstash.common.BufferedTokenizerExt.extract(BufferedTokenizerExt.java:83)", "usr.share.logstash.vendor.bundle.jruby.$3_dot_1_dot_0.gems.logstash_minus_codec_minus_json_lines_minus_3_dot_2_dot_2.lib.logstash.codecs.json_lines.RUBY$method$decode$0(/usr/share/logstash/vendor/bundle/jruby/3.1.0/gems/logstash-codec-json_lines-3.2.2/lib/logstash/codecs/json_lines.rb:69)", "org.jruby.internal.runtime.methods.CompiledIRMethod.call(CompiledIRMethod.java:165)", "org.jruby.internal.runtime.methods.MixedModeIRMethod.call(MixedModeIRMethod.java:185)",
```
The issue was fixed in https://github.com/elastic/logstash/pull/16760.

This problem is most likely to be seen when you are using the <<plugins-integrations-logstash,{ls} integration>> plugin to ship data between two {ls} instances, but may appear in other situations, too.

**Workaround for {ls}-to-{ls} communication**

The {ls}-to-{ls} issue can be mitigated by:

* Downgrading the _receiving_ {ls} to `8.16.2`, or any {ls} in the `8.15` series, **_AND/OR_**
* Upgrading the {ls} integration filter of the _sending_ {ls} to version `1.0.4`.

**Workaround for other `json_lines` codec situations**

Other `json_lines` codec issues can be mitigated by:

* Downgrading {ls} to `8.16.2`, or any {ls} in the `8.15` series.

[[notable-8-16-1]]
==== Notable issues fixed

Expand Down Expand Up @@ -155,14 +209,41 @@ The solution is to apply the default change contained in the newer 'jvm.options'
=== Logstash 8.16.0 Release Notes

[[known-issues-8-16-0]]
==== Known issue
==== Known issues

**{ls} may fail to shut down under some circumstances when you are using <<pipeline-to-pipeline>>.**
[[known-issue-8-16-0-shutdown-failure]]
===== {ls} may fail to shut down under some circumstances
{ls} may fail to shut down when you are using <<pipeline-to-pipeline>>.
Check out issue https://github.com/elastic/logstash/issues/16657[#16657] for details.

Workaround: Add `-Dlogstash.pipelinebus.implementation=v1` to `config/jvm.options`.
This change reverts the `PipelineBus` to `v1`, a version that does not exhibit this issue, but may impact performance in pipeline-to-pipeline scenarios.

[[known-issue-8-16-0-json_lines]]
===== "Input buffer full" error with {ls} 8.16.0, 8.16.1, or 8.17.0

If you are using `json_lines` codec 3.2.0 (or later) with {ls} 8.16.0, 8.16.1, or 8.17.0, you may see an error similar to this one, crashing the pipelines:
```
unable to process event. {:message=>"input buffer full", :class=>"Java::JavaLang::IllegalStateException", :backtrace=>["org.logstash.common.BufferedTokenizerExt.extract(BufferedTokenizerExt.java:83)", "usr.share.logstash.vendor.bundle.jruby.$3_dot_1_dot_0.gems.logstash_minus_codec_minus_json_lines_minus_3_dot_2_dot_2.lib.logstash.codecs.json_lines.RUBY$method$decode$0(/usr/share/logstash/vendor/bundle/jruby/3.1.0/gems/logstash-codec-json_lines-3.2.2/lib/logstash/codecs/json_lines.rb:69)", "org.jruby.internal.runtime.methods.CompiledIRMethod.call(CompiledIRMethod.java:165)", "org.jruby.internal.runtime.methods.MixedModeIRMethod.call(MixedModeIRMethod.java:185)",
```
The issue was fixed in https://github.com/elastic/logstash/pull/16760.

This problem is most likely to be seen when you are using the <<plugins-integrations-logstash,{ls} integration>> plugin to ship data between two {ls} instances, but may appear in other situations, too.

**Workaround for {ls}-to-{ls} communication**

The {ls}-to-{ls} issue can be mitigated by:

* Downgrading the _receiving_ {ls} to `8.16.2`, or any {ls} in the `8.15` series, **_AND/OR_**
* Upgrading the {ls} integration filter of the _sending_ {ls} to version `1.0.4`.

**Workaround for other `json_lines` codec situations**

Other `json_lines` codec issues can be mitigated by:

* Downgrading {ls} to `8.16.2`, or any {ls} in the `8.15` series.


[[health-api-8-16-0]]
==== Announcing the new {ls} Health Report API

Expand Down
Loading