Bug 862880
| Summary: | katello-agent throws trace, won't let user install packages on RHEL 6.1 system | ||||||
|---|---|---|---|---|---|---|---|
| Product: | Red Hat Satellite | Reporter: | Corey Welton <cwelton> | ||||
| Component: | katello-agent | Assignee: | Brad Buckingham <bbuckingham> | ||||
| Status: | CLOSED CURRENTRELEASE | QA Contact: | Og Maciel <omaciel> | ||||
| Severity: | high | Docs Contact: | |||||
| Priority: | unspecified | ||||||
| Version: | 6.0.1 | CC: | cpelland, jlaska, omaciel | ||||
| Target Milestone: | Unspecified | Keywords: | Triaged | ||||
| Target Release: | Unused | Flags: | cwelton:
needinfo-
|
||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Whiteboard: | |||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||
| Doc Text: |
Cause: yum and yum-rhn-plugins included in default/base RHEL 6.1 clients are not compatible with CFSE
Consequence: installing/removing packages using the CFSE web ui fail
Fix: update yum and yum-rhn-packages to latest available version
Result: installing/removing packages from the CFSE web ui works after packages are updated.
|
Story Points: | --- | ||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2012-12-12 15:23:56 UTC | Type: | Bug | ||||
| Regression: | --- | Mount Type: | --- | ||||
| Documentation: | --- | CRM: | |||||
| Verified Versions: | Category: | --- | |||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||
| Embargoed: | |||||||
| Attachments: |
|
||||||
it should be noted that after i updated this box to rhel 6.3 and restarted goferd, package(s) installed fine.
2012-10-04 11:57:52,405 [INFO][02013477-4fc4-44d3-8705-f21f6dfed26b] connect() @ broker.py:103 - {se-rhelbox.usersys.redhat.com:5671} connected to AMQP
2012-10-04 11:58:04,349 [INFO][worker-0] dispatch() @ dispatcher.py:530 - dispatching:{
"args": [
[
"atmel-firmware",
"acl"
]
],
"auth": {
"secret": "b3f5f02e05849afb9fa60ee847295bd48bde015dd42eea19039a2ac5025a5276",
"pam": null,
"uuid": "02013477-4fc4-44d3-8705-f21f6dfed26b"
},
"classname": "Packages",
"kws": {},
"cntr": null,
"method": "update"
}
2012-10-04 11:58:08,245 [INFO][worker-0] update() @ katelloplugin.py:198 - Packages updated: {'resolved': [], 'deps': []}
2012-10-04 11:58:08,245 [INFO][worker-0] sendreply() @ rmi.py:158 - 32c1a212-1aac-46bf-bc83-a97239a819d2 processed in: 3.954 (seconds)
I tried to reproduce the error; however, server initiated installs/updates worked successfully.
The following was the basic scenario:
- on my katello server
- create RHEL 6 provider/product/repo
- sync and promote RHEL 6
- client
- create RHEL 6.1 client (used our snap-guest config)
- # cat /etc/redhat-release
Red Hat Enterprise Linux Server release 6.1 (Santiago)
- register and subscribe the client to RHN
- yum repolist
- update subscription manager to (subscription-manager-0.99.19.4-1.el6_3.x86_64)
- install katello-agent (katello-agent-1.1.3-1.el6.noarch)
- unregister the client from RHN
- install bootstrap rpm from my katello server (e.g. candlepin-cert-consumer-myserver.com-1.0-1.noarch.rpm)
- service goferd start
- register the client to my katello server
- subscribe the client to my rhel6 repo
- yum repolist
At this point, I was able to successfully install packages (e.g. xterm) as well as update packages (e.g. abrt) from the server.
Packages installed on server:
-----------------------------
katello-all-1.1.12-14.el6cf.noarch
katello-cli-1.1.8-7.el6cf.noarch
katello-selinux-1.1.1-1.el6cf.noarch
katello-qpid-client-key-pair-1.0-1.noarch
katello-glue-candlepin-1.1.12-14.el6cf.noarch
katello-qpid-broker-key-pair-1.0-1.noarch
katello-certs-tools-1.1.8-1.el6cf.noarch
katello-common-1.1.12-14.el6cf.noarch
katello-1.1.12-14.el6cf.noarch
katello-candlepin-cert-key-pair-1.0-1.noarch
katello-configure-1.1.9-7.el6cf.noarch
katello-cli-common-1.1.8-7.el6cf.noarch
katello-glue-pulp-1.1.12-14.el6cf.noarch
pulp-admin-1.1.13-1.el6cf.noarch
m2crypto-0.21.1.pulp-7.el6.x86_64
python-oauth2-1.5.170-2.pulp.el6.noarch
pulp-1.1.13-1.el6cf.noarch
pulp-client-lib-1.1.13-1.el6cf.noarch
mod_wsgi-3.3-3.pulp.el6.x86_64
pulp-common-1.1.13-1.el6cf.noarch
pulp-selinux-server-1.1.13-1.el6cf.noarch
katello-glue-pulp-1.1.12-14.el6cf.noarch
Based on discussion with Corey, we are going to move this to ON_QA to see if we can get it reproduced. Created attachment 628636 [details]
rhsm.log
This issue still persists on RHEL 6.1 client. Attaching rhsm.log
Sadly this is still affecting RHEL 6.1 clients. Client: * subscription-manager-0.99.19.4-1.el6_3.x86_64 Server: * candlepin-0.7.8.1-1.el6cf.noarch * candlepin-selinux-0.7.8.1-1.el6cf.noarch * candlepin-tomcat6-0.7.8.1-1.el6cf.noarch * katello-1.1.12-16.el6cf.noarch * katello-all-1.1.12-16.el6cf.noarch * katello-candlepin-cert-key-pair-1.0-1.noarch * katello-certs-tools-1.1.8-1.el6cf.noarch * katello-cli-1.1.8-7.el6cf.noarch * katello-cli-common-1.1.8-7.el6cf.noarch * katello-common-1.1.12-16.el6cf.noarch * katello-configure-1.1.9-7.el6cf.noarch * katello-glue-candlepin-1.1.12-16.el6cf.noarch * katello-glue-pulp-1.1.12-16.el6cf.noarch * katello-qpid-broker-key-pair-1.0-1.noarch * katello-qpid-client-key-pair-1.0-1.noarch * katello-selinux-1.1.1-1.el6cf.noarch * pulp-1.1.13-1.el6cf.noarch * pulp-admin-1.1.13-1.el6cf.noarch * pulp-client-lib-1.1.13-1.el6cf.noarch * pulp-common-1.1.13-1.el6cf.noarch * pulp-selinux-server-1.1.13-1.el6cf.noarch I installed a new rhel6.1 client from an internal repo and was still unable to reproduce the error using my CFSE server.
I then went to look at the logs associated with the error and decided to register my rhel6.1 client to the QA CFSE server. Using this client I am still unable to reproduce the error. (i.e. I can install/update packages to my client using the QA CFSE server).
I then looked at the QA rhel6.1 client and in the /var/log/gofer/agent.log, I see that the initial failed attempts had the following traceback:
Traceback (most recent call last):
File "/usr/lib/python2.6/site-packages/gofer/rmi/dispatcher.py", line 485, in __call__
retval = method(*args, **keywords)
File "/usr/lib/gofer/plugins/katelloplugin.py", line 175, in install
installed = p.install(names)
File "/usr/lib64/gofer/plugins/package.py", line 176, in install
yb.install(pattern=info)
File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 3338, in install
mypkgs = self.pkgSack.returnPackages(patterns=pats,
File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 882, in <lambda>
pkgSack = property(fget=lambda self: self._getSacks(),
File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 668, in _getSacks
self.repos.populateSack(which=repos)
File "/usr/lib/python2.6/site-packages/yum/repos.py", line 260, in populateSack
self.doSetup()
File "/usr/lib/python2.6/site-packages/yum/repos.py", line 72, in doSetup
self.ayum.plugins.run('prereposetup')
File "/usr/lib/python2.6/site-packages/yum/plugins.py", line 184, in run
func(conduitcls(self, self.base, conf, **kwargs))
File "/usr/share/yum-plugins/rhnplugin.py", line 197, in prereposetup_hook
for opt, repoexp in opts.repos:
AttributeError: Values instance has no attribute 'repos'
The files at the bottom of the stack are part of the yum and yum-rhn-plugin rpms:
# rpm -qf /usr/lib/python2.6/site-packages/yum/__init__.py /usr/lib/python2.6/site-packages/yum/repos.py /usr/lib/python2.6/site-packages/yum/plugins.py /usr/share/yum-plugins/rhnplugin.py
yum-3.2.29-30.el6.noarch
yum-3.2.29-30.el6.noarch
yum-3.2.29-30.el6.noarch
yum-rhn-plugin-0.9.1-40.el6.noarch
I see on the QA client, that it was updated to rhel6.3 prior to being able to successfully install the vim rpm from the QA CFSE server. From the /var/log/yum.log, it appears that both yum and yum-rhn-plugin were part of this update.
I suspect the issue raised by this bugzilla is actually that the yum and yum-rhn-plugin packages included in the base rhel6.1 client that the test infrastructure is creating is not compatibile with CFSE. On my client, the install included yum-3.2.29 and yum-rhn-plugin-0.9.1-40.
I'd suggest the following:
- create the rhel 6.1 client using the same procedures that demonstrates the issue
- rpm -q yum yum-rhn-plugin
- yum update yum yum-rhn-plugin
And then try to perform the package actions again. If this demonstrates the issue, we will know that they yum pkgs need to be updated. If it does not, then we'll need to look at the other packages that were updated on the client (in /var/log/yum.log) to determine which might be the issue.
# rpm -q yum yum-rhn-plugin subscription-manager yum-3.2.29-17.el6.noarch yum-rhn-plugin-0.9.1-26.el6.noarch subscription-manager-0.95.11-1.el6.x86_64 # yum update yum yum-rhn-plugin subscription-manager # rpm -q yum yum-rhn-plugin subscription-manager yum-3.2.29-30.el6.noarch yum-rhn-plugin-0.9.1-40.el6.noarch subscription-manager-0.99.19.4-1.el6_3.x86_64 # yum -y install http://[CFSE]/pub/candlepin-cert-consumer-[CFSE]-1.0-1.noarch.rpm # subscription-manager --force register --username admin --password admin --org Starbucks --env Demi --autosubscribe Proceeded to install katello-agent and start goferd service. Then, from web ui was able to install packages. Verified that this is caused by older versions of yum and yum-rhn-plugin from RHEL 6.1 base installations. Updating them solves the problem. Filed https://bugzilla.redhat.com/show_bug.cgi?id=868984 to deal with documenting it. clearing needinfo flag. Marking CLOSED CURRENTRELEASE based on comment#10. Customers are advised to update their RHEL-6.1 systems prior to remotely scheduling package operations from System Engine. getting rid of 6.0.0 version since that doesn't exist |
Description of problem: When trying to install packages from katello on a rhel 6.1 system, errors are thrown in the agent log and packages fail to install. katello-agent-1.1.2-1.el6cf.noarch Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. sync and promote RHEL 6Server content 2. register a RHEL 6.1 system to katello. note that you may have to get a more recent subscription-manager package than the one that comes default with RHEL 6.1. Preferably do this with an activation key or just manually subscribe the system to RHEL 3. Systems > $system > Content > Packages 4. Choose a package (such as "abrt") from the content and attempt to update. Actual results: update/install fails: 2012-10-03 15:58:20,464 [ERROR][worker-0] __call__() @ dispatcher.py:488 - Values instance has no attribute 'repos' Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/gofer/rmi/dispatcher.py", line 485, in __call__ retval = method(*args, **keywords) File "/usr/lib/gofer/plugins/katelloplugin.py", line 197, in update updated = p.update(names) File "/usr/lib64/gofer/plugins/package.py", line 165, in update yb.update(pattern=info) File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 3676, in update m = self.pkgSack.returnNewestByNameArch(patterns=pats) File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 882, in <lambda> pkgSack = property(fget=lambda self: self._getSacks(), File "/usr/lib/python2.6/site-packages/yum/__init__.py", line 668, in _getSacks self.repos.populateSack(which=repos) File "/usr/lib/python2.6/site-packages/yum/repos.py", line 265, in populateSack self.doSetup() File "/usr/lib/python2.6/site-packages/yum/repos.py", line 72, in doSetup self.ayum.plugins.run('prereposetup') File "/usr/lib/python2.6/site-packages/yum/plugins.py", line 184, in run func(conduitcls(self, self.base, conf, **kwargs)) File "/usr/share/yum-plugins/rhnplugin.py", line 199, in prereposetup_hook for opt, repoexp in opts.repos: AttributeError: Values instance has no attribute 'repos' 2012-10-03 15:58:20,522 [INFO][worker-0] sendreply() @ rmi.py:158 - 67ba7ebd-37dc-4706-9b7d-9fa1459abc17 processed in: 778 (ms) 2012-10-03 15:58:20,643 [INFO][PathMonitor1] changed() @ katelloplugin.py:135 - changed: /etc/yum.repos.d/redhat.repo 2012-10-03 15:58:20,723 [INFO][PathMonitor1] report_enabled() @ katelloplugin.py:382 - reporting: {'enabled_repos': {'repos': []}} Expected results: should be able to install packages Additional info: Note: this seems specific to 6.1, and/or maybe earlier. I don't know the full range of distros this affects, but it doesn't seem to occur on 6.3, etc. I have tried this on at least two 6.1 boxes -- one old, one a fresh provision -- and get this error both places. I think I've tried it on another 6.1 too, but I can't remember.