[JBoss JIRA] (ISPN-6346) Zipkin based cluster-wide request tracing
by Tristan Tarrant (JIRA)
[ https://issues.jboss.org/browse/ISPN-6346?page=com.atlassian.jira.plugin.... ]
Tristan Tarrant updated ISPN-6346:
----------------------------------
Fix Version/s: (was: 9.0.0.Final)
> Zipkin based cluster-wide request tracing
> -----------------------------------------
>
> Key: ISPN-6346
> URL: https://issues.jboss.org/browse/ISPN-6346
> Project: Infinispan
> Issue Type: Feature Request
> Components: JMX, reporting and management
> Reporter: Galder Zamarreño
> Assignee: Galder Zamarreño
> Priority: Minor
>
> Zipkin (http://zipkin.io) is a distributed tracing system that started at Twitter.
> Infinispan should integrate with it so that we can trace individual cache requests in terms of which components it touches and how long it takes in each, and also when the cache requests end up going to other nodes (plus see which remote components it touches).
> The idea is that this should give us a better idea of what a particular operation does in terms of how many nodes it goes to, and what it does in each. This would be hugely beneficial to be able to better analyse performance issues.
> A key advantage of Zipkin is that it can be run in production since it can be configured to only track a fraction of all requests, making it even more valuable from the monitoring POV.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
7 years, 2 months
[JBoss JIRA] (ISPN-6494) Investigate bundler performance
by Dan Berindei (JIRA)
[ https://issues.jboss.org/browse/ISPN-6494?page=com.atlassian.jira.plugin.... ]
Dan Berindei commented on ISPN-6494:
------------------------------------
JGroups 4.0 no longer has {{sender-sends-with-timer}}, but {{no-bundler}} (the equivalent of {{sender-sends-with-timer}} + {{DONT_BUNDLE}}) is still much better than {{transfer-queue}} in some tests.
> Investigate bundler performance
> -------------------------------
>
> Key: ISPN-6494
> URL: https://issues.jboss.org/browse/ISPN-6494
> Project: Infinispan
> Issue Type: Task
> Components: Core
> Affects Versions: 9.0.0.Alpha1
> Reporter: Dan Berindei
> Assignee: Dan Berindei
>
> For ISPN-6027 we changed the default JGroups bundler to {{sender-sends-with-timer}}, because it was faster in some of the performance tests. However, IspnPerfTest shows {{transfer-queue-bundler}} to be consistently better, so we need to investigate the bundler choice again.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
7 years, 2 months
[JBoss JIRA] (ISPN-6494) Investigate bundler performance
by Tristan Tarrant (JIRA)
[ https://issues.jboss.org/browse/ISPN-6494?page=com.atlassian.jira.plugin.... ]
Tristan Tarrant updated ISPN-6494:
----------------------------------
Fix Version/s: (was: 9.0.0.Final)
(was: 9.0.0.Alpha2)
> Investigate bundler performance
> -------------------------------
>
> Key: ISPN-6494
> URL: https://issues.jboss.org/browse/ISPN-6494
> Project: Infinispan
> Issue Type: Task
> Components: Core
> Affects Versions: 9.0.0.Alpha1
> Reporter: Dan Berindei
> Assignee: Dan Berindei
>
> For ISPN-6027 we changed the default JGroups bundler to {{sender-sends-with-timer}}, because it was faster in some of the performance tests. However, IspnPerfTest shows {{transfer-queue-bundler}} to be consistently better, so we need to investigate the bundler choice again.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
7 years, 2 months
[JBoss JIRA] (ISPN-6478) Netty worker thread starvation with events
by Tristan Tarrant (JIRA)
[ https://issues.jboss.org/browse/ISPN-6478?page=com.atlassian.jira.plugin.... ]
Tristan Tarrant updated ISPN-6478:
----------------------------------
Priority: Critical (was: Blocker)
> Netty worker thread starvation with events
> ------------------------------------------
>
> Key: ISPN-6478
> URL: https://issues.jboss.org/browse/ISPN-6478
> Project: Infinispan
> Issue Type: Bug
> Components: Server
> Affects Versions: 8.2.1.Final
> Reporter: Galder Zamarreño
> Assignee: Galder Zamarreño
> Priority: Critical
>
> As a result of ISPN-6005, we decoupled the incoming Hot Rod server invocations from sending events by adding a intermediate queue that maintains the events to send to clients. However, this separation can lead to a Netty worker thread starvation issue since we add the events to the queue in Netty's IO thread, so if the queue is full, incoming requests are stuck.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
7 years, 2 months
[JBoss JIRA] (ISPN-6478) Netty worker thread starvation with events
by Tristan Tarrant (JIRA)
[ https://issues.jboss.org/browse/ISPN-6478?page=com.atlassian.jira.plugin.... ]
Tristan Tarrant updated ISPN-6478:
----------------------------------
Fix Version/s: (was: 9.0.0.Final)
(was: 9.0.0.CR3)
> Netty worker thread starvation with events
> ------------------------------------------
>
> Key: ISPN-6478
> URL: https://issues.jboss.org/browse/ISPN-6478
> Project: Infinispan
> Issue Type: Bug
> Components: Server
> Affects Versions: 8.2.1.Final
> Reporter: Galder Zamarreño
> Assignee: Galder Zamarreño
> Priority: Blocker
>
> As a result of ISPN-6005, we decoupled the incoming Hot Rod server invocations from sending events by adding a intermediate queue that maintains the events to send to clients. However, this separation can lead to a Netty worker thread starvation issue since we add the events to the queue in Netty's IO thread, so if the queue is full, incoming requests are stuck.
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)
7 years, 2 months