*** sshnaidm is now known as sshnaidm|afk | 00:28 | |
tristanC | jangutter: there seems to be some documentation about EFI with dib, e.g. https://kensavich.com/2018/07/03/building-a-customized-uefi-rhel-7-5-image-for-openstack-with-diskimage-builder/ | 01:45 |
---|---|---|
tristanC | jangutter: however we trying to move away from dib complexicity, and you can actually use virt-customize instead | 01:45 |
tristanC | jangutter: for example, our cloud-centos-7 image defined here: https://softwarefactory-project.io/cgit/config/tree/nodepool/rdo-cloud.yaml#n17 | 01:46 |
tristanC | jangutter: uses a trick with PATH env to run another disk-image-create script: | 01:46 |
tristanC | https://softwarefactory-project.io/cgit/config/tree/nodepool/elements/virt-customize | 01:47 |
tristanC | we have a story to integrate (or upstream) this trick, but i would recommend you try this element to build disk image | 01:48 |
*** bhavikdbavishi has joined #softwarefactory | 02:02 | |
*** bhavikdbavishi has quit IRC | 02:39 | |
*** bhavikdbavishi has joined #softwarefactory | 03:24 | |
*** bhavikdbavishi has quit IRC | 03:28 | |
*** bhavikdbavishi has joined #softwarefactory | 03:42 | |
sfbender | Merged software-factory/sf-ci master: Update tests for new ELK stack https://softwarefactory-project.io/r/15105 | 04:34 |
*** raukadah is now known as chandankumar | 04:56 | |
tristanC | jruzicka: but wait, isn't dlrn --order the same thing as i was asking before? :) | 06:09 |
sfbender | Merged software-factory/sf-config master: cgit: fix multinode deployment without gerrit https://softwarefactory-project.io/r/15125 | 07:07 |
sfbender | Merged software-factory/sf-config master: Set DocumentRoot explicitly in logserver conf https://softwarefactory-project.io/r/15135 | 07:11 |
jangutter | tristanC a bunch of nice people on #openstack-dib helped me to get something booting! | 07:49 |
jangutter | tristanC - but long story short, it looks like if you figure out the magic sequence, it's perfectly viable to enlist baremetal nodes in a nodepool, booted via ironic. | 07:50 |
jangutter | tristanC - with no funny hacks. | 07:50 |
tristanC | jangutter: that's great to hear! | 07:58 |
sfbender | Fabien Boucher created software-factory/sfinfo master: Add distroinfo in sfinfo https://softwarefactory-project.io/r/15155 | 08:03 |
sfbender | Merged software-factory/sfinfo master: Add distroinfo in sfinfo https://softwarefactory-project.io/r/15155 | 08:07 |
sfbender | Fabien Boucher created software-factory/sf-config master: DNM: validate DLRN/RDOPKG bump https://softwarefactory-project.io/r/15156 | 08:10 |
*** jpena|off is now known as jpena | 08:55 | |
jruzicka | tristanC, oh hello! | 09:18 |
jruzicka | I'm super early today ;) | 09:18 |
jruzicka | tristanC, distroinfo integration for rpmreq is definitely possible ;) | 09:19 |
tristanC | jruzicka: but it seems like dlrn order option would do the trick (to figure out distro rebuild order...) | 09:21 |
jruzicka | tristanC, but yes, order might do what you want :) | 09:22 |
jruzicka | dlrn --order that is | 09:22 |
jruzicka | tristanC, huh, and you managed to run rpmreq on py2 without changes to it? cause I see no open reviews... | 09:23 |
tristanC | jruzicka: it worked, but had to hack one of the requirements | 09:26 |
*** sshnaidm|afk has quit IRC | 09:26 | |
jruzicka | hawkey? | 09:26 |
tristanC | iirc pyrpm assumed unicode and used a py3 super() sugar | 09:26 |
jruzicka | ah pyrpm | 09:27 |
jruzicka | tristanC, u goona push those changes to master? | 09:28 |
jruzicka | *gonna | 09:28 |
tristanC | jruzicka: it was really dirty sed action to get it working, we just wanted to test the cli | 09:30 |
jruzicka | ack :) | 09:31 |
jruzicka | tristanC, so it's working as expected? | 09:31 |
tristanC | jruzicka: yes, after fixing pyrpm, we were able to run rpmreq with the spec files, and it found the order | 09:35 |
tristanC | jruzicka: but then we saw that dlrn could do that too, without having to gather all the spec file manually, so we went that way instead | 09:36 |
jruzicka | tristanC, nice validation, though :) thanks for info. | 09:51 |
tristanC | jpena: btw, we made some good progress on a rpm-build job, this time using the new zuul requires/provides logic | 10:10 |
jpena | nice! | 10:10 |
tristanC | jpena: the idea is to re-use the buildset repository of items ahead in zuul queues | 10:11 |
tristanC | jpena: so that, instead of building each items ahead, we can just add the items ahead artifact location to the mock config | 10:12 |
jruzicka | neat | 10:22 |
tristanC | jruzicka: hopefully there will be a demo by the end of the sprint | 10:23 |
jruzicka | tristanC, looking forward ;) | 10:23 |
tristanC | another cool thing is that we may be able to generate the rdo.yml file from the resources | 10:23 |
tristanC | when adding a repo, there may be a "dlrn-project-template" option, and when set it will be automatically set in further job | 10:24 |
*** sshnaidm|afk has joined #softwarefactory | 10:54 | |
*** sshnaidm|afk has quit IRC | 10:57 | |
*** bhavikdbavishi has quit IRC | 11:16 | |
*** sshnaidm|afk has joined #softwarefactory | 12:01 | |
*** bhavikdbavishi has joined #softwarefactory | 12:19 | |
*** jpena is now known as jpena|lunch | 12:31 | |
*** sshnaidm|afk is now known as sshnaidm | 13:32 | |
*** jpena|lunch is now known as jpena | 13:34 | |
*** bhavikdbavishi has quit IRC | 13:36 | |
*** rfolco|pto is now known as rfolco|ruck | 14:02 | |
*** chandankumar is now known as chkumar|pto | 15:03 | |
*** matburt has quit IRC | 15:47 | |
*** matburt has joined #softwarefactory | 17:00 | |
*** jpena is now known as jpena|off | 18:19 | |
*** rfolco|ruck has quit IRC | 19:26 | |
*** rfolco has joined #softwarefactory | 19:27 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!