From mboxrd@z Thu Jan 1 00:00:00 1970 From: Loic Dachary Subject: Re: Hammer backport and bypassing procedure Date: Sat, 29 Aug 2015 10:03:37 +0200 Message-ID: <55E16759.2020909@dachary.org> References: <55E0B396.2090602@dachary.org> <55E0C803.8020402@redhat.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="uagmuW45TjKKwmq17dqoCsoRj5OrAquBU" Return-path: Received: from mail2.dachary.org ([91.121.57.175]:45749 "EHLO smtp.dmail.dachary.org" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752261AbbH2IDq (ORCPT ); Sat, 29 Aug 2015 04:03:46 -0400 In-Reply-To: Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Abhishek Varshney Cc: Ceph Development This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --uagmuW45TjKKwmq17dqoCsoRj5OrAquBU Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable On 29/08/2015 07:46, Abhishek Varshney wrote:> Hi Loic, >=20 > How about marking the PRs which pass integration/upgrade tests with a p= refix PT (Passed Tests) or something in the title after they are merged i= nto the stable branch. This is probably how we can do it: >=20 > 1. While preparing an integration branch, also get all the PRs which a= re merged but do not have PT as prefix. These are essentially the PRs whi= ch have bypassed tests. Nathan had a similar idea and asked that the "Integration Passed" status = is added to the Backport tracker (see http://tracker.ceph.com/issues/1182= 4). We should start using it. > 2. Perform integration/update tests on all the other open PRs in the i= ntegration branch. > 3. If the integration branch passes all the tests, mark all the PRs as= PT, including the ones which had bypassed tests earlier. Now, we know th= at the PRs which had bypassed tests are bug free. That makes senses. We should amend http://tracker.ceph.com/projects/ceph-= releases/wiki/HOWTO_populate_the_integration_branch to also select merged= commits that have not seen tests. Note that the hammer branch also is te= sted, automatically (the test plan is documented at http://tracker.ceph.c= om/projects/ceph-releases/wiki/HOWTO_monitor_the_automated_tests_AKA_nigh= tlies). So, once a PR is merged, it will eventually be tested in this way= =2E > 4. If the integration branch encounters failures, we know it could be = because of the PRs which had bypassed tests and we know what those PRs ar= e from step 1. This is going in the right direction :-) That's a lot of manual updating = though and it would be great of teuthology could update the issues / PRs = with test results so we don't have to manually maintain that inventory.=20 Cheers >=20 > Thanks for the clarification on this scenario. >=20 > Regards > Abhishek >=20 > On Sat, Aug 29, 2015 at 2:13 AM, Josh Durgin > wrote: >=20 > On 08/28/2015 12:16 PM, Loic Dachary wrote: >=20 > Hi Abhishek, >=20 > We've just had an example of a backport merged into hammer alth= ough it did not follow the procedure : https://github.com/ceph/ceph/pull/= 5691 >=20 > It's a key aspect of backports : we're bound to follow procedur= e, but developers are allowed to bypass it entirely. It may seem like som= ething leading to chaos and frustration but it turns out to be exactly th= e opposite. In a nutshell, it would be constant source of frustration for= developers to learn and obey the rules documented at http://tracker.ceph= =2Ecom/projects/ceph-releases/wiki/HOWTO because it would not benefit the= m significantly. It would also be a problem for us, backporters, because = developers would not be as interested in backporting and our workload wou= ld significantly increase. >=20 > When a developer prepares a backport on his / her own, we updat= e the pull request and the issues to obey the procedure so the (s)he does= not have to. Sure, it's a little tedious but it's a small price to pay f= or the benefit of having a backport being dealt with. That's what I did f= or https://github.com/ceph/ceph/pull/5691 : updaging the corresponding is= sues, adding cross references to the pull request. >=20 > Samuel Just felt confident enough about the backport that it di= d not need a rados run to verify it does the right thing. Since it's ulti= mately Sam's responsibility, that's also ok. The only thing we need to ke= ep in mind when analyzing the next rados run is that this backport did no= t pass yet. We don't have a way to mark commits that bypassed tests just = yet, if you have ideas let us know :-) >=20 >=20 > That was me merging it based on my local testing. I'll keep an eye = out > for any fallout in the hammer runs. >=20 > Thanks for keeping everything updated Loic! > Josh >=20 >=20 >=20 > -----------------------------------------------------------------------= ------------------------------------------------------------------- >=20 > This email and any files transmitted with it are confidential and inten= ded solely for the use of the individual or entity to whom they are addre= ssed. If you have received this email in error please notify the system m= anager. This message contains confidential information and is intended on= ly for the individual named. If you are not the named addressee you shoul= d not disseminate, distribute or copy this e-mail. Please notify the send= er immediately by e-mail if you have received this e-mail by mistake and = delete this e-mail from your system. If you are not the intended recipien= t you are notified that disclosing, copying, distributing or taking any a= ction in reliance on the contents of this information is strictly prohibi= ted. Although Flipkart has taken reasonable precautions to ensure no viru= ses are present in this email, the company cannot accept responsibility f= or any loss or damage arising from the use of this email or attachments >=20 --=20 Lo=C3=AFc Dachary, Artisan Logiciel Libre --uagmuW45TjKKwmq17dqoCsoRj5OrAquBU Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iEYEARECAAYFAlXhZ1kACgkQ8dLMyEl6F23mlgCePJS0yM+17F+xabIvO3HukOwD /MUAnAj7xsw0JxaPqTm4l+Gzn/hBkt0F =SiGX -----END PGP SIGNATURE----- --uagmuW45TjKKwmq17dqoCsoRj5OrAquBU--