Red Hat Bugzilla – Bug 1200142
Content view republish does not make content available to content host immediately
Last modified: 2017-02-23 15:21:14 EST
Description of problem: Content view republish does not make content available to content host immediately Version-Release number of selected component (if applicable): Satellite-6.1.0-RHEL-6-20150303.0 How reproducible: Always Steps to Reproduce: 1. Create a content view cv1, Add repository(ies), publish 2. Add the content view to an activation key ak1 3. Create and register a content host ch1 using this activation key 4. Try to install an installable errata in ch1 - works well. 5. Try to add more content to cv1 and republish 6. Attempt to install an errata (which was made available from the new publish in cv1) Actual results: - Sat6 tasks complete with GREEN - Looking at /var/log/messages in the content host shows package not available: Mar 9 15:07:24 cisco-b200m1-03 goferd: [INFO][worker-0] gofer.agent.rmi:161 - sn=bff0fe9b-5e39-4c4b-8091-23f251c8adae processed in: 27.033 (seconds) Mar 9 15:10:38 cisco-b200m1-03 ntpd[22768]: 0.0.0.0 c615 05 clock_sync Mar 9 15:26:15 cisco-b200m1-03 goferd: [INFO][worker-0] gofer.rmi.dispatcher:610 - call: Content.install() sn=42187c59-88cb-427c-b127-fcdc3a484e71 data={u'task_id': u'872f7300-984c-4f5f-8408-d01fefdce472', u'consumer_id': u'aaa004a6-0cb6-4016-b110-c73ebeff90e9'} Mar 9 15:26:17 cisco-b200m1-03 goferd: [ERROR][worker-0] pulp.agent.lib.dispatcher:80 - handler failed Mar 9 15:26:17 cisco-b200m1-03 goferd: [ERROR][worker-0] pulp.agent.lib.dispatcher:80 - Traceback (most recent call last): Mar 9 15:26:17 cisco-b200m1-03 goferd: [ERROR][worker-0] pulp.agent.lib.dispatcher:80 - File "/usr/lib/python2.6/site-packages/pulp/agent/lib/dispatcher.py", line 76, in install Mar 9 15:26:17 cisco-b200m1-03 goferd: [ERROR][worker-0] pulp.agent.lib.dispatcher:80 - report = handler.install(conduit, units, dict(options)) Mar 9 15:26:17 cisco-b200m1-03 goferd: [ERROR][worker-0] pulp.agent.lib.dispatcher:80 - File "/usr/lib/python2.6/site-packages/pulp_rpm/handlers/rpm.py", line 100, in install Mar 9 15:26:17 cisco-b200m1-03 goferd: [ERROR][worker-0] pulp.agent.lib.dispatcher:80 - details = pkg.install(names) Mar 9 15:26:17 cisco-b200m1-03 goferd: [ERROR][worker-0] pulp.agent.lib.dispatcher:80 - File "/usr/lib/python2.6/site-packages/pulp_rpm/handlers/rpmtools.py", line 136, in install Mar 9 15:26:17 cisco-b200m1-03 goferd: [ERROR][worker-0] pulp.agent.lib.dispatcher:80 - raise caught Mar 9 15:26:17 cisco-b200m1-03 goferd: [ERROR][worker-0] pulp.agent.lib.dispatcher:80 - InstallError: 0:alsa-utils-1.0.22-9.el6_6.x86_64: No package(s) available to install Mar 9 15:26:17 cisco-b200m1-03 goferd: [INFO][worker-0] gofer.agent.rmi:161 - sn=42187c59-88cb-427c-b127-fcdc3a484e71 processed in: 2.720 (seconds) Expected results: Errata should be installed successfully Additional info:
It requires 'yum clean all' in the content host to make the newly published content available.
*** Bug 1194718 has been marked as a duplicate of this bug. ***
This will be resolved by a candlepin change here: https://github.com/candlepin/candlepin/pull/844/files assigning to devan
Verified. Satellite-6.1.0-RHEL-6-20150317.0
(In reply to sthirugn from comment #7) > Verified. > > Satellite-6.1.0-RHEL-6-20150317.0 Correction: This was verified in Satellite-6.1.0-RHEL-7-20150320.1
When I tested last week (see Comment 7 and Comment 8), I think the packages were available instantly because of the content host's previous regitration to library. Now I retested it with clean slate content host - (yum clean all), now it failed. Satellite-6.1.0-RHEL-7-20150324.0
It appears that it it is no longer set for 24 hours, but it just isn't set at all.
My apologies I tested to the point of certs, not on actual clients. Something in subscription-manager is assuming 0 means unset somehow and then just ignoring it. (which is wrong) However we also can't fix that in any timely fashion as those clients are in the wild and must be supported. Instead I have modified to set the expiry to 1 second, which should do the trick. I have confirmed this value does make it from the entitlement cert to the redhat.repo file. Fixed in candlepin.git master: c34f5bfb3a29f7dffe13e581e80fd3b60212d4a1 In candlepin-0.9.48-1
Fixed in candlepin-0.9.48-1.el6
Verified. - I am able to install the applicable errata to a content host using incremental updates. - Also I see that metadata_expire = 1 is set in the content host Version Tested: Satellite-6.1.0-RHEL-6-20150331.1
This bug is slated to be released with Satellite 6.1.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2015:1592