Bug 732350
| Summary: | Tomcat failed to start properly or the installer ran out of tries | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | [Community] Spacewalk | Reporter: | Phil <phil.ingram> | ||||||||
| Component: | Installation | Assignee: | Jan Pazdziora (Red Hat) <jpazdziora> | ||||||||
| Status: | CLOSED CURRENTRELEASE | QA Contact: | Red Hat Satellite QA List <satqe-list> | ||||||||
| Severity: | high | Docs Contact: | |||||||||
| Priority: | unspecified | ||||||||||
| Version: | 1.5 | CC: | hughpearse, jpazdziora, tlestach | ||||||||
| Target Milestone: | --- | ||||||||||
| Target Release: | --- | ||||||||||
| Hardware: | x86_64 | ||||||||||
| OS: | Linux | ||||||||||
| Whiteboard: | |||||||||||
| Fixed In Version: | spacewalk-java-1.6.39-1 | Doc Type: | Bug Fix | ||||||||
| Doc Text: | Story Points: | --- | |||||||||
| Clone Of: | Environment: | ||||||||||
| Last Closed: | 2011-12-22 16:49:41 UTC | Type: | --- | ||||||||
| Regression: | --- | Mount Type: | --- | ||||||||
| Documentation: | --- | CRM: | |||||||||
| Verified Versions: | Category: | --- | |||||||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||||||
| Embargoed: | |||||||||||
| Bug Depends On: | 735460 | ||||||||||
| Bug Blocks: | 723481 | ||||||||||
| Attachments: |
|
||||||||||
|
Description
Phil
2011-08-22 06:02:12 UTC
When navigating to my server I get redirected to /rhn/Login.do but am greeted with a 404 saying: HTTP Status 404 - type Status report message description The requested resource () is not available. Apache Tomcat/6.0.32 We were able to reproduce the problem internally with Spacewalk nightly. Tomáš, could you investigate more? Yes, the trouble is c3p0 that is available now in F15 repo in a higher version that the package we'd like to use from jpackage repo. The quickest workaround is to step back f.e. # yum downgrade c3p0 -y to use c3p0-0.9.1.2-2.jpp5 from jpackage repo. It seem that the fedora package cannot find appropriate logger ... java.lang.ClassNotFoundException: com.mchange.v2.log.MLog I created: Bug 735460 - java.lang.ClassNotFoundException: com.mchange.v2.log.MLog for c3p0 fedora 15 component. (In reply to comment #3) > Yes, the trouble is c3p0 that is available now in F15 repo in a higher version > that the package we'd like to use from jpackage repo. > > The quickest workaround is to step back f.e. > # yum downgrade c3p0 -y > to use c3p0-0.9.1.2-2.jpp5 from jpackage repo. Can we just require (in rpm dependencies) the exact c3p0 version on Fedora 15 for now? (In reply to comment #6) > (In reply to comment #3) > > Yes, the trouble is c3p0 that is available now in F15 repo in a higher version > > that the package we'd like to use from jpackage repo. > > > > The quickest workaround is to step back f.e. > > # yum downgrade c3p0 -y > > to use c3p0-0.9.1.2-2.jpp5 from jpackage repo. > > Can we just require (in rpm dependencies) the exact c3p0 version on Fedora 15 > for now? Actually, mchange-commons is already installed on my Fedora 15, so the trick is to do # ln -s /usr/share/java/mchange-commons.jar /var/lib/tomcat6/webapps/rhn/WEB-INF/lib (In reply to comment #7) > > Actually, mchange-commons is already installed on my Fedora 15, so the trick is > to do > > # ln -s /usr/share/java/mchange-commons.jar > /var/lib/tomcat6/webapps/rhn/WEB-INF/lib And this is what I did for Fedora 15 in Spacewalk master, 5024fea0db061d82a0ba61d6b80c7843c34678b4. Tagged as spacewalk-java-1.6.39-1. nice work people :). To test do I just have to enable the nightly builds as per * https://fedorahosted.org/spacewalk/wiki/HowToInstall (In reply to comment #9) > nice work people :). To test do I just have to enable the nightly builds as > per > > * https://fedorahosted.org/spacewalk/wiki/HowToInstall Well, if you don't want to pollute your released installation with nightly bits, just do # ln -s /usr/share/java/mchange-commons.jar /var/lib/tomcat6/webapps/rhn/WEB-INF/lib/mchange-commons.jar # service tomcat6 restart -- that should do the trick. excellent. well, i can confirm that after doing the symlink everything now works on a fresh install; including a dependency issue I used to have with tanukiwrapper which is no doubt unrelated here.. is this where I change the status of this bug to CLOSED?? (In reply to comment #11) > excellent. well, i can confirm that after doing the symlink everything now > works on a fresh install; including a dependency issue I used to have with > tanukiwrapper which is no doubt unrelated here.. > > is this where I change the status of this bug to CLOSED?? We will close this bugzilla once the fix is in released Spacewalk -- so far we only have it in nightly builds (and we have this manual workaround). Please leave it open, we will take care of it later. Created attachment 521801 [details]
rhn_taskomatic_daemon.log
hmmm, well even though the installation seemed to work and I have been able to create a new channel with 3 repositories bound to it there is an issue related to the scheduler where according to spacewalk "The scheduling service appears down. Please contact your Satellite administrator.".
[root@spacewalk ~]# service taskomatic status
RHN Taskomatic is running (6632).
[root@spacewalk ~]# netstat -pant|grep task
tcp 0 0 127.0.0.1:32001 127.0.0.1:31003 ESTABLISHED 6632/taskomaticd
From looking at taskomatic's log output I noticed that it referenced "mchange", being the library I manually linked before.
For this new installation I did a dvd installation, updated, installed spacewalk as per the fedorahosted guide on to postgres and then went through the creation of channels and repos.
should I file a separate bug and if so what extra information would be required?
thanks again for your assistance
Yes, we need to add the /usr/share/java/mchange-commons.jar to taskomatic wrapper.java.classpath as well (in /etc/rhn/default/rhn_taskomatic_daemon.conf). (I'll fix and test the change as soon as I get my F15 machine running.) Created attachment 523466 [details]
patch for rhn_taskomatic_daemon.conf
I can confirm that including the mchange jar fixes taskomatic. All i did was modify the conf file and restarted taskomatic. within seconds it started working through its schedule and now all is good.
Thanks again everyone for your assistance :)
(In reply to comment #15) > Created attachment 523466 [details] > patch for rhn_taskomatic_daemon.conf > > I can confirm that including the mchange jar fixes taskomatic. All i did was > modify the conf file and restarted taskomatic. within seconds it started > working through its schedule and now all is good. > > Thanks again everyone for your assistance :) Tomáš, is it OKay / harmless to have this patch applied even on OSes where the thing does not exist as a separate .jar? Yes, it's safe. And also much better than to link different jars for every OS. spacewalk.git: 2f93f6226b9634c1b1b34c86bef6ece21e342e8e *** Bug 742576 has been marked as a duplicate of this bug. *** Spacewalk 1.6 has been released. |