From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nathan Cutler Subject: Untested PRs merged to jewel before 10.2.6 release? Date: Tue, 7 Mar 2017 22:38:52 +0100 Message-ID: <713de223-60a8-7def-5e59-c6c09b52dc06@suse.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mx2.suse.de ([195.135.220.15]:58747 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933439AbdCHBOw (ORCPT ); Tue, 7 Mar 2017 20:14:52 -0500 Received: from relay2.suse.de (charybdis-ext.suse.de [195.135.220.254]) by mx2.suse.de (Postfix) with ESMTP id DA3B9ABCD for ; Tue, 7 Mar 2017 21:38:53 +0000 (UTC) Sender: ceph-devel-owner@vger.kernel.org List-ID: To: "ceph-devel@vger.kernel.org" I noticed that the v10.2.6 tag was added today and, as far as I can tell, it was added to the tip of the jewel branch. That would be fine, except that the tip of the jewel branch at the time the tag was added (420a9a0796e327676cb704bc18b2020f9941d7b6) is several PRs ahead of the jewel SHA1 that underwent integration testing (d9eaab456ff45ae88e83bd633f0c4efb5902bf07). $ git log --oneline --no-merges d9eaab456ff45ae88e83bd633f0c4efb5902bf07..420a9a0796e327676cb704bc18b2020f9941d7b6 88f4895 qa/suites/upgrade/hammer-x: Add some volumes 0c242d1 qa/suites/ceph-deploy: Drop OpenStack volume count ccd0265 qa: replace centos 7.2 with centos 7.3 2cbec5b Removed dumplin test 13234.yaml as not needed anymore 771e1d9 qa/suites/rest: Openstack volumes a18640f qa/suites/ceph-ansible: Openstack volumes 841688b qa/suites/fs: Add openstack volume configuration 9778743 qa/suites/samba: Openstack volume configuration cd1e8ef qa/suites/hadoop: Openstack volume configuration ac7add1 qa/suites/knfs: Add openstack volume configuration ba35859 qa/suites/kcephfs: Openstack volume configuration aced718 qa/suites/krbd: Add openstack volume configuration 94d5888 qa/suites/rgw: Add openstack volume configuration Am I seeing that right? If so, that appears to potentially defeat the purpose of the integration testing we do. Maybe I'm overstating the danger here -- the commits look innocuous enough -- but it seems to set a rather dangerous precedent. Nathan