[JBoss JIRA] (JGRP-1989) Bundlers: reuse send buffer when transport == sync.
by Bela Ban (JIRA)
[ https://issues.jboss.org/browse/JGRP-1989?page=com.atlassian.jira.plugin.... ]
Bela Ban resolved JGRP-1989.
----------------------------
Resolution: Done
> Bundlers: reuse send buffer when transport == sync.
> ---------------------------------------------------
>
> Key: JGRP-1989
> URL: https://issues.jboss.org/browse/JGRP-1989
> Project: JGroups
> Issue Type: Enhancement
> Reporter: Bela Ban
> Assignee: Bela Ban
> Fix For: 3.6.7
>
>
> With the addition of {{TCP_NIO2}}, all bundlers now create new send buffers for every message (or message list). This generates a lot of memory allocations, perhaps it is better to revert this change for *synchronous transports* such as {{UDP}} and {{TCP}}, and still create new buffers for *asynchronous transports* such as {{TCP_NIO2}}.
> Synchronous transports guarantee a message has been put on the wire when {{TP.send()}} returns, whereas asynchronous transports may only have completed a partial write (so we cannot reuse the buffer).
> The code in the bundler should check for this, and copy if async or not copy if sync.
> Whether or not a transport is sync is determined by a new abstract method that needs to be overridden by every transport.
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
10 years, 7 months
[JBoss JIRA] (JGRP-1989) Bundlers: reuse send buffer when transport == sync.
by Bela Ban (JIRA)
[ https://issues.jboss.org/browse/JGRP-1989?page=com.atlassian.jira.plugin.... ]
Bela Ban commented on JGRP-1989:
--------------------------------
OK, so the continuation of this issue is JGRP-1998, which looks into reusing buffers on the receive side.
Preliminary results with JMC and UPerf for 2M messages in a cluster of 4 shows a reduction of allocated memory from *16.51GB* to *10.94*!
Performance of UPerf didn't increase though, but apps running on JGroups might benefit from this reduction.
> Bundlers: reuse send buffer when transport == sync.
> ---------------------------------------------------
>
> Key: JGRP-1989
> URL: https://issues.jboss.org/browse/JGRP-1989
> Project: JGroups
> Issue Type: Enhancement
> Reporter: Bela Ban
> Assignee: Bela Ban
> Fix For: 3.6.7
>
>
> With the addition of {{TCP_NIO2}}, all bundlers now create new send buffers for every message (or message list). This generates a lot of memory allocations, perhaps it is better to revert this change for *synchronous transports* such as {{UDP}} and {{TCP}}, and still create new buffers for *asynchronous transports* such as {{TCP_NIO2}}.
> Synchronous transports guarantee a message has been put on the wire when {{TP.send()}} returns, whereas asynchronous transports may only have completed a partial write (so we cannot reuse the buffer).
> The code in the bundler should check for this, and copy if async or not copy if sync.
> Whether or not a transport is sync is determined by a new abstract method that needs to be overridden by every transport.
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
10 years, 7 months
[JBoss JIRA] (JGRP-1998) Transport: reuse of incoming buffers
by Bela Ban (JIRA)
Bela Ban created JGRP-1998:
------------------------------
Summary: Transport: reuse of incoming buffers
Key: JGRP-1998
URL: https://issues.jboss.org/browse/JGRP-1998
Project: JGroups
Issue Type: Enhancement
Reporter: Bela Ban
Assignee: Bela Ban
Fix For: 3.6.7
Investigate whether buffers can be reused on the receive side. JGRP-1989 already provides this on the send side. Goal: reduction of memory allocation rate.
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
10 years, 7 months
[JBoss JIRA] (WFLY-5758) Getting [Management resource '[(\"subsystem\" => \"undertow\")]' not found] when removing ressources from Undertow
by Montpa Pasteur (JIRA)
[ https://issues.jboss.org/browse/WFLY-5758?page=com.atlassian.jira.plugin.... ]
Montpa Pasteur commented on WFLY-5758:
--------------------------------------
I don't know if this relevant, but when i start my wildfly and run my script to get the list of datasources, i have to wait like a minute to get this list, before that i get an empty list.
> Getting [Management resource '[(\"subsystem\" => \"undertow\")]' not found] when removing ressources from Undertow
> ------------------------------------------------------------------------------------------------------------------
>
> Key: WFLY-5758
> URL: https://issues.jboss.org/browse/WFLY-5758
> Project: WildFly
> Issue Type: Bug
> Components: CLI
> Affects Versions: 8.x.x TBD
> Environment: Wildfly 8.2.1 on SUSE Linux 11.1 and Solaris 11
> Reporter: Montpa Pasteur
> Assignee: Alexey Loubyansky
> Priority: Minor
>
> When trying remove more that one resource from Undertow using the CLI, I am getting this message: *[Management resource '[(\"subsystem\" => \"undertow\")]' not found*]
> This doesn't happen all time, it looks like random. Adding a sleep between the CLI command helps some time.
> I get this problem when trying to remove http request header configuration and when trying to remove static ressource serving configuration(location, file handler, filter ..)
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
10 years, 7 months
[JBoss JIRA] (JGRP-1997) Bundler: find new default bundler
by Bela Ban (JIRA)
Bela Ban created JGRP-1997:
------------------------------
Summary: Bundler: find new default bundler
Key: JGRP-1997
URL: https://issues.jboss.org/browse/JGRP-1997
Project: JGroups
Issue Type: Task
Reporter: Bela Ban
Assignee: Bela Ban
Fix For: 4.0
Before removing the deprecated sender-sends-with-timer bundler, assess which bundler is the fastest across a range of applications (UPerf, Infinispan etc).
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
10 years, 7 months
[JBoss JIRA] (WFLY-5758) Getting [Management resource '[(\"subsystem\" => \"undertow\")]' not found] when removing ressources from Undertow
by Montpa Pasteur (JIRA)
[ https://issues.jboss.org/browse/WFLY-5758?page=com.atlassian.jira.plugin.... ]
Montpa Pasteur commented on WFLY-5758:
--------------------------------------
No, I am using a python script that call jboss-cli.sh with the paramters. Each command executed is basically a call to jboss-cli.sh with the name of a file containing the command to execute. The use of a sleep between commands seems to help sometimes but .. sometimes not. My python script calls the commands sequentially.
I am having this problem when I try to remove more than one part of the config, right now i am having the problem when i try to remove RBAC authentification ressources.
It looks like wildfly have some problem refreshing its config. I don't know if this is relevant but the execution of these commands is really slow.
I tried the same command set on my Windows machine but i can't reproduce the problem.
> Getting [Management resource '[(\"subsystem\" => \"undertow\")]' not found] when removing ressources from Undertow
> ------------------------------------------------------------------------------------------------------------------
>
> Key: WFLY-5758
> URL: https://issues.jboss.org/browse/WFLY-5758
> Project: WildFly
> Issue Type: Bug
> Components: CLI
> Affects Versions: 8.x.x TBD
> Environment: Wildfly 8.2.1 on SUSE Linux 11.1 and Solaris 11
> Reporter: Montpa Pasteur
> Assignee: Alexey Loubyansky
> Priority: Minor
>
> When trying remove more that one resource from Undertow using the CLI, I am getting this message: *[Management resource '[(\"subsystem\" => \"undertow\")]' not found*]
> This doesn't happen all time, it looks like random. Adding a sleep between the CLI command helps some time.
> I get this problem when trying to remove http request header configuration and when trying to remove static ressource serving configuration(location, file handler, filter ..)
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
10 years, 7 months
[JBoss JIRA] (WFCORE-1232) Adjust patching tests to use merged patches
by Jan Martiska (JIRA)
[ https://issues.jboss.org/browse/WFCORE-1232?page=com.atlassian.jira.plugi... ]
Jan Martiska moved JBEAP-2430 to WFCORE-1232:
---------------------------------------------
Project: WildFly Core (was: JBoss Enterprise Application Platform)
Key: WFCORE-1232 (was: JBEAP-2430)
Workflow: GIT Pull Request workflow (was: CDW v1)
Component/s: Patching
Test Suite
(was: Patching)
(was: Test Suite)
Target Release: (was: 7.0.0.GA)
Affects Version/s: 2.0.4.Final
(was: 7.0.0.ER3)
> Adjust patching tests to use merged patches
> -------------------------------------------
>
> Key: WFCORE-1232
> URL: https://issues.jboss.org/browse/WFCORE-1232
> Project: WildFly Core
> Issue Type: Bug
> Components: Patching , Test Suite
> Affects Versions: 2.0.4.Final
> Reporter: Jan Martiska
> Assignee: Jan Martiska
>
> WFCORE-838 introduced a new way of joining multiple patches together - merging them into one. This needs some test coverage in patching TS.
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
10 years, 7 months
[JBoss JIRA] (WFCORE-1231) Remove OSGi bundle related patching tests
by Jan Martiska (JIRA)
Jan Martiska created WFCORE-1231:
------------------------------------
Summary: Remove OSGi bundle related patching tests
Key: WFCORE-1231
URL: https://issues.jboss.org/browse/WFCORE-1231
Project: WildFly Core
Issue Type: Bug
Components: Patching , Test Suite
Affects Versions: 2.0.4.Final
Reporter: Jan Martiska
Assignee: Jan Martiska
Priority: Minor
OSGi is no longer supported, there's no point in keeping these tests.
This affects unit tests in the 'patching' module and the patching module of integration TS.
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)
10 years, 7 months