devel January 2016

devel@lists.centos.org

48 participants
32 discussions

CentOS 7.2 cloud images for ppc64le
by Xiandong Meng 14 Jun '16

14 Jun '16

Hi all, I noticed that there are centos 7 cloud images for x86. But i did not find the cloud images for ppc64le. Are there any work on it within this community? Or is there any plan to work on it? Regards, Xiandong Meng <mengxiandong(a)gmail.com> mengxiandong(a)gmail.com

2 6

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

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

[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

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

How to create CentOS 7.2 ppc64 image
by Vit Ry 30 Jan '16

30 Jan '16

Hi all. How can I create CentOS 7.2 image for ppc64 arch? Are there any scripts, guides, etc..(is it open source..?)? What is difference between build envs for RH 7.0 and CentOS 7.2 ..? I see, RH uses yaboot as bootloader and so on, when C7.2 - uses grub... --- Regards, Vit Ry.

2 2

Fwd: sig-core-t_functional suite for 7.2.1511
by Thierry Fauck＠linux.vnet.ibm.com 29 Jan '16

29 Jan '16

As talked earlier today, Here are my comments - I understand they are mostly related to the environment it is running into, but it may be interesting to look into it to improve the cases. Regards Thierry (nick - thf ) -------- Forwarded Message -------- Subject: sig-core-t_functional suite for 7.2.1511 Date: Tue, 22 Dec 2015 11:14:29 +0100 From: Thierry Fauck(a)linux.vnet.ibm.com <thierry(a)linux.vnet.ibm.com> To: centos-qa(a)centos.org Hello, While checking the suite for ppc64 and ppc64le, I got the following changes and comment - let me know if you want a diff file. - runtest.sh If script fails and you relaunch it a second time, you may get a failure because tests users are already created - I suggested to remove all test users in runtests.sh to have a clean environment. In t_CheckExitStatus shouldn't we add the test name being run to ease tracing and script debugging like ? [ $1 -eq 0 ] && { t_Log "$0 PASS"; return $PASS; } - Depending on your setting, you may come with LC_ALL setting as utf8 or C and 10-cmp-tests may fails - checking for byte or char would allow both settings to be run. cmp -i 15:16 ${FILE}a ${FILE}b | egrep -q "byte 1, line 1|char 1, line 1"|| ExitFail - tests/p_mtr/mtr_test.sh fails if you are not in the centos.org env as it tries to reach the qa host. As this test is done at start of runtest.sh, in case it is not true, shouldn't you use another hosts found on your local network (an easy bypass is to use like the local DNS server or the gateway ?). - p_shadow-utils/20-chage_tests may fails because of capital letters jan versus Jan - I would suggest to use chage -l testshadow | grep Last | grep -qi "Jan 01, 2013" - tests/p_shim/01_shim_secureboot_signed.sh tests/p_kernel/02_kernel_secureboot_signed.sh are x86_64 dependant - Had to increase -T delay from 2 to 8 in tests/p_squid/squid_test.sh - tests/p_systemd/25-journalctl.sh - for fast processors, you need to had a delay for journalctl to cope with your message - I added a sleep 1 after writting to /dev/kmsg - tests/p_which/which_basic.sh - sounds like wich is now in /bin instead of /usr/bin - /usr/bin/which which | grep -q "/bin/which" would cope with both cases Hope that helps -- Thierry

1 0

Why is systemd user instance support disabled in CentOS 7 / RHEL 7?
by Tadej Janež 28 Jan '16

28 Jan '16

Hi! I'm preparing a backup package for Fedora/EPEL 7 [1] that provides a web server with a systemd unit file for running with the systemd user instance (i.e. 'systemd --user'). As reported in [2], systemd user instance support is disabled in CentOS 7 due to being disabled in the upstream RHEL package. It is the same with the latest systemd for CentOS 7 [3]. Unfortunately, the original Red Hat Bugzilla report [4] is not public so I'm asking here if anyone has more information why this is disabled? Thanks and regards, Tadej [1] https://apps.fedoraproject.org/packages/bup [2] https://bugs.centos.org/view.php?id=8767 [3] https://git.centos.org/blob/rpms! systemd/1ff63606a64d814f37c4956854107e049a684754/SOURCES! 0004-remove-user-.service.patch [4] https://bugzilla.redhat.com/show_bug.cgi?id=1019738

4 5

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

devel January 2016