From mboxrd@z Thu Jan 1 00:00:00 1970 From: Willem Jan Withagen Subject: Re: Bumping my jenkins builds to the wall in radable.sh Date: Thu, 17 Mar 2016 10:29:40 +0100 Message-ID: <56EA7904.4000807@digiware.nl> References: <56E9A585.9000309@digiware.nl> <56E9BBD2.5050700@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from smtp.digiware.nl ([31.223.170.169]:59448 "EHLO smtp.digiware.nl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932329AbcCQJ3t (ORCPT ); Thu, 17 Mar 2016 05:29:49 -0400 In-Reply-To: Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Gregory Farnum , Josh Durgin Cc: Ceph Development , Yehuda Sadeh-Weinraub On 16-3-2016 21:06, Gregory Farnum wrote: > On Wed, Mar 16, 2016 at 1:02 PM, Josh Durgin wrote: >> On 03/16/2016 12:34 PM, Gregory Farnum wrote: >>> >>> On Wed, Mar 16, 2016 at 11:27 AM, Willem Jan Withagen >>> wrote: >>>> >>>> >>>> ../ceph-object-corpus/archive/0.61.4-60-g24c59be/objects/RGWBucketEnt >>>> >>>> ../ceph-object-corpus/archive/0.61.4-60-g24c59be/objects/RGWBucketInfo >>>> error: buffer::end_of_buffer >>>> **** failed to decode >>>> >>>> ../ceph-object-corpus/archive/0.61.4-60-g24c59be/objects/RGWBucketInfo/25f5ff183bba6da708243a0bbbe40a2e >>>> **** >>>> error: buffer::end_of_buffer >>>> **** failed to decode >>>> >>>> ../ceph-object-corpus/archive/0.61.4-60-g24c59be/objects/RGWBucketInfo/54f53e3e301071f48247d59272b27b53 >>>> **** >>>> >>>> >>>> ../ceph-object-corpus/archive/0.61.4-60-g24c59be/objects/rgw_bucket_pending_info >>>> >>>> "Just" a matter of rebase? >>>> Or does it need more tinkering. >>> >>> >>> Well, ceph-object-corpus is a bunch of old encoded versions of our >>> objects, and it's validating that current code is still capable of >>> decoding them. 0.61.4 is pretty old but I *think* should still work >>> (maybe check that your submodules are all up to date; maybe some of >>> them got dropped?). Assuming that still works on Linux, it's not a >>> rebase problem; I would guess maybe these types just never worked >>> because it wasn't BSD cross-compatible until more recently. You might >>> want to explore how they're failing exactly; obviously it's running >>> out of data to decode but the *why* is interesting. >> >> >> It's broken on linux on master: >> >> http://gitbuilder.sepia.ceph.com/gitbuilder-ceph-tarball-trusty-amd64-basic/log.cgi?log=188644cb753fe304d941051b75b9eac49ede1086 >> https://jenkins.ceph.com/job/ceph-pull-requests/3220/console >> >> Probably snuck in while 'make check' was broken from the crush update. > > Oh, and the very next commit after that is the fix, so yes, a rebase > should fix it. (Thanks Kefu+Casey!) > Rebased, and pushed.... Lets see what that brings.