Red Hat Bugzilla – Bug 1418455
unable to unregister host if candlepin record is missing
Last modified: 2017-06-27 08:40:17 EDT
Description of problem: ****** downstream-only bug! do not clone. This is already fixed upstream via a larger refactor. Its very likely not worth cherrypicking the entire change for this. ****** In certain cases, users may have hosts that are registered in Katello but not Candlepin. This typically occurs due to bugs related to registration. When this happens, the host cannot be removed from Katello. There is some code that checks for the host's entitlements, but if it gets a 404 from Candlepin, it will error with the following: RestClient::ResourceNotFound: Katello::Resources::Candlepin::CandlepinResource: 404 Resource Not Found {"displayMessage":"Runtime Error RESTEASY001185: Could not find resource for relative : /consumers//entitlements of full path: https://localhost:8443/candlepin/consumers//entitlements at org.jboss.resteasy.core.registry.PathParamSegment.matchPattern:209","requestUuid":"e7511d96-c594-48b2-b572-f5cb1eddd579"} (GET /candlepin/consumers//entitlements) Version-Release number of selected component (if applicable): 6.2.7 How reproducible: every time Steps to Reproduce: 1. load database with affected records 2. attempt to delete host via foreman-rake console Actual results: error Expected results: successful unregister Additional info:
Please add verifications steps for this bug to help QE verify
repro steps: NOTE: the repro steps involve causing some data corruption to occur! * create fresh satellite, register one host * log into postgres db (sudo su - postgres; psql candlepin) * run this sql: delete from cp_consumer_facts; delete from cp_installed_products ; delete from cp_consumer; * now, attempt to delete host via web UI without the patch, this will fail and the host will stick around. With the patch, you'll still get an error raised in the web UI, but the delete will succeed.
Verified in Satellite 6.2.9 Snap 2. Followed the steps outlined in #5, but was more selective with the cp_consumer delete. The host deletion task still produced an error, as expected (see attached), however the host was deleted. To get past the paused deletion task, you can go into the dynflow console, skip the "Actions::Candlepin::Consumer::Destroy" step, then navigate back to the task and click resume.
Created attachment 1270547 [details] verification screenshot
Hello Team, Is there any workaround for this?
Hi Satish, Please refer the below KCS: ==> https://access.redhat.com/solutions/2991221 Regards, Nithin
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-2017:1191