Red Hat Bugzilla – Bug 1134100
When publishing/promoting CV, while capsule exists, content gets synced twice to capsule
Last modified: 2017-02-23 16:01:37 EST
Description of problem: When a user has capsule(s) attached to an instance, and attempts to publish/promote content, said content is synced twice to capsule -- once during the initial publish process and then again afterwards. Version-Release number of selected component (if applicable): Satellite-6.0.4-RHEL-6-20140823.0 How reproducible: Steps to Reproduce: 1. Attach a capsule to your sat 2. Create a content view and attempt to publish/promote it. 3. Note dynflow tasks and/or possibly just talk w/ a knowledgeable developer who understands what the heck is going on (jsherrill!) Actual results: Apparently, this is the situation. * Previously, as part of a publish, new content was immediately pushed out to capsules as part of that publish * There was discussion to break this into two tasks - do initial pub/promote to library and spawn a second task to sync. * I am not sure whether a real decision was made on the behavior in the bullet above, but whatever the case, we're apparently still doing a sync during the initial publish -- followed by a second sync immediately after. Expected results: Apparently there was a lot of discussion between dev on how to handle the sync, whether to keep it as one big monolithic function (promote/publish/capsule sync) or break it out (promote/pub THEN content sync). As an end user I think the second behavior is more palatable because otherwise user is sitting there on a publish that takes ages and user has no idea why. Regardless, we shouldn't be syncing twice, independent of workflow decision. Additional info:
To clarify it was split into two tasks and then accidentally added back as part of the publish/promote process as well. So the node metadata generate and sync are now done twice during a publish or promote. Once as part of publish/promote and a 2nd time as a separate task that gets started at the very end of the publish/promote task.
As an aside, if a user has like 10, 20 capsules, following the monolithic model, the publish could possibly take a Very Long Time as each capsule gets synced.
Created redmine issue http://projects.theforeman.org/issues/7273 from this bug
Moving to POST since upstream bug http://projects.theforeman.org/issues/7273 has been closed ------------- Justin Sherrill Applied in changeset commit:katello|b50dc0e4e85bd0f46d65775c35f0050c4a62c988.
This appears fixed in recent capsule sync efforts. VERIFIED Satellite-6.1.0-RHEL-7-20150320.1
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