If a repo sync task fails in succession more times than the configured failure_threshold, then there is no way to restart it using rhui-manager. Because it's number of successive failures has exceeded the threshold, it is not scheduled again. However, when you go into rhui-manager and try to sync an individual repo, it does not show up in the list of available repos to sync. I'm guessing this is because it's in the "Unknown" state, this is what you see for it's state if you get the repo's status via either the dr or vr command from the sync screen. The only way to kick it back off is to use pulp-admin. I don't think we want to be telling customers to do this.
This issue has been mitigated by the pulp rhui branch commit e9b58eccb42a5d71b8d30fb9964b865f8ccffffa, which disables the failure_threshold setting in rhui. To verify, a repo sync can fail greater than 5 times, and it will keep sync'ing on its schedule.
set tracker bug. 746803
To verify this defect , I stopped the httpd service more than 5 times. And Once I start it, sync starts perfectly.. Once I start the httpd service, grinder starts updating logs and fetching metadata: 2011-10-21 13:35:55,207 21859:140663093253888: grinder.RepoFetch:INFO: RepoFetch:455 fetchYumRepo() basepath = /var/lib/pulp//repos/content/dist/rhel/rhui/server/6/6.1/i386/optional/os 2011-10-21 13:35:55,268 21859:140663093253888: grinder.RepoFetch:INFO: RepoFetch:468 Fetching repo metadata... 2011-10-21 13:36:44,496 21859:140663093253888: grinder.RepoFetch:INFO: RepoFetch:474 Determining downloadable Content bits... 2011-10-21 13:36:47,136 21859:140663093253888: grinder.RepoFetch:INFO: RepoFetch:319 2656 packages have been marked to be fetched 2011-10-21 13:36:47,136 21859:140663093253888: grinder.RepoFetch:INFO: RepoFetch:341 Preparing to fetch any available trees.. 2011-10-21 13:36:47,341 21859:140663093253888: grinder.BaseFetch:INFO: BaseFetch:243 Fetching Unknown bytes: .treeinfo from https://cdn.redhat.com/content/dist/rhel/rhui/server/6/6.1/i386/optional/os/.treeinfo when httpd stop: ===================== 2011-10-21 13:39:17,395 21928:140010967791360: grinder.BaseFetch:INFO: activeobject:169 Fetching 2654332 bytes: inkscape-docs-0.47-6.el6.i686.rpm from https://cdn.redhat.com/content/dist/rhel/rhui/server/6/6.1/i386/optional/os/Packages/inkscape-docs-0.47-6.el6.i686.rpm 2011-10-21 13:39:17,400 21928:140011049559808: grinder.BaseFetch:INFO: activeobject:169 Fetching 73164 bytes: ilmbase-devel-1.0.1-6.1.el6.i686.rpm from https://cdn.redhat.com/content/dist/rhel/rhui/server/6/6.1/i386/optional/os/Packages/ilmbase-devel-1.0.1-6.1.el6.i686.rpm Again start the httpd: ======================== 2011-10-21 13:39:34,977 21996:140372684568320: grinder.RepoFetch:INFO: RepoFetch:455 fetchYumRepo() basepath = /var/lib/pulp//repos/content/dist/rhel/rhui/server/6/6.1/i386/optional/os 2011-10-21 13:39:35,045 21996:140372684568320: grinder.RepoFetch:INFO: RepoFetch:468 Fetching repo metadata... And here what I observed is, when I stop the httpd service to stop the running sync, rhui-manager fails with unexpected error. however after restarting the httpd service, when I check the sync status the status remains in same "in-progress" state and grinder starts downloading packages.
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/RHBA-2017:0367