February 2016 - devel - lists.centos.org

Request to Join Altarch SIG - AArch64
by Steve Capper 18 Feb '16

18 Feb '16

Hello, I would like to join the AltArch Special Interest Group for AArch64 please. I am a software engineer at ARM working mainly on the kernel and virtualisation, but would be able to relay queries about other areas to my colleagues and potentially involve them too in future. In the office I have an AMD Seattle system for my development work. Thanks, -- Steve Capper

2 1

CentOS Atomic SIG Meeting Minutes 2015-12-03
by Jason Brooks 17 Feb '16

17 Feb '16

Last week the meetbot died during our CentOS Atomic SIG meeting. This is the log from the meeting, and the one action item we set. See you at the meeting tomorrow at 16:00 UTC in #centos-devel. Jason Actions: jbrooks to poke around about upstream cockpit-docker exclusion 2015-12-03T16:02:30 <jbrooks> #startmeeting CentOS Atomic SIG 2015-12-03T16:02:30 <centbot> Meeting started Thu Dec 3 16:02:30 2015 UTC. The chair is jbrooks. Information about MeetBot at http://wiki.debian.org/MeetBot. 2015-12-03T16:02:30 <centbot> Useful Commands: #action #agreed #help #info #idea #link #topic. 2015-12-03T16:02:31 <dustymabe> for thanksgiving 2015-12-03T16:02:58 <dustymabe> lalatend1M: meeting time 2015-12-03T16:03:01 *** dneary <dneary!~dneary@Maemo/community/docmaster/dneary> has quit IRC (Ping timeout: 250 seconds) 2015-12-03T16:03:07 <jbrooks> #chair dustymabe maxamillion kbsingh lalatend1M jzb 2015-12-03T16:03:07 <centbot> Current chairs: dustymabe jbrooks jzb kbsingh lalatend1M maxamillion 2015-12-03T16:03:32 <jbrooks> who else? 2015-12-03T16:03:40 <tmrts> o/ 2015-12-03T16:03:53 <jbrooks> #chair tmrts 2015-12-03T16:03:53 <centbot> Current chairs: dustymabe jbrooks jzb kbsingh lalatend1M maxamillion tmrts 2015-12-03T16:04:27 <jbrooks> #topic CentOS Atomic Host Rebuild status 2015-12-03T16:04:49 * lalatend1M i shere 2015-12-03T16:05:01 <jbrooks> We had a rebuild release on Nov 24 2015-12-03T16:05:11 <jbrooks> That was the last of the "7.1" rebuilds 2015-12-03T16:05:33 <jbrooks> Currently, I'm iterating through getting the rebuild based on the new centos pkgs built 2015-12-03T16:05:38 *** lalatend1M is now known as lalatenduM 2015-12-03T16:05:47 *** jfilak <jfilak!jfilak@nat/redhat/x-tnaxbhzqcclqnkyz> has quit IRC (Remote host closed the connection) 2015-12-03T16:06:09 <jbrooks> I'm using the CR repo and sort of a extras-next temp repo set up by Johnny 2015-12-03T16:06:37 <dustymabe> nice 2015-12-03T16:06:50 <jbrooks> So that's moving along decently 2015-12-03T16:07:07 *** kbreeman <kbreeman!~rooot(a)64.238.140.42> has joined #centos-devel 2015-12-03T16:07:22 *** jchaloup <jchaloup!Jan@nat/redhat/x-irqgrnsnztmbmexf> has quit IRC (Ping timeout: 246 seconds) 2015-12-03T16:07:34 <hhorak> RemiFedora: kbsingh: guys here are working on it, but they need to move a lot of data and re-set up new server. 2015-12-03T16:07:54 <jbrooks> That brings up sort of a subtopic -- there's been some interest expressed in adding the cockpit-docker and cockpit-kubernetes pkgs into our rebuild 2015-12-03T16:08:12 <kbsingh> hhorak: time to introduce some automation 2015-12-03T16:08:21 <hhorak> RemiFedora: I'd like to make the official announcements together for el6 and el7, and el6 is still not ready. 2015-12-03T16:08:22 <kbsingh> hhorak: if we can do something to help, do tell 2015-12-03T16:08:40 <jbrooks> RHEL Atomic doesn't include these, so that'd be a divergence, but the pkgs themselves are in the standard centos repos 2015-12-03T16:08:43 <dustymabe> jbrooks: and those packages aren't in RHELAH 2015-12-03T16:08:45 <dustymabe> ? 2015-12-03T16:09:06 <kbsingh> are they shipped as containers ? 2015-12-03T16:09:09 *** lars_kurth <lars_kurth!chrome@nat/citrix/x-sexywyygzwzdqpcu> has quit IRC (Ping timeout: 260 seconds) 2015-12-03T16:09:14 <jbrooks> dustymabe, Right, not in the host, but are available, as in, not from the CBS or something 2015-12-03T16:09:15 <kbsingh> i.e why are they released if they are not included in the AH 2015-12-03T16:09:15 *** fzdarsky_ is now known as fzdarsky 2015-12-03T16:09:32 <jbrooks> They, like cockpit-bridge and cockpit-shell, control host things 2015-12-03T16:09:46 <hhorak> kbsingh: well, the sclo6 packages are still not signed/released, right? if they were, we can point people to use those, as RemiFedora proposed :) 2015-12-03T16:09:51 <jbrooks> Really, only cockpit-ws, the web app, is containerizable, as I understand it 2015-12-03T16:09:55 <dustymabe> jbrooks: was a decision made to exclude them from RHEL-AH? 2015-12-03T16:10:10 <dustymabe> or would they consider including them as well? 2015-12-03T16:10:22 <jbrooks> dustymabe, I don't know, I assume so, since they're not there. I don't know why 2015-12-03T16:10:42 <kbsingh> hhorak: they were added to the configs last night, but the job didnt run through cleanly. 2015-12-03T16:10:47 <jbrooks> Maybe not fitting the SKU vision or something? 2015-12-03T16:10:49 <kbsingh> hhorak: i will get them out tomorrow 2015-12-03T16:10:55 <dustymabe> might be worth asking - if there is a good enough reason then that might be a good reason we shouldn't do it 2015-12-03T16:11:11 <jbrooks> OK 2015-12-03T16:11:26 <jbrooks> #action jbrooks to poke around about upstream cockpit-docker exclusion 2015-12-03T16:12:13 <jbrooks> Anything else on this topic? 2015-12-03T16:12:36 <hhorak> kbsingh: ok, thx.. I'll at least point people to el7 builds at least :) 2015-12-03T16:12:49 <dustymabe> jbrooks: nope 2015-12-03T16:13:54 <jbrooks> #topic Four Week Build 2015-12-03T16:14:03 <jbrooks> So, I haven't done anything new on this front 2015-12-03T16:14:23 <kbsingh> me neither 2015-12-03T16:14:24 <jbrooks> I do see that the fedora two week has been firming up, and that's important for the four week 2015-12-03T16:14:55 <jbrooks> The trick now is figuring out a semi-automated flow of the fedora two week "atomic" pkgs into the CBS 2015-12-03T16:14:56 <dustymabe> yeah. honestly I wish I had more time to devote to this, but i've been stretched thin 2015-12-03T16:15:33 <jbrooks> OK, I figured I'd bump it up in ppl's minds, at least :) 2015-12-03T16:15:55 <jbrooks> Other topics before open floor? lalatenduM, do you have adb-related issues? 2015-12-03T16:15:57 *** deryni-work <deryni-work!~ereisner(a)63-148-88-20.dia.static.qwest.net> has joined #centos-devel 2015-12-03T16:16:12 <lalatenduM> jbrooks: yeah 2015-12-03T16:16:21 <jbrooks> #topic ADB 2015-12-03T16:16:36 <lalatenduM> is centos 7.2 out? 2015-12-03T16:16:43 <kbsingh> we have an arm build getting ready - and quite a few people have expressed interest in getting atomic host there as well - is that something we might want to take up in the scope of this group ? 2015-12-03T16:16:43 <jbrooks> Not yet 2015-12-03T16:16:48 *** Humble <Humble!hchiramm@nat/redhat/x-sqybcsspxrljnigq> has quit IRC (Ping timeout: 245 seconds) 2015-12-03T16:17:03 <dustymabe> kbsingh: I have thought about that before.. I like the idea 2015-12-03T16:17:10 <kbsingh> not yet, but eta is a week or so from now 2015-12-03T16:17:10 <lalatenduM> kbsingh: yup I think so 2015-12-03T16:17:14 *** oere <oere!~oere(a)botanix.informatik.tu-chemnitz.de> has quit IRC (Quit: Leaving.) 2015-12-03T16:17:14 <jbrooks> kbsingh, sure 2015-12-03T16:17:30 <kbsingh> the ostree patches needed to support uboot already exist... 2015-12-03T16:17:30 <lalatenduM> I am going to do a ADB build once 7.2 is out 2015-12-03T16:17:46 <jbrooks> lalatenduM, cool 2015-12-03T16:17:50 <lalatenduM> but looks like we have to manage with scratch build 2015-12-03T16:17:59 <kbsingh> lalatenduM: at the moment, the CR repos have all the content already seeded up, so if you can use taht - it might give you a base to test / trial waiting for main line release 2015-12-03T16:18:03 <lalatenduM> as there s a SCm error with teh non scratch builds 2015-12-03T16:18:25 <lalatenduM> is a scm error* 2015-12-03T16:18:28 <lalatenduM> kbsingh: yeah 2015-12-03T16:18:42 *** vpavlin <vpavlin!vpavlin@nat/redhat/x-mikwxkdugihlswrr> has joined #centos-devel 2015-12-03T16:18:48 <kbsingh> i think we will get the scm side of things figured out in the coming days as well, once the git <-> Accounts.centos.org stuff is in place, the git stuff becomes easy ( and the plan is to get that git stuff integrated in the next 10 days or so) 2015-12-03T16:18:54 <jbrooks> lalatenduM, Is there a bug for that error that we can link here 2015-12-03T16:19:01 <lalatenduM> kbsingh: cool 2015-12-03T16:19:11 <lalatenduM> jbrooks: good idea I will file one. 2015-12-03T16:19:23 <kbsingh> jbrooks: tl'dr, cbs needs scm == git.centos.org; but we have no clear path to hosting non rpm content for sig's yet 2015-12-03T16:19:24 <jbrooks> Anything more on that? 2015-12-03T16:19:25 <lalatenduM> I had sent a mail to centos-devel about it 2015-12-03T16:19:37 <jbrooks> Ah, ok 2015-12-03T16:19:41 <kbsingh> ie. we need to get the auth layer in place, then see how that maps to branch names, sig space etc. 2015-12-03T16:20:16 <lalatenduM> Also planning to build teh openshift origin docker image to include in the adb box, but thats not a blocker 2015-12-03T16:20:46 <jbrooks> https://lists.centos.org/pipermail/centos-devel/2015-November/014034.html 2015-12-03T16:20:47 <lalatenduM> till we have a better caching solution for docker image 2015-12-03T16:20:57 <jbrooks> ^ that's the centos-devel msg, I believe 2015-12-03T16:21:03 <lalatenduM> jbrooks: yup 2015-12-03T16:21:20 <jbrooks> OK, any more on that? 2015-12-03T16:21:28 <lalatenduM> nope I am done 2015-12-03T16:21:38 <jbrooks> #topic Open Floor 2015-12-03T16:21:42 <tmrts> I'd like to talk about replacing cloud-init 2015-12-03T16:21:54 <jbrooks> tmrts, Cool, go ahead 2015-12-03T16:22:05 <tmrts> As you guys know I developed https://github.com/tmrts/flamingo to be a replacement for cloud-init 2015-12-03T16:22:32 <tmrts> We were in the process of integrating it to Atomic Host but I've had some health issues 2015-12-03T16:23:09 *** vpavlin <vpavlin!vpavlin@nat/redhat/x-mikwxkdugihlswrr> has quit IRC (Ping timeout: 260 seconds) 2015-12-03T16:23:11 <tmrts> I was away from the SIG for a while and I'd like to complete the process if the intention is still the same 2015-12-03T16:23:50 <jbrooks> tmrts, Have you made any test images w/ flamingo built-in? 2015-12-03T16:24:24 *** hhorak <hhorak!hhorak@nat/redhat/x-vkqtjbqanjtplabh> has quit IRC (Quit: Leaving.) 2015-12-03T16:24:25 <tmrts> I tried to rebuild the image with flamingo rpm (both on copr and centos build system) but I've had libvirtd issues Is there any developments on the build system for the images or is it still done through scripts? 2015-12-03T16:24:48 <jbrooks> Still scripts 2015-12-03T16:25:05 <jbrooks> Point me at your copr or cbs location and I can try to do a build 2015-12-03T16:25:19 <tmrts> jbrooks, I performed integration tests with the custom built image but It needs to be done again as I understand there was updates 2015-12-03T16:25:57 <dustymabe> did any of us ever evaluate the coreos rewrite of cloud-init 2015-12-03T16:26:06 <tmrts> in cbs you can find it with atomic-testing label and I'll link the copr in a sec 2015-12-03T16:26:16 <dustymabe> if we could combine efforts then I think we could gain momentum 2015-12-03T16:26:36 <tmrts> I've had discussions with the core-os team 2015-12-03T16:26:47 <kbsingh> the intel guys are also redoing things, again in go. for their clear containers effort 2015-12-03T16:26:52 *** juergh <juergh!~juergh(a)adsl-84-227-41-33.adslplus.ch> has quit IRC (Quit: Leaving.) 2015-12-03T16:27:04 <jbrooks> dustymabe, Not me -- it'd be good to hook up w/ the fedora cloud folks and do some sort of info/testing roundup of these options 2015-12-03T16:27:18 <jbrooks> ID the issues w/ cloud-init, the options, etc. 2015-12-03T16:27:24 <tmrts> They want to merge efforts as well. But, for us it would mean jumping on coreos-cloudinit and contributing to it so that we can use it 2015-12-03T16:27:56 <jbrooks> tmrts, does it fit w/ what you've been doing? 2015-12-03T16:28:01 <tmrts> #info copr repository https://copr.fedoraproject.org/coprs/tamertas/flamingo/ 2015-12-03T16:28:07 <jbrooks> thx 2015-12-03T16:28:36 <tmrts> jbrooks, do you mean getting in touch with fedora folks? 2015-12-03T16:28:39 <dustymabe> tmrts: nice so you have reached out to them? 2015-12-03T16:28:45 <dustymabe> them == coreos 2015-12-03T16:28:49 <jbrooks> tmrts, the coreos effort 2015-12-03T16:29:01 <jbrooks> Like, does merging efforts make sense to you? 2015-12-03T16:29:12 <tmrts> dustymabe, they've reached out to me 2015-12-03T16:29:19 <dustymabe> oh, even better 2015-12-03T16:29:32 <lalatenduM> cool 2015-12-03T16:30:11 <tmrts> I evaluated coreos-cloudinit and basically they hardwired it to coreos (some parts). ultimately a joint effort would be better 2015-12-03T16:30:42 <tmrts> But, we'd need to contribute to coreos-cloudinit and make it work for us 2015-12-03T16:31:10 *** sshnaidm <sshnaidm!sshnaidm@nat/redhat/x-yzjqlttdrtvsnbca> has quit IRC (Ping timeout: 246 seconds) 2015-12-03T16:31:13 <dustymabe> tmrts: right. so what I would say is that it's going to be hard to convince them to abandon what they have. what I would propose is that we actually evaluate what they have, remove the hardwired parts, but still allow them to achieve their goals with those hardwirings 2015-12-03T16:31:18 *** jstribny <jstribny!jstribny@nat/redhat/x-cmyzhkpgavztclxi> has quit IRC (Quit: Leaving) 2015-12-03T16:31:32 <dustymabe> and then we can together launch a new project together with them 2015-12-03T16:31:39 *** rssfed23 <rssfed23!~rssfed23(a)91.217.86.20> has quit IRC (Quit: My Mac has gone to sleep. ZZZzzz…) 2015-12-03T16:31:43 <dustymabe> under some different branding 2015-12-03T16:31:44 <tmrts> dustymabe, absolutely it makes no sense for them to abandon it 2015-12-03T16:32:21 <dustymabe> honestly if we get centos and coreos on-board many others will follow suit 2015-12-03T16:32:24 <dustymabe> IMHO 2015-12-03T16:32:40 <dustymabe> because many people (myself included) view cloud-init from ubuntu as dead 2015-12-03T16:33:17 <dustymabe> thoughts from anyone else on this? 2015-12-03T16:33:24 <jbrooks> Agreed, though we really need to get fedora into the loop -- that's where centos ultimately comes from. For the rebuild, certainly, and we've been envisioning the four-week in that way, as well 2015-12-03T16:34:10 <jbrooks> Do we have a next action on this? 2015-12-03T16:34:12 <dustymabe> jbrooks: agree 2015-12-03T16:34:23 <dustymabe> probably would be better to do it in fedora first 2015-12-03T16:34:48 <dustymabe> tmrts: are you interested in combining efforts with coreos or are you more interested in flamingo as it exists today? 2015-12-03T16:35:09 <dustymabe> you never know. they might even be willing to keep the flamingo name for the launch of the combined effort 2015-12-03T16:35:20 <dustymabe> but most likely would want to have the new project be based largely on their code 2015-12-03T16:35:28 <dustymabe> makes sense since it is already being used and testeed 2015-12-03T16:36:00 <tmrts> dustymabe, Here's my viewpoint a collaboration would be very helpful, but as it stands it would mean splitting flamingo into parts and contribute it to coreos-cloudinit 2015-12-03T16:36:31 *** foxhunt72 is now known as foxhunt_away 2015-12-03T16:36:39 <dustymabe> right 2015-12-03T16:37:02 <dustymabe> which I think probably makes the most sense, because it is already in production and being used 2015-12-03T16:37:25 <kbsingh> i need to rebase away to another thing, thanks fuys 2015-12-03T16:37:26 <kbsingh> guys even 2015-12-03T16:37:28 <dustymabe> if it was the other way around where flamingo was out there and had users and coreos didn't then probably would go the other way 2015-12-03T16:37:47 <dustymabe> tmrts: any chance you will be at fosdem? 2015-12-03T16:38:19 <tmrts> dustymabe, yeah I agree with you it is battle-tested 2015-12-03T16:38:34 <tmrts> dustymabe, unfortunately I won't be able to attend 2015-12-03T16:38:53 <dustymabe> maybe the architecture needs rework (to match your design better), but most of the code will probably come from coreos-cloudinit 2015-12-03T16:38:58 <dustymabe> which is actually probably a good thing for us 2015-12-03T16:39:16 <dustymabe> since they probably support way more functionality at this point (not sure on this, just guessing) 2015-12-03T16:39:55 <dustymabe> tmrts: so I guess what I am asking is "is the proposal I made acceptable to you?" 2015-12-03T16:40:00 <dustymabe> since you wrote flamingo 2015-12-03T16:41:04 <tmrts> dustymabe, the most acceptable choice would be to create a new organization(or effort) and merge what they have with flamingo along with them 2015-12-03T16:41:27 <dustymabe> right. and what I am proposing is that we do that 2015-12-03T16:41:47 <tmrts> dustymabe, yes I agree with you 2015-12-03T16:41:52 <dustymabe> where the new org has probably most of the new code coming from coreos-cloud-init and some of it coming from flamingo 2015-12-03T16:42:24 <dustymabe> either way we want to merge our efforts 2015-12-03T16:42:33 <dustymabe> some of the coreos devs live in NYC 2015-12-03T16:42:39 <dustymabe> and I see them often at meetups 2015-12-03T16:44:07 *** centbot <centbot!~centbot@centos/bot/centbot> has joined #centos-devel 2015-12-03T16:44:08 <dustymabe> and how long ago was that? 2015-12-03T16:44:41 *** Bahhumbug <Bahhumbug!jrd@serentos/admin/jrd> has joined #centos-devel 2015-12-03T16:44:44 <tmrts> I replied that we'd like to merge efforts as well and what do you think the next action should be 2015-12-03T16:45:06 <Bahhumbug> My apologies for dropping the bot in the middle of your meeting. I don't believe meetbot will be able to recover gracefully from this action. 2015-12-03T16:45:15 <tmrts> this was a week ago and he said that we should start by contributing to coreos-ci 2015-12-03T16:45:34 <dustymabe> tmrts: ok cool 2015-12-03T16:46:05 <dustymabe> basically what we need to do is float the idea of the combined organization with them 2015-12-03T16:46:34 <dustymabe> and as long as they understand that is our goal and they agree it is their goal too then I don't see any problem with contributing code to their GH for now 2015-12-03T16:46:40 <Bahhumbug> If anyone needs raw bot logs from the meeting (sans the ~2 minutes it was down, obviously) please ping me and I will be happy to provide them. 2015-12-03T16:46:46 <dustymabe> to ease the migration to the new org 2015-12-03T16:47:06 <tmrts> I'll mention that we'd be more comfortable with a combined organization and help with the migration 2015-12-03T16:47:16 <dustymabe> Bahhumbug: jbrooks will probably need them to post to the list 2015-12-03T16:47:19 *** paragan <paragan!~paragan@unaffiliated/paragan> has quit IRC (Remote host closed the connection) 2015-12-03T16:47:27 <dustymabe> tmrts: feel free to add me to the emails 2015-12-03T16:47:41 <tmrts> yeah I'll cc you 2015-12-03T16:47:56 <dustymabe> thanks 2015-12-03T16:47:56 <Bahhumbug> Ping me at the end of your meeting and I'll make them available. I no longer camp this channel. 2015-12-03T16:47:56 <jbrooks> Bahhumbug, yes please, I'm at jbrooks(a)redhat.com 2015-12-03T16:47:56 <dustymabe> I like this strategy 2015-12-03T16:47:56 <Bahhumbug> Again, my apologies. 2015-12-03T16:47:59 *** Bahhumbug <Bahhumbug!jrd@serentos/admin/jrd> has left #centos-devel 2015-12-03T16:48:12 *** lars_kurth <lars_kurth!chrome@nat/citrix/x-lkwhcqeojptdzhen> has joined #centos-devel 2015-12-03T16:48:24 <dustymabe> tmrts++ 2015-12-03T16:48:47 <dustymabe> tmrts: how much time do you have to devote to this effort? 2015-12-03T16:49:02 <dustymabe> is it a primary thing or is it something that is a side project for you? 2015-12-03T16:49:50 *** jchaloup <jchaloup!~Jan(a)147.229.63.11> has joined #centos-devel 2015-12-03T16:49:51 <dustymabe> once we get a clear path to success what we can do is promote what we are doing and get more people interested in the project 2015-12-03T16:50:32 <tmrts> dustymabe, I was worked on it during the summer as part of GSoC. At the moment it is a side-project. That is one of my concerns as well I might not be able to devote enough time 2015-12-03T16:51:00 *** outi <outi!~outi(a)HSI-KBW-217-008-049-015.hsi.kabelbw.de> has quit IRC (Quit: leaving) 2015-12-03T16:51:15 <dustymabe> tmrts: got ya 2015-12-03T16:51:20 <dustymabe> ok I'll keep that in mind 2015-12-03T16:52:02 <jbrooks> OK, are we all set for now? 2015-12-03T16:52:12 <dustymabe> I think so 2015-12-03T16:52:17 <kbsingh> can Arrfab / Evolution, not push the logs directly to www.centos.org ? 2015-12-03T16:52:18 <tmrts> yes I think so as well 2015-12-03T16:52:29 <jbrooks> I guess there isn't a meeting to end? 2015-12-03T16:52:37 <jbrooks> #endmeeting

1 1

Thanks for CentOS 7 ppc64!
by Bryson Lee 17 Feb '16

17 Feb '16

Just a quick note to thank all involved in the ppc64 AltArch effort! Was able to get one of our IBM 812L’s up and running very quickly after importing the Everything ISO into our Cobbler environment. Next I’ll try rebuilding some of the packages that we need for 32-bit ppc… Thanks again for all the great work! Regards, -Bryson ________________________________ This message (including any attachments) may contain confidential information intended for a specific individual and purpose. If you are not the intended recipient, you should delete this message and any attachments.

3 8

ppc64le. Team looking for opal-prd package
by Bill Buros 17 Feb '16

17 Feb '16

From peers working with CentOS on Power8, /"I am trying to install OPAL-PRD on cent-os. Do you know where to get opal-prd package ?"/ Presently opal-prd is not available in the CentOS repos or in EPEL, but is expected in a future release of RHEL. In the meantime, it is possible to build the rpm from source and install on CentOS. Latest source rpm : https://hegdevasant.fedorapeople.org/opal-prd/v5/opal-prd-5.1.13-1.fc22.src… Thanks to Vasant Hegde for pointer to files.

4 5

seven.centos.org
by Karanbir Singh 16 Feb '16

16 Feb '16

hi We have seven.centos.org which runs more or less as an announcement and community blog - but the word 'seven' in there highlights its focus around CentOS Linux 7. As we promote that, this mindset goes with it. Am wondering if we need a better worded name for the blog, or to move it under www.centos.org/blog/ ( yet retain the lose'ish perms for contributors.ie. SIGs are welcome too ) ? Regards -- Karanbir Singh +44-207-0999389 | http://www.karan.org/ | twitter.com/kbsingh GnuPG Key : http://www.karan.org/publickey.asc

2 1

RPM repo targets for Atomic SIG
by Karanbir Singh 15 Feb '16

15 Feb '16

hi, We are working to get the ADB content released and have settled on mirror.centos.org/centos/7/atomic/x86_64/adb/ as the destination ( you can see the content set at buildlogs.centos.org/centos/7/atomic/x86_64/adb/ now ) This would then land at the same level as the /repo directory for atomic host ( which contains the ostree repo ). For the rpms that are used to build the ostree repo, we were using buildlogs.centos.org/centos/7/atomic/x86_64/Packages/ directory - I want to propose : 1) move the signed content to mirror.centos.org 2) put it under /centos/7/atomic/x86_64/{name}/ where name is either rpm-ostree or atomic-host or something suiteable. Would that work for everyone ? Note that the ostree repo ( therefore the url in the deployed atomic host instances ) does not change. Regards -- Karanbir Singh +44-207-0999389 | http://www.karan.org/ | twitter.com/kbsingh GnuPG Key : http://www.karan.org/publickey.asc

1 0

ABRT reporting workflow
by Jakub Filak 15 Feb '16

15 Feb '16

Hello, Let me briefly re-introduce my self. I work on ABRT and I was involved in the integration of ABRT with CentOS Bug Tracker. I want to let you know that we are re-working ABRT reporting workflow in order to reduce the number of opened bug reports. Our current mission is to encourage users to open "full bug report" (i.e. issue in CentOS Bug Tracker) only if they want to participate in bug resolving process and submit anonymous crash statics [1][2] otherwise. We are going to let users choose from reporting options similar the following ones: * [Submit anonymous crash details] - [I do not want to participate in bug resolution process] * [Enter new bug report] - [I want to help to fix this bug and I need to have it fixed as soon as possible] These changes should land in CentOS 7 too. Ideas and suggestions are more than welcome. Best regards, Jakub 1: https://github.com/abrt/faf/wiki/uReport 2: https://retrace.fedoraproject.org/faf/summary/?opsysreleases=71

2 3

[StorageSIG] state of Ceph
by François Cami 14 Feb '16

14 Feb '16

Hi, This is the state of the Ceph effort so far: Ceph-specific dist-gits are hosted at https://github.com/CentOS-Storage-SIG Ceph Hammer (0.94.5) packages have been built and tagged storage7-ceph-hammer-testing: http://cbs.centos.org/koji/taginfo?tagID=646 These work well enough for test and development use cases. A centos-release-ceph-hammer rpm has been built but is not yet properly tagged to go into extras: http://cbs.centos.org/koji/packageinfo?packageID=2497 A Ceph quickstart howto (WIP) was added to the wiki: https://wiki.centos.org/SpecialInterestGroup/Storage/ceph-Quickstart Next steps: * properly tag the centos-release-ceph-hammer rpm so that it lands into extras ( https://bugs.centos.org/view.php?id=10285 ), * Ceph upstream is preparing a Hammer dot release ( http://tracker.ceph.com/versions/505 ) which will be rebuilt and tested as soon as it is available, * decide whether to build infernalis or skip it and rebuild Jewel instead, * complete and proofread the howto, * add CI jobs to test releases. Thanks to everyone involved. François

1 0

Not doing GSoC this year
by Karsten Wade 12 Feb '16

12 Feb '16

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Last year we had mixed success with Google Summer of Code. On one side, we had successful projects and some happy students and satisfied mentors. However, on the back end (often not visible to others here) we were plagued with inefficiencies and mistakes that made the experience a bit of a PITA all around. The good thing is we now have a good idea of what it would take to run a successful program: * Keep it as a year long process, so the post-review leads right in to planning for the next year. * Have Admins and Mentors who are strongly engaged from early on. * Improve the process to bring in students doing work. Bottom line is that it's only worth doing if we do it right. That is the consensus from many discussions I've been involved in over the last six months or so. So the deadline for proposing being a mentoring org is approaching next week, and there is no way we can sanely put together a bid. Instead, I'd like us to ponder if there is value for us, what that value is, and get ourselves prepared for a run in the future ... if that is what we decide makes sense. Regards, - - Karsten - -- Karsten 'quaid' Wade .^\ CentOS Doer of Stuff http://TheOpenSourceWay.org \ http://community.redhat.com @quaid (identi.ca/twitter/IRC) \v' gpg: AD0E0C41 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iEYEARECAAYFAla+JhIACgkQ2ZIOBq0ODEEttACgo1P8GKi8AE5dAfePUvlC7nhd y5EAn3L/HMzc2y1+Oy/BhSsjrTB9roT5 =CDJG -----END PGP SIGNATURE-----

1 0

Atomic SIG Chair Change
by Joe Brockmeier 12 Feb '16

12 Feb '16

Hey all, Discussed at a meeting a few weeks ago, but I failed to send email on this. (And hence another good reason for the change...) Jason Brooks has been heavily active in keeping the rebuilds going and generally moving things forward. I proposed that he take over as chair, and he didn't run screaming - nor did anyone else object. So, I'd like this to be officially official, officially. Lazy consensus rules, so unless there's a -1 from one of the members of the SIG, this should be effective in 72 hours. Apologies again for tardiness in sending this email. Best, jzb -- Joe Brockmeier | Community Team, OSAS jzb(a)redhat.com | http://community.redhat.com/ Twitter: @jzb | http://dissociatedpress.net/

2 1

official centos-7 docker image are wrong
by Farkas Levente 12 Feb '16

12 Feb '16

hi, according to this bug https://bugzilla.redhat.com/show_bug.cgi?id=1142311 rh silently replace iputils with the same version with a fixed capabilities. unfortunately official base centos-7 (ie. latest 7.2) docker images are build with the wrong version, so these images should have to be rebuild with the proper version. currently in the official centos7 docker images: ------------------------------------- # rpm -qi iputils Name : iputils Version : 20121221 Release : 7.el7 Architecture: x86_64 Install Date: Wed 23 Dec 2015 06:09:24 PM UTC Group : System Environment/Daemons Size : 368577 License : BSD and GPLv2+ Signature : RSA/SHA256, Wed 25 Nov 2015 02:43:41 PM UTC, Key ID 24c6a8a7f4a80eb5 Source RPM : iputils-20121221-7.el7.src.rpm Build Date : Fri 20 Nov 2015 07:12:19 PM UTC Build Host : worker1.bsys.centos.org Relocations : (not relocatable) Packager : CentOS BuildSystem <http://bugs.centos.org> Vendor : CentOS URL : http://www.skbuff.net/iputils Summary : Network monitoring tools including ping Description : The iputils package contains basic utilities for monitoring a network, including ping. The ping command sends a series of ICMP protocol ECHO_REQUEST packets to a specified network host to discover whether the target machine is alive and receiving network traffic. ------------------------------------- while in the centos repo: ------------------------------------- # rpm -qi iputils Name : iputils Version : 20121221 Release : 7.el7 Architecture: x86_64 Install Date: Tue 15 Dec 2015 03:15:22 PM CET Group : System Environment/Daemons Size : 368577 License : BSD and GPLv2+ Signature : RSA/SHA256, Wed 25 Nov 2015 03:43:41 PM CET, Key ID 24c6a8a7f4a80eb5 Source RPM : iputils-20121221-7.el7.src.rpm Build Date : Fri 20 Nov 2015 08:12:19 PM CET Build Host : worker1.bsys.centos.org Relocations : (not relocatable) Packager : CentOS BuildSystem <http://bugs.centos.org> Vendor : CentOS URL : http://www.skbuff.net/iputils Summary : Network monitoring tools including ping Description : The iputils package contains basic utilities for monitoring a network, including ping. The ping command sends a series of ICMP protocol ECHO_REQUEST packets to a specified network host to discover whether the target machine is alive and receiving network traffic. ------------------------------------- imho it's a bug! without it ping not working from any docker container based on centos7. -- Levente "Si vis pacem para bellum!"

3 8

New CentOS Atomic Host Available
by Jason Brooks 12 Feb '16

12 Feb '16

An updated version of CentOS Atomic Host (version 7.20160203) is now available for download. CentOS Atomic Host is a lean operating system designed to run Docker containers, built from standard CentOS 7 RPMs, and tracking the component versions included in Red Hat Enterprise Linux Atomic Host 7.2. CentOS Atomic Host is available as a VirtualBox or libvirt-formatted Vagrant box, or as an installable ISO, qcow2 or Amazon Machine image. These images are available for download at cloud.centos.org(http://cloud.centos.org/centos/7/atomic/images/). The backing ostree repo is published to mirror.centos.org(http://mirror.centos.org/centos/7/atomic/x86_64/repo). CentOS Atomic Host includes these core component versions: * kernel-3.10.0-327.4.5.el7.x86_64 * cloud-init-0.7.5-10.el7.centos.1.x86_64 * atomic-1.6-6.gitca1e384.el7.x86_64 * kubernetes-1.0.3-0.2.gitb9a88a7.el7.x86_64 * etcd-2.1.1-2.el7.x86_64 * ostree-2015.9-2.atomic.el7.x86_64 * docker-1.8.2-10.el7.centos.x86_64 * flannel-0.5.3-8.el7.x86_64 ## Upgrading If you're running a previous version of CentOS Atomic Host, you can upgrade to the current image by running the following command: $ sudo atomic host upgrade ## Images ### Vagrant CentOS-Atomic-Host-7-Vagrant-Libvirt.box (http://cloud.centos.org/centos/7/atomic/images/CentOS-Atomic-Host-7-Vagrant -Libvirt.box) (416 MB) and CentOS-Atomic-Host-7-Vagrant-Virtualbox.box (http://cloud.centos.org/centos/7/atomic/images/CentOS-Atomic-Host-7-Vagrant -Virtualbox.box) (428 MB) are Vagrant boxes for Libvirt and Virtualbox providers. The easiest way to consume these images is via the Atlas / Vagrant Cloud setup (see https://atlas.hashicorp.com/centos/boxes/atomic-host) For example, getting the VirtualBox instance up would involve running the following two commands on a machine with vagrant installed: $ vagrant init centos/atomic-host && vagrant up --provider virtualbox ### ISO The installer ISO (http://cloud.centos.org/centos/7/atomic/images/CentOS-Atomic-Host-7-Install er.iso) (737 MB) can be used via regular install methods (PXE, CD, USB image, etc.) and uses the Anaconda installer to deliver the CentOS Atomic Host. This allows flexibility to control the install using kickstarts and define custom storage, networking and user accounts. This is the recommended process for getting CentOS Atomic Host onto bare metal machines, or to generate your own image sets for custom environments. ### QCOW2 The CentOS-Atomic-Host-7-GenericCloud.qcow2 (http://cloud.centos.org/centos/7/atomic/images/CentOS-Atomic-Host-7-Generic Cloud.qcow2) (1 GB) is suitable for use in on-premise and local virtualized environments. We test this on OpenStack, AWS and local Libvirt installs. If your virtualization platform does not provide its own cloud-init metadata source, you can create your own(http://www.projectatomic.io/blog/2014/10/getting-started-with-cloud-init /) NoCloud iso image. ### Amazon Machine Images Region Image ID ------ -------- us-east-1 ami-896653e3 us-west-2 ami-1f94747f us-west-1 ami-dae791ba eu-west-1 ami-430aba30 eu-central-1 ami-40d6cd2c ap-southeast-1 ami-4319d720 ap-northeast-1 ami-b54d4adb ap-southeast-2 ami-27123544 ap-northeast-2 ami-961fd1f8 sa-east-1 ami-238d0e4f ### SHA Sums 4062ef213eed698ac8ec03b32a55dd6903721a44dc8d54a18513644f160ca7d4 CentOS-Atomic-Host-7.20160130-GenericCloud.qcow2 a7dd91736f45101e95e7d9a80c2eede9164eb0392c8c4748b08c98a42d3eda39 CentOS-Atomic-Host-7.20160130-GenericCloud.qcow2.gz 9eca81d3638e4e00fc734d7233b47a3af803237cc82e5a66b3a587552232dcdc CentOS-Atomic-Host-7.20160130-GenericCloud.qcow2.xz be3c1a3326c04026f37bd6b6c2fccca3a285ea40ac663230624854abeaaee135 CentOS-Atomic-Host-7.20160130-Installer.iso 90942c3599e15ae21cdc0b1682b8e0d3fa88f8db2f6fdca0ece28c2bffdbb34f CentOS-Atomic-Host-7.20160130-Vagrant-Libvirt.box ce674573f6d7020b3d04c51f070d7172e71b6a4316c1495c238a7eac0260cb5a CentOS-Atomic-Host-7.20160130-Vagrant-Virtualbox.box ## Release Cycle The CentOS Atomic Host image follows the upstream Red Hat Enterprise Linux Atomic Host cadence. After sources are released, they're rebuilt and included in new images. After the images are tested by the SIG and deemed ready, we announce them. ## Getting Involved CentOS Atomic Host is produced by the CentOS Atomic SIG (http://wiki.centos.org/SpecialInterestGroup/Atomic) based on upstream work from Project Atomic (http://www.projectatomic.io/) If you'd like to work on testing images, help with packaging, documentation -- join us! The SIG meets weekly on Thursdays at 16:00 UTC in the #centos-devel channel, and you'll often find us in #atomic and/or #centos-devel if you have questions. You can also join the atomic-devel (https://lists.projectatomic.io/mailman/listinfo/atomic-devel) mailing list if you'd like to discuss the direction of Project Atomic, its components, or have other questions. ## Getting Help If you run into any problems with the images or components, feel free to ask on the centos-devel (http://lists.centos.org/mailman/listinfo/centos-devel) mailing list. Have questions about using Atomic? See the atomic (https://lists.projectatomic.io/mailman/listinfo/atomic) mailing list or find us in the #atomic channel on Freenode. --- Jason Brooks Red Hat Open Source and Standards @jasonbrooks | @redhatopen http://community.redhat.com

2 1

Registries
by Karanbir Singh 11 Feb '16

11 Feb '16

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Guys, Zeeshan has offered to look at the wider state of registries and do a short briefing on them ( oci/docker specific as a place to start ) at the next centos buildsystems irc meeting. Brian: what would be the best way to host this ? regards - -- Karanbir Singh, Project Lead, The CentOS Project +44-207-0999389 | http://www.centos.org/ | twitter.com/CentOS GnuPG Key : http://www.karan.org/publickey.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAEBAgAGBQJWu0VjAAoJEI3Oi2Mx7xbtzq4H/36aBI4WM1Nmu73xcEzhiC7M t3tMsPK7/N9AqckjFQFSe00mJmIVE9ibzNIxmlTzzreHzlxjaKdDaMDES26+Op3e hOPWqsMoQ44wEJ6b7T6t7oP7pnOHo/dIdBwU+KirRfTHDKS/AcMH50GjrleqBIAm Ff7WluJQFwIO1CXH2jLYqbYCNCDdMxG932VnOgCZU0mOSPHS7cZ4VkzvDXnPdey8 VIwZKZwXYl+MBQDqohYJcSmUobiWYeEXlBOKNu6Q9J1+lYrfnrQq6/cpW6ETtpjB NdZfDf2bh7E1h88TpB0N93zPyel30Qkp9Pbe/uKF+fzP0jw6QdFQJLkiTSDUaW0= =Y6hq -----END PGP SIGNATURE-----

3 2

Re: [CentOS-devel] [atomic] official centos-7 docker image are wrong
by Jason Brooks 11 Feb '16

11 Feb '16

----- Original Message ----- > From: "Eric Paris" <eparis(a)redhat.com> > To: "Farkas Levente" <lfarkas(a)lfarkas.org>, "CentOS devel" <centos-devel(a)centos.org>, atomic(a)projectatomic.io > Sent: Wednesday, February 10, 2016 5:00:31 PM > Subject: Re: [atomic] official centos-7 docker image are wrong > > I'm willing to bet one week's pay that RH didn't silently release the > same version twice with different code... > > Wasn't there a problem that one of the base image build systems somehow > didn't support file caps and so the caps were being silently lost? > Maybe something about squashfs not understanding filecaps and rpm not > considering that a permanent failure... Very very hazy... There was this: http://www.projectatomic.io/blog/2015/04/problems-with-ping-in-containers-o… > > I know I remember this bug in the back of my head... > > In any case if you look in the centos base image (after installing > attr) > > # getfattr -n security.capability /usr/bin/ping > /usr/bin/ping: security.capability: No such attribute > > # rpm -V iputils > ........P /usr/bin/ping > ........P /usr/bin/ping6 > ........P /usr/sbin/arping > ........P /usr/sbin/clockdiff > > So while the right version/rpm was installed the actual bits in the > base image aren't what the rpm says they should be! The bug is in how > the base image is created... I don't know where, but just building it > again will likely just keep doing the wrong thing... > > I know this will trigger someone's mind who looking into this months > and months ago... > > -Eric > > On Wed, 2016-02-10 at 11:15 +0100, Farkas Levente wrote: > > hi, > > according to this bug > > https://bugzilla.redhat.com/show_bug.cgi?id=1142311 > > rh silently replace iputils with the same version with a fixed > > capabilities. > > unfortunately official base centos-7 (ie. latest 7.2) docker images > > are > > build with the wrong version, so these images should have to be > > rebuild > > with the proper version. > > currently in the official centos7 docker images: > > ------------------------------------- > > # rpm -qi iputils > > Name : iputils > > Version : 20121221 > > Release : 7.el7 > > Architecture: x86_64 > > Install Date: Wed 23 Dec 2015 06:09:24 PM UTC > > Group : System Environment/Daemons > > Size : 368577 > > License : BSD and GPLv2+ > > Signature : RSA/SHA256, Wed 25 Nov 2015 02:43:41 PM UTC, Key ID > > 24c6a8a7f4a80eb5 > > Source RPM : iputils-20121221-7.el7.src.rpm > > Build Date : Fri 20 Nov 2015 07:12:19 PM UTC > > Build Host : worker1.bsys.centos.org > > Relocations : (not relocatable) > > Packager : CentOS BuildSystem <http://bugs.centos.org> > > Vendor : CentOS > > URL : http://www.skbuff.net/iputils > > Summary : Network monitoring tools including ping > > Description : > > The iputils package contains basic utilities for monitoring a > > network, > > including ping. The ping command sends a series of ICMP protocol > > ECHO_REQUEST packets to a specified network host to discover whether > > the target machine is alive and receiving network traffic. > > ------------------------------------- > > > > while in the centos repo: > > ------------------------------------- > > # rpm -qi iputils > > Name : iputils > > Version : 20121221 > > Release : 7.el7 > > Architecture: x86_64 > > Install Date: Tue 15 Dec 2015 03:15:22 PM CET > > Group : System Environment/Daemons > > Size : 368577 > > License : BSD and GPLv2+ > > Signature : RSA/SHA256, Wed 25 Nov 2015 03:43:41 PM CET, Key ID > > 24c6a8a7f4a80eb5 > > Source RPM : iputils-20121221-7.el7.src.rpm > > Build Date : Fri 20 Nov 2015 08:12:19 PM CET > > Build Host : worker1.bsys.centos.org > > Relocations : (not relocatable) > > Packager : CentOS BuildSystem <http://bugs.centos.org> > > Vendor : CentOS > > URL : http://www.skbuff.net/iputils > > Summary : Network monitoring tools including ping > > Description : > > The iputils package contains basic utilities for monitoring a > > network, > > including ping. The ping command sends a series of ICMP protocol > > ECHO_REQUEST packets to a specified network host to discover whether > > the target machine is alive and receiving network traffic. > > ------------------------------------- > > imho it's a bug! without it ping not working from any docker > > container > > based on centos7. > > > >

1 0

SIGs, versions, and yum update
by George Dunlap 10 Feb '16

10 Feb '16

A couple of weeks ago I was just about to push an update of Xen 4.4 to Xen 4.6 to the Virt SIG repositories, and I suddenly got some push-back from the users, because Xen 4.6 has some fairly significant changes -- particularly the removal of the old toolstack daemon called xend. (xend had been disabled by default and had lots of messages warning about the upcoming deprecation, even in 4.4.) It seems some users have the expectation that they should be able to run "yum update -y" on their servers on a regular basis, without having to worry about accidentally updating to a new major version that break things. So one thing that was proposed was that we make separate packages somehow, such that users would have to actively request the newer version rather than getting it automatically. (This could be done either by having a package named xen46, or by having separate centos-release-xen-46 package that pointed to an entirely different repository.) But of course, other users pushed back on that idea, saying they would always rather have the latest version Just Work, and didn't like the idea of having to manually keep track of what version was installed on any given system, or of actually finding out that there's a new package and what that new package name is. KB and I chatted about it at the Virt SIG meeting, and decided that we should touch base with the other SIGs to have a consistent message. As for my own opinion: I can see both sides of the story, and as a package maintainer I'd be willing to do either one. But on the whole I tend to side with the "latest version" crowd. I think you should always be looking at what "yum update" gives you before installing it; even minor updates I can't guarantee are going to break things. And I'd rather the installation instructions be simple (don't have to mess around with version numbers), and I'd rather people not have to fish around for major updates (or run packages with unpatched security vulnerabilities because they didn't realize their packages were now obsolete). What do other SIGs think? -George

5 6

[Atomic SIG] Signing and Publishing RPMs for ADB
by Lalatendu Mohanty 10 Feb '16

10 Feb '16

For ADB (Atomic Developer Bundle) Vagrantbox we want to add a new RPM i.e. adb-utils [2]. For that to happen we need to sign the adb-utils which we have built in cbs [3] and making it available in CentOS projects official YUM repositories (e.g. buildlogs). So that we can include the repo while building the Vagrantbox. We need a key to sign the content and also need automatic push of builds when it is tagged with "atomic7-adb-common-testing". I have also raised issues in bugs.centos.org [4] [5] for this. At this point the RPMs only make sense inside of the ADB Vagrantbox. So we expect user to use the RPMs through the Vagrantbox. [1] https://github.com/projectatomic/adb-atomic-developer-bundle [2] https://github.com/projectatomic/adb-utils [3] http://cbs.centos.org/koji/buildinfo?buildID=8707 [4] https://bugs.centos.org/view.php?id=10345 [5] https://bugs.centos.org/view.php?id=10346 Thanks, Lala

2 1

adb effort and the PaaS SIG
by Karanbir Singh 10 Feb '16

10 Feb '16

hi, With the PaaS SIG onboarding, and being centralised around OpenShift - does it make sense to consider ADB moving from the Atomic SIG into this space ? Regards -- Karanbir Singh +44-207-0999389 | http://www.karan.org/ | twitter.com/kbsingh GnuPG Key : http://www.karan.org/publickey.asc

1 0

[altarch] multiarch docker images
by Manfred Touron 08 Feb '16

08 Feb '16

Hi everyone, I'm working at Scaleway on the multiarch images system, I'm @moul on IRC and talked with @Arrfab and @Evolution (thanks a lot, guys). --- Context: Today, we are using Docker as a build-step to generate our baremetal servers rootfs, what we call, the "Scaleway images". Here is an example of "1 click app" image: - Public Hub: https://hub.scaleway.com/wordpress.html - DevHub (aggregate APIs): http://devhub.scaleway.com/#/images/wordpress:latest - Source (GitHub): https://github.com/scaleway-community/scaleway-wordpress/ - Docker Hub: https://hub.docker.com/r/scaleway/wordpress/ - Build logs: https://travis-ci.org/scaleway/qa/builds/98515979 We are quite happy today with this method and are succesfully running plenty Linux distributions (Ubuntu, openSUSE, Gentoo, Fedora, Debian, Arch Linux and Alpine Linux). Actually we also have a proof-of-concept of CentOS image, but based on an old development CubieTruck snapshot with no public mirrors. Here is how I built this POC image (I wasn't aware of virt-tar-out): https://github.com/armbuild/centos-img/blob/master/buildcontainers.sh --- We will soon release new server models with new architectures, our next priorities are: `arm64` and `x86_64`, but we plan to support more architectures if we feel enough people will need it so the development worth it. I started to work on "multiarch" images (one image per architecture). Here are some images: - https://github.com/multiarch/alpine -> built with `apk-static` in a chroot - https://github.com/multiarch/busybox -> static binaries extracted from latest busybox-static.deb - https://github.com/multiarch/ubuntu-debootstrap -> qemu-debootstrap - https://github.com/multiarch/debian-debootstrap -> same as above - https://github.com/multiarch/ubuntu-core -> based on released rootfs tarballs at https://partner-images.canonical.com/core/wily/current/ All those images also contains an `x86_64` `qemu-${ARCH}-static` so you can run them on an `x86_64` host with `binfmt-support` available. Look at the README.md files for usage and examples. I began to work on the Centos image https://github.com/multiarch/centos, I successfully built the `armhf` and `arm64` images by `virt-tar-out`ing the CubieTruck released image, i.e.: http://mirror.centos.org/altarch/7.2.1511/isos/armhfp/ I also successfully built the `x86_64` and `i386` images using kickstart: livemedia-creator --arch=i386 --make-disk \ --iso=/tmp/boot7-i386.iso \ --ks=$(pwd)/centos-7.ks \ --image-name=centos-7-base-i386 based on https://github.com/CentOS/sig-cloud-instance-build/tree/livemedia_builds/do… For now I failed to build the `ppc64` nor `ppc64le` images, but that's not a priority. I have now an issue, my `armhf` and `x86_64` images are not the same, the ones based on real boards .img.xz embed about 300M of kernel images, dtb, kernel modules, firmware (I don't want them) + a lot more apps that we can find by default in the <200Mb `x86_64` Docker image. I would like to find a solution to build "same" images for all the targets and avoid as much as possible the need of diffing the outputs of `yum list installed`. My favorite solution would be to use kickstart files, which are very cool to put on GitHub. But the non `x86_64` architectures lack of a `boot.iso` file packing `isolinux`. Is it possible to add those `boot.iso` files on the mirror or tell me how to build them ? Do you see alternative ways to build identical images ? -- Manfred Touron - @moul CTO @ scaleway.com

7 16

NO CBS/Infra Meeting 08-Feb-2016
by Brian Stinson 07 Feb '16

07 Feb '16

Hi Folks, I know some of us (myself included) are traveling or recovering from the Dojo -> FOSDEM -> cfgmgmt -> Devconf marathon, so let's cancel this week's meeting. I'll send out a list of open action items here to the list sometime between airplanes tomorrow. Cheers! --Brian

1 0

release process for SIGs
by François Cami 07 Feb '16

07 Feb '16

Hi, I'd like the centos-release-ceph-hammer rpm to be included in the extras (1) repository like the other centos-release-* packages. First things first: some of these centos-release* packages are tagged el7.centos and others el7. Niels told me el7.centos was the right tag and since this matches the build target name (core7-extras-common-el7.centos), el7.centos seems like the right target. Could I get confirmation that I should use el7.centos for centos-release-ceph-*? Then, the main problem is that packages in the extras repository should be built against core7-extras-common-el7.centos but I cannot use that build target (2). Is there an accepted way of doing this? Thanks, François (1) http://mirror.centos.org/centos-7/7.2.1511/extras/x86_64/Packages/ (2) "ActionNotAllowed: tag requires build-core permission" http://cbs.centos.org/koji/taskinfo?taskID=66398

2 1

Code review platform
by Julien Pivotto 05 Feb '16

05 Feb '16

Hi all, For my project of SIG, I would be interested to get some kind of code-review tool. It would be great to be able to review the contributions before they get accepted. Also, we could hook that into cbs and/or public ci. I have no preference regarding the tool, even if several projects I use are using gerrit. Do you think that the CentOS platform could welcome such a tool? We could imagine that we limit its scope at the beginning, then we extend that to any SIG/CentOS project that wants that. -- (o- Julien Pivotto //\ Open-Source Consultant V_/_ Inuits - https://www.inuits.eu

2 1

missing updates for CentOS6
by Jacco Ligthart 05 Feb '16

05 Feb '16

Hi All, I've been busy building updated RH packages for armv5tel (http://www.redsleeve.org/) During that process I often use repodiff to check if there are updates. And I also use repodiff to the CentOS repos. (of you didn't build it it's often also of no use to me :) ) this time around I noticed that CentOS6 was missing a couple of updates. So here a modified output of repodiff. repodiff -o ftp://ftp.redhat.com/redhat/linux/enterprise/6Server/en/os/SRPMS/ -n http://vault.centos.org/6.7/os/Source/ -n http://vault.centos.org/6.7/updates/Source/ --downgrade --simple Downgraded Packages: cups-pk-helper: cups-pk-helper-0.0.4-13.el6 -> cups-pk-helper-0.0.4-12.el6 dvd+rw-tools: dvd+rw-tools-7.1-7.el6 -> dvd+rw-tools-7.1-6.el6 environment-modules: environment-modules-3.2.10-3.el6 -> environment-modules-3.2.10-2.el6 findutils: findutils-4.4.2-8.el6 -> findutils-4.4.2-6.el6 freeipmi: freeipmi-1.2.1-7.el6 -> freeipmi-1.2.1-6.el6_5 gedit: gedit-2.28.4-4.el6 -> gedit-2.28.4-3.el6 ibus: ibus-1.3.4-9.el6 -> ibus-1.3.4-8.el6 libcanberra: libcanberra-0.22-3.el6 -> libcanberra-0.22-1.el6.centos libipathverbs: libipathverbs-1.3-3.el6_5 -> libipathverbs-1.3-3.el6 logwatch: logwatch-7.3.6-54.el6 -> logwatch-7.3.6-52.el6 pango: pango-1.28.1-11.el6 -> pango-1.28.1-10.el6 paps: paps-0.6.8-13.el6.3 -> paps-0.6.8-13.el6.2 passwd: passwd-0.77-7.el6 -> passwd-0.77-4.el6_2.2 perl-LDAP: perl-LDAP-0.40-2.el6 -> perl-LDAP-0.40-1.el6 pulseaudio: pulseaudio-0.9.21-24.el6 -> pulseaudio-0.9.21-21.el6 unzip: unzip-6.0-4.el6 -> unzip-6.0-2.el6_6 urw-fonts: urw-fonts-2.4-11.el6 -> urw-fonts-2.4-10.el6 watchdog: watchdog-5.6-5.el6 -> watchdog-5.6-4.el6_6 wget: wget-1.12-8.el6 -> wget-1.12-5.el6_6.1 Jacco

2 3

SIG-Paas request
by Troy Dawson 04 Feb '16

04 Feb '16

Hi, As a member of the OpenShift project, I am trying to get OpenShift Origin into CentOS. This effort would create CentOS OpenShift Origin rpm's, along with their dependencies. It would also involve creating CentOS OpenShift Origin docker images. It could also involve creating virtual machine and/or cloud images. Virtual machine and cloud images are not part of our original wants and needs, but if someone were to help create them, we wouldn't say no. I am requesting a SIG, named SIG-Paas. If the committee feels that is too broad, then we would like SIG-OpenShift. Thank You Troy Dawson

7 10

Yum missing from Atomic image
by Bent Terp 04 Feb '16

04 Feb '16

Hi, this may be a deliberate choice but I can't really figure out why one would make that decision... [centos@atomic ~]$ sudo yum update sudo: yum: command not found [centos@atomic ~]$ rpm -ql yum | grep bin /usr/bin/yum [centos@atomic ~]$ ls -l /usr/bin/yum ls: cannot access /usr/bin/yum: No such file or directory [centos@atomic ~]$ I used the latest http://cloud.centos.org/centos/7/atomic/images/CentOS-Atomic-Host-7-Generic… in an RDO deploy. I don't experience the same issue with the regular image only with the Atomic. regards, Bent

3 3

[altarch?] Naming scheme and lorax parameters for userland + non-distro kernel builds
by Brian Stinson 03 Feb '16

03 Feb '16

Hi All, I'm working on putting some things together for a couple of the Intel boards (Edison, Jaguarboard, etc.) and was wondering if there's a standard way we should be naming things on the install media. For the rpi2, we use 'CentOS Altarch 7 (1511) Userland for armhfp' A lot of the Intel boards out there require a newer or otherwise non-distro kernel so I was considering titling the install media like this: CentOS Altarch 7 (1511) Userland for x86_64 Not sure if the 'Altarch' bit applies because it's the standard x86_64 userland (no rebuilding needed). Cheers! --Brian

3 2

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

devel February 2016