August 2014 - mod_cluster-issues - Jboss List Archives

[JBoss JIRA] (MODCLUSTER-199) Segmentation fault in Apache worker when using different balancernames and Aliases.

by Michal Babacek (JIRA)

[ https://issues.jboss.org/browse/MODCLUSTER-199?page=com.atlassian.jira.pl... ] Michal Babacek closed MODCLUSTER-199. ------------------------------------- Closing. Clean-up. At least one of the following applies: * the issue has been thoroughly tested as a part of one of the current releases or * it hasn't occurred in ~2 years or * it's utterly harmless > Segmentation fault in Apache worker when using different balancernames and Aliases. > ----------------------------------------------------------------------------------- > > Key: MODCLUSTER-199 > URL: https://issues.jboss.org/browse/MODCLUSTER-199 > Project: mod_cluster > Issue Type: Bug > Security Level: Public(Everyone can see) > Affects Versions: 1.1.0.Final > Environment: RedHat EL 5.5 x86_64, Apache 2.2.3, mod_cluster-1.1.0.Final-linux2-x64 > Reporter: Geert Pante > Assignee: Jean-Frederic Clere > Attachments: core.zip > > > Hi all, > We did a test setup with 2 apache servers and 2 backend JBoss servers. One JBoss Server has 2 applications deployed on one alias, and the other one has one application deployed as ROOT application context. In the jbossweb server.xml we set Aliases, and this worked most of the time, with applications http://alias1/app1, http://alias1/app2 and http://alias2/ all proxied nicely through both apache servers. The names alias1 and alias2 were mapped to a virtual IP that could be switched dynamically between the apache servers. However, we did have some problems that after a restart of one of the apache server, it would get 'All workers are in error state' for one of the applications, and it would no longer send ajp ping requests. The JBoss server continues to send STATUS messages happily. In mod_cluster-manager all seems well, too. > Initially we used a default ManagerBalancerName in the Apache config, so both were using the same balancer. As a test we tried to remove the ManagerBalancerName in the apache config, and set a balancername in the mod_cluster-beans.xml. In mod_cluster-manager we saw the different balancers now, but again, on one of the apache servers, the proxying does not work, this time with a segmentation fault. > See attached coredump > =============================================== > Core was generated by `/usr/sbin/httpd'. > Program terminated with signal 11, Segmentation fault. > (gdb) thread apply all bt full > Thread 2 (Thread 31682): > #0 0x00002b81583146f7 in kill () from /lib64/libc.so.6 > No symbol table info available. > #1 <signal handler called> > No symbol table info available. > #2 0x00002b815835d7c0 in strlen () from /lib64/libc.so.6 > No symbol table info available. > #3 0x00002b8157ca840e in ?? () from /usr/lib64/libapr-1.so.0 > No symbol table info available. > #4 0x7069626d00000000 in ?? () > No symbol table info available. > #5 0x00002b815f9755c8 in ?? () > No symbol table info available. > #6 0x00002b815669e230 in ?? () > No symbol table info available. > #7 0x00002b81566a1e08 in ?? () > No symbol table info available. > #8 0x00002b815f973f90 in ?? () > No symbol table info available. > #9 0x00002b815f32365f in cluster_get_sessionid (r=0x2b815f973f18, > stickyval=0x1 <Address 0x1 out of bounds>, uri=0x2b815f9755c8 "/", > sticky_used=0x7) at mod_proxy_cluster.c:1032 > sticky = <value optimized out> > sticky_path = <value optimized out> > route = <value optimized out> > #10 0x00002b815f324e92 in get_route_balancer (r=0x2b815f973f90, > conf=0x2b815669d1a8) at mod_proxy_cluster.c:1788 > balancer = 0x2b81566a1e08 > route = <value optimized out> > sessionid = <value optimized out> > sticky_used = <value optimized out> > i = 1 > #11 0x00002b815f3251d5 in proxy_cluster_trans (r=0x1) > at mod_proxy_cluster.c:1938 > balancer = 0x2b815f973f90 "\030?\227_\201+" > sconf = 0x2b815f9755c8 > conf = 0x1 > #12 0x00002b81563e4ba2 in ap_run_translate_name () > No symbol table info available. > #13 0x00002b81563e5e62 in ap_process_request_internal () > No symbol table info available. > #14 0x00002b81563f7968 in ap_should_client_block () > No symbol table info available. > #15 0x00002b815f963870 in ?? () > No symbol table info available. > #16 0x00002b815f973f90 in ?? () > No symbol table info available. > #17 0x00002b815f963870 in ?? () > No symbol table info available. > #18 0x0000000000000000 in ?? () > No symbol table info available. > Thread 1 (Thread 31683): > #0 0x00002b81583b0fc2 in select () from /lib64/libc.so.6 > No symbol table info available. -- This message was sent by Atlassian JIRA (v6.2.6#6264)

9 years, 8 months

1
0
0 / 0

[JBoss JIRA] (MODCLUSTER-367) Backport missing commits from 1.2.x branch to master

by Michal Babacek (JIRA)

[ https://issues.jboss.org/browse/MODCLUSTER-367?page=com.atlassian.jira.pl... ] Michal Babacek closed MODCLUSTER-367. ------------------------------------- Closing. Clean-up. At least one of the following applies: * the issue has been thoroughly tested as a part of one of the current releases or * it hasn't occurred in ~2 years or * it's utterly harmless > Backport missing commits from 1.2.x branch to master > ---------------------------------------------------- > > Key: MODCLUSTER-367 > URL: https://issues.jboss.org/browse/MODCLUSTER-367 > Project: mod_cluster > Issue Type: Task > Security Level: Public(Everyone can see) > Affects Versions: 1.3.0.Alpha1 > Environment: all > Reporter: Radoslav Husar > Assignee: Radoslav Husar > Priority: Blocker > Fix For: 1.3.0.Final > > > Since my first attempt to address this via manageable rebase went ignored, this needs to be tediously hand picked (ready in my branch). -- This message was sent by Atlassian JIRA (v6.2.6#6264)

9 years, 8 months

1
0
0 / 0

[JBoss JIRA] (MODCLUSTER-70) openssl doesn't build on platfrom

by Michal Babacek (JIRA)

[ https://issues.jboss.org/browse/MODCLUSTER-70?page=com.atlassian.jira.plu... ] Michal Babacek closed MODCLUSTER-70. ------------------------------------ Closing. Clean-up. At least one of the following applies: * the issue has been thoroughly tested as a part of one of the current releases or * it hasn't occurred in ~2 years or * it's utterly harmless > openssl doesn't build on platfrom > --------------------------------- > > Key: MODCLUSTER-70 > URL: https://issues.jboss.org/browse/MODCLUSTER-70 > Project: mod_cluster > Issue Type: Bug > Security Level: Public(Everyone can see) > Affects Versions: 1.0.0.CR1 > Environment: hudson build > Reporter: Jean-Frederic Clere > Assignee: Jean-Frederic Clere > Fix For: 1.0.0.CR2 > > > Since the change to the lastest version openssl the ssl is not build on most of the supported plaforms -- This message was sent by Atlassian JIRA (v6.2.6#6264)

9 years, 8 months

1
0
0 / 0

[JBoss JIRA] (MODCLUSTER-261) segmentation fault with nofailover=On and proxy in virtual host

by Michal Babacek (JIRA)

[ https://issues.jboss.org/browse/MODCLUSTER-261?page=com.atlassian.jira.pl... ] Michal Babacek closed MODCLUSTER-261. ------------------------------------- Closing. Clean-up. At least one of the following applies: * the issue has been thoroughly tested as a part of one of the current releases or * it hasn't occurred in ~2 years or * it's utterly harmless > segmentation fault with nofailover=On and proxy in virtual host > --------------------------------------------------------------- > > Key: MODCLUSTER-261 > URL: https://issues.jboss.org/browse/MODCLUSTER-261 > Project: mod_cluster > Issue Type: Bug > Security Level: Public(Everyone can see) > Affects Versions: 1.1.3.Final > Environment: RHEL 5.7, Apache HTTPD 2.2.3 (packaged) or Apache HTTPD 2.2.21 (compiled), EAP 5.1.1 > Reporter: Tomas Huryn > Assignee: Jean-Frederic Clere > Priority: Blocker > > we have > - two clusters (2 JBoss instances for cluster1 (inst1 and inst2), 1 JBoss instance for cluster2 (inst3)). All instances run on the same machine (as) with binding manager > - AJP on JBoss (port 8009 (inst1), 8109 (inst2), 8209 (inst3)) > - root context enabled, application is in ROOT.war > - defined jvmRoute in server.xml (inst1 -Djboss.web.jvmRoute=as.jb1 , inst2 -Djboss.web.jvmRoute=as.jb2 , inst3 -Djboss.web.jvmRoute=as.cms1) > - Apache HTTPD runs on separate machine (ws) > we need > - cluster1 has connection to inst1 and inst2 with sticky session (client with one session will be routed to one instance only) > - cluster2 has connection to inst3 > We had tryed 2 solutions with no success: > SOLUTION 1 > - all in virtual host > <VirtualHost externalip:8080> > ServerName web1 > <Proxy balancer://cluster1/> > Order deny,allow > Allow from all > BalancerMember ajp://as:8009/ route=as.jb1 > BalancerMember ajp://as:8109/ route=as.jb2 > </Proxy> > ProxyPass / balancer://cluster1/ stickysession=JSESSIONID|jsessionid nofailover=On > ProxyPassReverse / balancer://cluster1/ > </VirtualHost> > <VirtualHost externalip:8080> > ServerName web2 > <Proxy balancer://cluster2/> > Order deny,allow > Allow from all > BalancerMember ajp://as:8209/ route=as.cms1 > </Proxy> > ProxyPass / balancer://cluster2/ stickysession=JSESSIONID|jsessionid nofailover=On > ProxyPassReverse / balancer://cluster2/ > </VirtualHost> > Result: > Solution generate segmentation fault in error log, where somebody try to browse web. If we remove nofailover=On, all works fine except stick session - requests are splited to inst1 and inst2. > SOLUTION2 > - apache wide configuration > <Proxy balancer://cluster1/> > Order deny,allow > Allow from all > BalancerMember ajp://as:8009/ route=as.jb1 > BalancerMember ajp://as:8109/ route=as.jb2 > </Proxy> > <Proxy balancer://cluster2/> > Order deny,allow > Allow from all > BalancerMember ajp://as:8209/ route=as.cms1 > </Proxy> > <VirtualHost externalip:8080> > ServerName web1 > ProxyPass / balancer://cluster1/ stickysession=JSESSIONID|jsessionid nofailover=On > ProxyPassReverse / balancer://cluster1/ > </VirtualHost> > <VirtualHost externalip:8080> > ServerName web2 > ProxyPass / balancer://cluster2/ stickysession=JSESSIONID|jsessionid nofailover=On > ProxyPassReverse / balancer://cluster2/ > </VirtualHost> > Result: > Sticky session works fine (no segmentation fault), but requests for web1 is routed to inst3 (cluster2) too and requests for web2 is routed to inst1 (cluster1) too. So there is problem with existence of two proxy cluster conffiguration directive. -- This message was sent by Atlassian JIRA (v6.2.6#6264)

9 years, 8 months

1
0
0 / 0

[JBoss JIRA] (MODCLUSTER-189) Enable/Disable does not work in mod_cluster_manager

by Michal Babacek (JIRA)

[ https://issues.jboss.org/browse/MODCLUSTER-189?page=com.atlassian.jira.pl... ] Michal Babacek closed MODCLUSTER-189. ------------------------------------- Closing. Clean-up. At least one of the following applies: * the issue has been thoroughly tested as a part of one of the current releases or * it hasn't occurred in ~2 years or * it's utterly harmless > Enable/Disable does not work in mod_cluster_manager > --------------------------------------------------- > > Key: MODCLUSTER-189 > URL: https://issues.jboss.org/browse/MODCLUSTER-189 > Project: mod_cluster > Issue Type: Bug > Security Level: Public(Everyone can see) > Affects Versions: 1.0.4.GA, 1.1.0.Final > Environment: Linux x86_64 > Reporter: Henrik Aronsen > Assignee: Jean-Frederic Clere > Labels: advertise, mod_cluster, mod_cluster_manager, unicast > > When I disable a context in mod_cluster_manager nothing happens. The context is still available even though it has Status: DISABLED. > I am using mod_cluster 1.1.0 with JBoss EAP 5.1. I've tried mod_cluster 1.0.4 and JBoss EAP 5.0.0 as well. It's a pretty default setup, other than that I don't use multicast: ServerAdvertise is Off and I've set the proxyList in mod_cluster-jboss-beans.xml which points to my Apache frontend. -- This message was sent by Atlassian JIRA (v6.2.6#6264)

9 years, 8 months

1
0
0 / 0

[JBoss JIRA] (MODCLUSTER-252) routing table lookup creates performance issues.

by Michal Babacek (JIRA)

[ https://issues.jboss.org/browse/MODCLUSTER-252?page=com.atlassian.jira.pl... ] Michal Babacek closed MODCLUSTER-252. ------------------------------------- Closing. Clean-up. At least one of the following applies: * the issue has been thoroughly tested as a part of one of the current releases or * it hasn't occurred in ~2 years or * it's utterly harmless > routing table lookup creates performance issues. > ------------------------------------------------ > > Key: MODCLUSTER-252 > URL: https://issues.jboss.org/browse/MODCLUSTER-252 > Project: mod_cluster > Issue Type: Bug > Security Level: Public(Everyone can see) > Affects Versions: MOD_CLUSTER_1_0_10_GA_CP02, 1.1.3.Final > Environment: use ~ 200 workers > Reporter: Jean-Frederic Clere > Assignee: Jean-Frederic Clere > Fix For: 1.2.0.Beta4 > > Attachments: mod_proxy_cluster.patch > > > the throughput goes down to ~ 5 req/s -- This message was sent by Atlassian JIRA (v6.2.6#6264)

9 years, 8 months

1
0
0 / 0

[JBoss JIRA] (MODCLUSTER-19) Investigate auto-registration of JBW listener/dependency

by Michal Babacek (JIRA)

[ https://issues.jboss.org/browse/MODCLUSTER-19?page=com.atlassian.jira.plu... ] Michal Babacek closed MODCLUSTER-19. ------------------------------------ Closing. Clean-up. At least one of the following applies: * the issue has been thoroughly tested as a part of one of the current releases or * it hasn't occurred in ~2 years or * it's utterly harmless > Investigate auto-registration of JBW listener/dependency > -------------------------------------------------------- > > Key: MODCLUSTER-19 > URL: https://issues.jboss.org/browse/MODCLUSTER-19 > Project: mod_cluster > Issue Type: Feature Request > Security Level: Public(Everyone can see) > Affects Versions: 1.0.0.Beta2 > Reporter: Jean-Frederic Clere > Assignee: Paul Ferraro > -- This message was sent by Atlassian JIRA (v6.2.6#6264)

9 years, 8 months

1
0
0 / 0

[JBoss JIRA] (MODCLUSTER-98) Linux mod_cluster_manager display zero instead values

by Michal Babacek (JIRA)

[ https://issues.jboss.org/browse/MODCLUSTER-98?page=com.atlassian.jira.plu... ] Michal Babacek closed MODCLUSTER-98. ------------------------------------ Closing. Clean-up. At least one of the following applies: * the issue has been thoroughly tested as a part of one of the current releases or * it hasn't occurred in ~2 years or * it's utterly harmless > Linux mod_cluster_manager display zero instead values > ----------------------------------------------------- > > Key: MODCLUSTER-98 > URL: https://issues.jboss.org/browse/MODCLUSTER-98 > Project: mod_cluster > Issue Type: Bug > Security Level: Public(Everyone can see) > Affects Versions: 1.0.1.GA > Environment: On 32 bits Linux mod_cluster_manager display zero instead values > Reporter: Jean-Frederic Clere > Assignee: Jean-Frederic Clere > Fix For: 1.0.2.GA, 1.1.0.Beta1 > > > try http://localhost/mod_cluster_manager/ you will see: > Load: 0 for example INFO and DUMP are also wrong. -- This message was sent by Atlassian JIRA (v6.2.6#6264)

9 years, 8 months

1
0
0 / 0

[JBoss JIRA] (MODCLUSTER-131) fix JBPAPP-3463 a clean way.

by Michal Babacek (JIRA)

[ https://issues.jboss.org/browse/MODCLUSTER-131?page=com.atlassian.jira.pl... ] Michal Babacek closed MODCLUSTER-131. ------------------------------------- Closing. Clean-up. At least one of the following applies: * the issue has been thoroughly tested as a part of one of the current releases or * it hasn't occurred in ~2 years or * it's utterly harmless > fix JBPAPP-3463 a clean way. > ---------------------------- > > Key: MODCLUSTER-131 > URL: https://issues.jboss.org/browse/MODCLUSTER-131 > Project: mod_cluster > Issue Type: Feature Request > Security Level: Public(Everyone can see) > Environment: See JBPAPP-3463 > Reporter: Jean-Frederic Clere > Assignee: Paul Ferraro > Fix For: 1.1.0.CR1 > > -- This message was sent by Atlassian JIRA (v6.2.6#6264)

9 years, 8 months

1
0
0 / 0

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

by Michal Babacek (JIRA)

[ https://issues.jboss.org/browse/MODCLUSTER-105?page=com.atlassian.jira.pl... ] Michal Babacek closed MODCLUSTER-105. ------------------------------------- Closing. Clean-up. At least one of the following applies: * the issue has been thoroughly tested as a part of one of the current releases or * it hasn't occurred in ~2 years or * it's utterly harmless > Refactor mod_cluster to allow implementations for other application servers > --------------------------------------------------------------------------- > > Key: MODCLUSTER-105 > URL: https://issues.jboss.org/browse/MODCLUSTER-105 > Project: mod_cluster > Issue Type: Feature Request > Security Level: Public(Everyone can see) > Affects Versions: 1.0.2.GA > Reporter: Paul Ferraro > Assignee: Paul Ferraro > Fix For: 1.2.0.Beta1 > > > Split current jar into a -core.jar containing APIs, and isolate catalina implementation to a separate -catalina.jar. -- This message was sent by Atlassian JIRA (v6.2.6#6264)

9 years, 8 months

1
0
0 / 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

mod_cluster-issues August 2014