Jean-Frederic Clere <jfclere> made a comment on jira MODCLUSTER-335

I need the httpd.conf corresponding to the test and the error_log (with debug) to investigate the problem.

Michal Babacek <mbabacek> made a comment on jira MODCLUSTER-335

These are present in the zips I attached yesterday, sorry for not pointing it out explicitly.
Paths:

[^logs-and-configs-m_c-1.2.4.zip]
{{target-ER6/jboss-reports/logs-archive/noe.ews.modcluster.tests.ModProxyTest/modProxyOnlyPartOfUrlSpaceTest/HttpdRhel-2/ews-httpd/logs/error_log}}

{{target-ER6/jboss-reports/confs-archive/noe.ews.modcluster.tests.ModProxyTest/modProxyOnlyPartOfUrlSpaceTest/HttpdRhel-2/ews-httpd/mod_cluster.conf/mod_cluster.conf}}

{{target-ER6/jboss-reports/confs-archive/noe.ews.modcluster.tests.ModProxyTest/modProxyOnlyPartOfUrlSpaceTest/HttpdRhel-2/ews-httpd/proxy1.conf/proxy1.conf}}

{{target-ER6/jboss-reports/confs-archive/noe.ews.modcluster.tests.ModProxyTest/modProxyOnlyPartOfUrlSpaceTest/HttpdRhel-2/ews-httpd/httpd.conf/httpd.conf}}

Analogically in [^logs-and-configs-m_c-1.2.3.zip] but for having "target-ER5" instead of "target-ER6".

Jean-Frederic Clere <jfclere> made a comment on jira MODCLUSTER-335

Hm nothing more than logs in  /target-ER6/jboss-reports/logs-archive/noe.ews.modcluster.tests.ModProxyTest/modProxyOnlyPartOfUrlSpaceTest/HttpdRhel-2/ews-httpd/

Michal Babacek <mbabacek> made a comment on jira MODCLUSTER-335

I have confirmed the regression lays in *mod_proxy_cluster.so*, not in httpd itself or any other module.
With the same *httpd*, I was updating module by module from 1.2.3 to 1.2.4 with this result:

 * (x) mod_proxy_cluster.so  -- broken  
 * (/) mod_advertise.so  
 * (/) mod_manager.so  
 * (/) mod_slotmem.so  

I understand it does not look like so, but I can not find anything but this diff in [mod_proxy_cluster.c|https://gist.github.com/Karm/5530970] that would be even remotely relevant to the issue on hand.

Jean-Frederic Clere <jfclere> made a comment on jira MODCLUSTER-335

That is a regression from the fix of MODCLUSTER-328.

Jean-Frederic Clere <jfclere> made a comment on jira MODCLUSTER-335

Fixed by https://github.com/modcluster/mod_cluster/commit/6610301d2dfbf8d19b2f5edb0328a870eaf78d82

Michal Babacek <mbabacek> updated the status of jira MODCLUSTER-335 to Closed

Michal Babacek <mbabacek> made a comment on jira MODCLUSTER-335

Verified on [modcluster/mod_cluster/tree/1.2.x|https://github.com/modcluster/mod_cluster/tree/1.2.x]. Closing.

not EAP6.1 = not EWS 2.0.1 = known bug.

(In reply to Jean-frederic Clere from comment #9)
> not EAP6.1 = not EWS 2.0.1 = known bug.

This bug _is_ present both in EWS 2.0.1 and EAP6.1, so I don't understand this comment:

> Jean-frederic Clere 2013-05-28 11:57:59 EDT
> not EAP6.1 = not EWS 2.0.1 = known bug.
> Flags: devel_ack-

not fixed EAP6.1 = not fixed EWS 2.0.1 = known bug.

Need further information about this problem and check from SME. Also, if any workaround is available, please let us know.

doc text look ok.

Already fixed by https://github.com/modcluster/mod_cluster/commit/6610301d2dfbf8d19b2f5edb0328a870eaf78d82

Verified in EWS 2.1.0: 

mod_proxy_cluster weinanli$ grep 'uri = luri;' mod_proxy_cluster.c
       uri = luri;

It's cool now. EWS 2.1.0.ER2

Doc type changed to Bug Fix.
Updated Doc text.