We should do that for RHN Satellite as well. +++ This bug was initially created as a clone of Bug #795680 +++ Description of problem: While regisered to spacewalk proxy, osad is not picking events. Version-Release number of selected component (if applicable): Proxy 1.6.5-1.el6 # rpm -qa | grep proxy spacewalk-proxy-package-manager-1.6.5-1.el6.noarch spacewalk-proxy-installer-1.6.7-1.el6.noarch spacewalk-proxy-common-1.6.5-1.el6.noarch spacewalk-proxy-docs-1.1.1-1.el6.noarch spacewalk-proxy-broker-1.6.5-1.el6.noarch spacewalk-proxy-redirect-1.6.5-1.el6.noarch spacewalk-proxy-management-1.6.5-1.el6.noarch spacewalk-proxy-monitoring-1.6.1-1.el6.noarch spacewalk-proxy-html-1.1.1-1.el6.noarch spacewalk-proxy-selinux-1.1.1-1.el6.noarch (against Spacewalk 1.6, partly 1.7 ...) # rpm -qa | grep spacewalk spacewalk-selinux-1.6.2-1.el5 spacewalk-backend-config-files-common-1.7.19-1.el5 spacewalk-backend-package-push-server-1.7.19-1.el5 spacewalk-java-config-1.6.102-1.el5 spacewalk-setup-jabberd-1.6.5-1.el5 spacewalk-monitoring-selinux-1.6.2-1.el5 spacewalk-admin-1.6.3-1.el5 spacewalk-search-1.6.8-1.el5 spacewalk-common-1.5.1-1.el5 spacewalk-backend-sql-1.7.19-1.el5 spacewalk-backend-config-files-tool-1.7.19-1.el5 spacewalk-branding-1.6.4-1.el5 spacewalk-config-1.6.4-1.el5 spacewalk-slf4j-1.6.1-1.el5 spacewalk-pxt-1.6.38-1.el5 spacewalk-backend-libs-1.7.19-1.el5 spacewalk-grail-1.6.38-1.el5 spacewalk-java-oracle-1.6.102-1.el5 spacewalk-taskomatic-1.6.102-1.el5 spacewalk-backend-server-1.7.19-1.el5 spacewalk-backend-xmlrpc-1.7.19-1.el5 spacewalk-backend-iss-1.7.19-1.el5 spacewalk-backend-xp-1.7.19-1.el5 spacewalk-repo-1.7-4.el5 spacewalk-client-repo-1.7-4.el5 spacewalk-java-lib-1.6.102-1.el5 spacewalk-base-1.6.38-1.el5 spacewalk-java-1.6.102-1.el5 rh-tests-Spacewalk-Installer-Sanity-Yum-install-spacewalk-oracle-1.0-5 spacewalk-backend-1.7.19-1.el5 spacewalk-backend-xml-export-libs-1.7.19-1.el5 spacewalk-backend-applet-1.7.19-1.el5 spacewalk-backend-iss-export-1.7.19-1.el5 spacewalk-certs-tools-1.6.7-1.el5 spacewalk-base-minimal-1.6.38-1.el5 spacewalk-doc-indexes-1.1.1-1.el5 spacewalk-monitoring-1.4.1-1.el5 spacewalk-schema-1.6.42-1.el5 spacewalk-backend-sql-oracle-1.7.19-1.el5 spacewalk-backend-config-files-1.7.19-1.el5 spacewalk-html-1.6.38-1.el5 spacewalk-setup-1.6.5-1.el5 spacewalk-sniglets-1.6.38-1.el5 spacewalk-oracle-1.5.1-1.el5 spacewalk-backend-app-1.7.19-1.el5 spacewalk-backend-tools-1.7.19-1.el5 (on client we had osad-5.10.34-1.el5) How reproducible: Always, deterministic Steps to Reproduce: 1. Register client to proxy. 1.5. (Make sure osad is running on client, you may want to restart osad if it was already running.) 2. Schedule event (e.g. enable monitoring and then schedule remote command). Actual results: Nothing happens (I have been observing only for something more that half an hour, but I think that it does not work). Expected results: Scheduled event is picked up. Additional info: I have checked, that it works directly against spacewalk. When client is registered to proxy, OSA status is shown unknown. After (re)starting osad status changes to offline. Pinging system (via WebUI) adds info about last pinged, restarting osad typically removes this info. (So it seems that at least some information is propagated to spacewalk.) Looking at connection reveals, that osad process has open (TCP) connection to proxy, process c2s in particular. There does not seems to be any (persistent) connection directly to server. (But that may be ok(?)) This was all run over IPv6. --- Additional comment from mkollar on 2012-02-21 04:42:30 EST --- Created attachment 564620 [details] Output of "osad -N -vvvv" while registered to proxy. Attaching output of "osad -N -vvvv" while registered to proxy. --- Additional comment from mkollar on 2012-02-21 04:43:52 EST --- Created attachment 564621 [details] Output of "osad -N -vvvv" while registered to directly to spacewalk. Attaching output of "osad -N -vvvv" while registered to directly to spacewalk. --- Additional comment from msuchy on 2012-03-01 10:17:39 EST --- This is caused by s2s not communicating between two instances of jabberd server. I find this relevant report in upstream: https://bugs.launchpad.net/jabberd2/+bug/374687 Flipping to jabberd. Can you please communicate with upstream to do some release and rebase to it?
Where do you need to fix? rawhide only?
This bug is for RHN Satellite on RHEL. You meant fix for Bug #795680? Then yes, rawhide would be good start.
Ah, okay. I way confused why there were two open bugs.
spacewalk.git master: 6c97648eb2900209a85ca997f87c35c0c8920d4b
To make the thing work, we don't need the patch mentioned in https://bugs.launchpad.net/jabberd2/+bug/374687 at all. All we need is to enable /s2s/lookup/resolve-ipv6 in jabberd's setup.
The following network setup: Satellite 5.5 -> | IPv6 connection | -> RHN Proxy 5.5 -> -> | IPv4 connection | -> Client works just fine for me. I'm guessing problem from comment #12 is a network configuration problem. I'm returning this back to ON_QA.