July 2011 - mod_cluster-issues - Jboss List Archives

[JBoss JIRA] Created: (MODCLUSTER-151) Modify jbossweb metrics to use service provider spi, instead of jmx

by Paul Ferraro (JIRA)

Modify jbossweb metrics to use service provider spi, instead of jmx ------------------------------------------------------------------- Key: MODCLUSTER-151 URL: https://jira.jboss.org/jira/browse/MODCLUSTER-151 Project: mod_cluster Issue Type: Feature Request Affects Versions: 1.1.0.CR1 Reporter: Paul Ferraro Assignee: Paul Ferraro Currently, the jbossweb load metrics (i.e. ActiveSessionsLoadMetric, BusyConnectorsLoadMetric, RequestCountLoadMetric, ReceiveTrafficLoadMetric, SendTrafficLoadMetric) use jmx to generate their load values. This is potentially fragile. Instead, these load metrics should use org.jboss.mod_cluster.Engine as a load context. This raises the issue of load value scope. Currently, load is scoped to a server. Really, this should be scoped to an engine. While server:engine is usually a 1:1 relationship, this is technically a 1:N relationship. Suggested API change: class LoadMetricSource<C extends LoadContext> { C createContext(Engine engine); } Where there exists: class EngineLoadMetricSource<EngineLoadContext> { public EngineLoadContext createContext(Engine engine) { return new EngineLoadContext(engine); } } class EngineLoadContext implements LoadContext { private final Engine engine; public EngineLoadContext(Engine engine) { this.engine = engine; } public Engine getEngine() { return this.engine; } public void close() { // Nothing to close } } The various jbossweb load metrics would use this source. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

13 years

6
6
0 / 0

[JBoss JIRA] Created: (MODCLUSTER-105) Refactor mod_cluster to allow implementations for other application servers

by Paul Ferraro (JIRA)

Refactor mod_cluster to allow implementations for other application servers --------------------------------------------------------------------------- Key: MODCLUSTER-105 URL: https://jira.jboss.org/jira/browse/MODCLUSTER-105 Project: mod_cluster Issue Type: Feature Request Affects Versions: 1.0.2.GA Reporter: Paul Ferraro Assignee: Paul Ferraro Split current jar into a -core.jar containing APIs, and isolate catalina implementation to a separate -catalina.jar. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

13 years

2
1
0 / 0

[JBoss JIRA] Created: (MODCLUSTER-243) mod_cluster throws warnings at shutdown in AS7

by Jean-Frederic Clere (JIRA)

mod_cluster throws warnings at shutdown in AS7 ---------------------------------------------- Key: MODCLUSTER-243 URL: https://issues.jboss.org/browse/MODCLUSTER-243 Project: mod_cluster Issue Type: Bug Affects Versions: 1.1.2.Final Reporter: Jean-Frederic Clere Assignee: Jean-Frederic Clere Priority: Minor Fix For: 1.1.3.Final At shutdown you get an warning: +++ 21:37:37,537 WARN [org.jboss.modcluster.advertise.impl.AdvertiseListenerImpl] (MSC service thread 1-1) Failed to interrupt socket reception: java.io.IOException: Invalid argument at java.net.PlainDatagramSocketImpl.send(Native Method) [:1.6.0_22] at java.net.DatagramSocket.send(DatagramSocket.java:629) [:1.6.0_22] at org.jboss.modcluster.advertise.impl.AdvertiseListenerImpl.interruptDatagramReader(AdvertiseListenerImpl.java:240) at org.jboss.modcluster.advertise.impl.AdvertiseListenerImpl.stop(AdvertiseListenerImpl.java:262) at org.jboss.modcluster.advertise.impl.AdvertiseListenerImpl.destroy(AdvertiseListenerImpl.java:278) at org.jboss.modcluster.ModClusterService.shutdown(ModClusterService.java:214) at org.jboss.modcluster.catalina.CatalinaEventHandlerAdapter.destroy(CatalinaEventHandlerAdapter.java:374) at org.jboss.modcluster.catalina.CatalinaEventHandlerAdapter.stop(CatalinaEventHandlerAdapter.java:144) at org.jboss.as.modcluster.ModClusterService.stop(ModClusterService.java:236) at org.jboss.msc.service.ServiceControllerImpl$StopTask.run(ServiceControllerImpl.java:1819) at org.jboss.msc.service.ServiceControllerImpl$ClearTCCLTask.run(ServiceControllerImpl.java:2302) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1110) [:1.6.0_22] at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:603) [:1.6.0_22] at java.lang.Thread.run(Thread.java:679) [:1.6.0_22] +++ In fact that should be a debug... And the fix should go in mod_cluster. -- This message is automatically generated by JIRA. For more information on JIRA, see: http://www.atlassian.com/software/jira

13 years

3
4
0 / 0

[JBoss JIRA] Created: (MODCLUSTER-219) AS7 schema requires properties.

by Jean-Frederic Clere (JIRA)

AS7 schema requires properties. ------------------------------- Key: MODCLUSTER-219 URL: https://issues.jboss.org/browse/MODCLUSTER-219 Project: mod_cluster Issue Type: Feature Request Affects Versions: 1.1.1.Final Reporter: Jean-Frederic Clere Assignee: Paul Ferraro Fix For: 1.1.2.Final In the xsd we have: +++ <xs:complexType name="common-load-metricType" abstract="true"> <xs:sequence> <xs:element name="property" type="tns:propertyType" minOccurs="0" maxOccurs="unbounded"/> </xs:sequence> <xs:attribute name="weight" type="xs:int"/> <xs:attribute name="capacity" type="xs:double"/> </xs:complexType> +++ So a getProperties() and a setProperties() should be added AbstractLoadMetric for example? -- This message is automatically generated by JIRA. For more information on JIRA, see: http://www.atlassian.com/software/jira

13 years

3
7
0 / 0

[JBoss JIRA] Created: (MODCLUSTER-209) mod_cluster_manager need better presentation

by Anders Welen (JIRA)

mod_cluster_manager need better presentation -------------------------------------------- Key: MODCLUSTER-209 URL: https://issues.jboss.org/browse/MODCLUSTER-209 Project: mod_cluster Issue Type: Enhancement Affects Versions: 1.1.1.Final Reporter: Anders Welen Assignee: Jean-Frederic Clere Priority: Minor The mod_cluster_manager "show INFO output" status page has the following problems: 1. "Transferred" misspelled 2. Context is missing the number of requests PS. Why not make it easier to parse? look at mod_jk that support both properties and XML DS. -- This message is automatically generated by JIRA. - For more information on JIRA, see: http://www.atlassian.com/software/jira

13 years, 3 months

2
1
0 / 0

[JBoss JIRA] Created: (MODCLUSTER-242) requests routed to nodes where the context doesn't exist

by Samuel Mendenhall (JIRA)

requests routed to nodes where the context doesn't exist -------------------------------------------------------- Key: MODCLUSTER-242 URL: https://issues.jboss.org/browse/MODCLUSTER-242 Project: mod_cluster Issue Type: Bug Affects Versions: 1.0.10 Reporter: Samuel Mendenhall Assignee: Jean-Frederic Clere Fix For: 1.0.11 Env: 4 JBoss servers w/ mod_cluster sitting behind 2 Apache servers. The 2 Apache servers are proxying to the 4 JBoss servers. Those 4 JBoss servers are split into 2 separate clusters, each cluster serving different applications. For example, Cluster1 has AppA and AppB, but Cluster2 has AppC and Appd. When attempting to connect to any application through the Apache server, we randomly see 404 errors. Is this because Apache/mod_cluster is trying to route the request to all servers, instead of just the servers with the right web context enabled. -- This message is automatically generated by JIRA. For more information on JIRA, see: http://www.atlassian.com/software/jira

13 years, 4 months

3
4
0 / 0

[JBoss JIRA] Created: (MODCLUSTER-220) Extra REMOVE-APP event after failover

by Radoslav Husar (JIRA)

Extra REMOVE-APP event after failover ------------------------------------- Key: MODCLUSTER-220 URL: https://issues.jboss.org/browse/MODCLUSTER-220 Project: mod_cluster Issue Type: Enhancement Reporter: Radoslav Husar Assignee: Jean-Frederic Clere Priority: Trivial Trivial thing: I think there is an extra REMOVE-APP event happening after fail-over. The test scenario is simple - start a node with 6 context, kill the node and reconnect. The apps are unchanged but 5 are remembered as were and one is removed and added back? Since they happen at the same time it looks like it happens uselessly and can cause race condition. {code} 127.0.0.1 - - [04/Apr/2011:19:18:34 +0200] "INFO / HTTP/1.0" 200 - "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:18:34 +0200] "CONFIG / HTTP/1.0" 200 - "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:18:34 +0200] "ENABLE-APP / HTTP/1.0" 200 - "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:18:34 +0200] "ENABLE-APP / HTTP/1.0" 200 - "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:18:34 +0200] "ENABLE-APP / HTTP/1.0" 200 - "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:18:34 +0200] "ENABLE-APP / HTTP/1.0" 200 - "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:18:34 +0200] "ENABLE-APP / HTTP/1.0" 200 - "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:18:34 +0200] "ENABLE-APP / HTTP/1.0" 200 - "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:18:34 +0200] "STATUS / HTTP/1.0" 200 58 "-" "ClusterListener/1.0" ... 127.0.0.1 - - [04/Apr/2011:19:19:03 +0200] "GET /SessionTest/SessionTestServlet? HTTP/1.1" 200 1 "-" "Jakarta Commons-HttpClient/3.1" ... 127.0.0.1 - - [04/Apr/2011:19:20:44 +0200] "INFO / HTTP/1.0" 200 575 "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:20:44 +0200] "CONFIG / HTTP/1.0" 200 - "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:20:44 +0200] "ENABLE-APP / HTTP/1.0" 200 - "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:20:44 +0200] "REMOVE-APP / HTTP/1.0" 200 - "-" "ClusterListener/1.0" 127.0.0.1 - - [04/Apr/2011:19:20:44 +0200] "STATUS / HTTP/1.0" 200 58 "-" "ClusterListener/1.0" {code} -- This message is automatically generated by JIRA. For more information on JIRA, see: http://www.atlassian.com/software/jira

13 years, 5 months

3
21
0 / 0

[JBoss JIRA] Created: (MODCLUSTER-241) kill -HUP httpd process increase the open locks.

by Jean-Frederic Clere (JIRA)

kill -HUP httpd process increase the open locks. ------------------------------------------------ Key: MODCLUSTER-241 URL: https://issues.jboss.org/browse/MODCLUSTER-241 Project: mod_cluster Issue Type: Bug Affects Versions: 1.1.2.Final , 1.0.10 Reporter: Jean-Frederic Clere Assignee: Jean-Frederic Clere Fix For: 1.1.3.Final start httpd and kill -HUP the process. -- This message is automatically generated by JIRA. For more information on JIRA, see: http://www.atlassian.com/software/jira

13 years, 6 months

1
3
0 / 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

mod_cluster-issues July 2011