Wednesday, 2014-01-29

openstackgerrit	Tzu-Mainn Chen proposed a change to openstack/tuskar-ui: Combined Instance and Node objects into Node https://review.openstack.org/69724	00:02
*** CaptTofu has joined #tripleo		00:02
*** derekh has quit IRC		00:05
*** jcooley_ has quit IRC		00:10
openstackgerrit	lifeless proposed a change to openstack/tripleo-heat-templates: Update overcloud to support N compute hosts. https://review.openstack.org/69548	00:14
lifeless	SpamapS: I missed a ref in https://review.openstack.org/#/c/69548/	00:16
*** derekh has joined #tripleo		00:17
*** jcooley_ has joined #tripleo		00:17
openstackgerrit	Derek Higgins proposed a change to openstack/diskimage-builder: Run environment.d scripts outside of chroot https://review.openstack.org/69750	00:23
openstackgerrit	Derek Higgins proposed a change to openstack/diskimage-builder: Add pip support to source-repositories https://review.openstack.org/69751	00:23
openstackgerrit	Steve Kowalik proposed a change to openstack/tripleo-image-elements: Add a debian-mirror element https://review.openstack.org/68308	00:24
openstackgerrit	Derek Higgins proposed a change to openstack/tripleo-image-elements: Set PIP_INDEX_URL to the openstack pypi mirror https://review.openstack.org/69752	00:26
openstackgerrit	Derek Higgins proposed a change to openstack/tripleo-image-elements: Use source-repositories to download os-apply-config https://review.openstack.org/69753	00:26
openstackgerrit	Derek Higgins proposed a change to openstack/tripleo-image-elements: Use source-repositories to download os-refresh-config https://review.openstack.org/69754	00:26
openstackgerrit	Derek Higgins proposed a change to openstack/tripleo-image-elements: Use source-repositories to download os-collect-config https://review.openstack.org/69755	00:26
*** matsuhashi has joined #tripleo		00:30
*** markmcclain has quit IRC		00:31
*** markmcclain has joined #tripleo		00:31
derekh	lifeless: pleia2: I'm looking at saving logs from the seed for the ci job , any idea if writing to the logs dir like http://paste.openstack.org/show/62056/ and changing the scp macro like http://paste.openstack.org/show/62057/ will do the trick	00:37
openstackgerrit	lifeless proposed a change to openstack/tripleo-heat-templates: Prep work for a scalable control plane. https://review.openstack.org/69758	00:37
derekh	just don't know if I in the correct working directory to create the logs dir in	00:37
*** CaptTofu has quit IRC		00:38
lifeless	derekh: I think it would be useful	00:39
lifeless	derekh: but why are you discarding the pip log ?	00:40
lifeless	derekh: https://review.openstack.org/#/c/69752/ is buggy/ unneeded	00:40
lifeless	derekh: we already have pypi mirror support	00:40
lifeless	derekh: https://review.openstack.org/#/c/69309/	00:40
*** cd-undercloud has joined #tripleo		00:41
cd-undercloud	************ overcloud complete status=1 **********	00:41
*** cd-undercloud has quit IRC		00:41
derekh	according to the logs the pip log regex isn't finding anything	00:41
lifeless	derekh: I'm confused by putting pip into source-repositories	00:41
derekh	2014-01-28 14:22:57.480 \| [SCP] No file(s) found: **/pip.log	00:41
lifeless	derekh: ah, fair enough	00:41
lifeless	derekh: but we probably do want to gather e.g. the dib build logs	00:41
derekh	lifeless: we currently install os-*-config with a pip install we need to install from git in reviews for thos projects so we need a way to switch from pip to git	00:42
derekh	lifeless: the dib logs are currently echo'ed to console.html	00:43
lifeless	derekh: there's a fundamental conceptual problem with installing from git though	00:44
lifeless	derekh: dependencies will be pulled in from pip	00:44
derekh	lifeless: yes they will but in the same way nova deps are, taking os-apply-config as an example , we would want to test o-a-c from git for o-a-c reviews and pull all deps from releases	00:46
derekh	I submitted https://review.openstack.org/#/c/69752/1 so that pip support in source-repositories would pull form pypi.openstack.org instead of the default, why is it buggy?	00:47
lifeless	derekh: so in gate you can have a branch of o-a-c being tested with a branch of o-r-c	00:47
openstackgerrit	James Slagle proposed a change to openstack/diskimage-builder: Add package map for stunnel4 https://review.openstack.org/69762	00:47
lifeless	derekh: because we already have pypi mirror support setup for the gate - see the review I linked	00:47
lifeless	derekh: it make all our pip calls use the pypi.openstack.org mirror	00:48
SpamapS	lifeless: didn't you rename 'ComputeKey' to 'NovaCompute0Key' ?	00:48
lifeless	SpamapS: yes, I missed one instance of it in my first push	00:49
openstackgerrit	James Slagle proposed a change to openstack/tripleo-image-elements: SSL support for openstack services. https://review.openstack.org/67343	00:49
derekh	lifeless: it would for pip calls inside the chroot, my change alters the default for pip calls outside the chroot (i.e. in source-repositories)	00:49
lifeless	derekh: right, but source-repositories doesn't /install/ anywhere else, so I think this is a confusing design	00:50
derekh	lifeless: my change to source-repositories doesn't install, it downloads the tarball and unzipps it, the change to os-apply-config then installs the unzipped directory	00:51
lifeless	derekh: I don't think that will work as binary packages evolve w/pip - wheels	00:52
SpamapS	lifeless: https://review.openstack.org/#/c/69548/3/overcloud-source.yaml <-- if I am calling with -P ComputeKey=baz , I'll get a fail after this merges	00:52
derekh	lifeless: the os-apply-config elements can then get its source from either pip tarball (unzipped to /opt/stack/os-apply-config) or git (cloned to /opt/stack/os-apply-config)	00:52
lifeless	derekh: so hang on	00:52
lifeless	derekh: lets take a step back here	00:52
lifeless	derekh: I think the basic design approach is broken, and arguing about the details won't fix it	00:53
SpamapS	lifeless: we don't do that, but I don't know if tuskar does.	00:53
lifeless	SpamapS: ComputeKey isn't a parameter.	00:53
lifeless	SpamapS: so HTF could we do that today?	00:53
lifeless	derekh: the basic failure design problem is that we are tackling the one element but we need to override potentially N different elements	00:54
SpamapS	oh man I read that entirely wrong	00:55
SpamapS	lifeless: sorry I was reading it upside down	00:55
lifeless	derekh: we need to hook in somewhere where pip install <name>- which is what requirements files use!	00:56
lifeless	derekh: will DTRT.	00:56
derekh	lifeless: sorry I'm lost, I still don't see the problem with what I've done	00:59
*** meera has quit IRC		00:59
lifeless	derekh: so the goal is to make the $tool baked into images be the one from the ZUUL_REF right ?	00:59
SpamapS	hm	01:00
lifeless	derekh: so for instance, python-neutronclient should be the ZUUL_REF'd python-neutronclient when the test is a python-neutronclient one, systematically across all elements	01:00
SpamapS	can't ssh to the spawned instance	01:00
lifeless	like, the one nova has in it's vms	01:00
derekh	lifeless: yes, so the follow on from this would be , if ZUUL_REF == os-apply-config then DIB_REPOTYP=git DIB_REPOLOC=/opt/stc/new/o-a-c	01:01
lifeless	derekh: the one in nova's venv is installed recursively by pip install -r requirements.txt	01:01
SpamapS	root@overcloud-novacompute0-ils4ty7674ol:~# ip netns	01:01
lifeless	derekh: the problem is that you've taken a point-by-point solution rather than a systematic one	01:01
SpamapS	root@overcloud-novacompute0-ils4ty7674ol:~#	01:01
SpamapS	no namespaces	01:01
lifeless	SpamapS: normal	01:01
lifeless	SpamapS: no netns's on the compute node, they are all on the network node	01:02
SpamapS	oh we just attach right to tuns?	01:02
SpamapS	or taps or whatever those things ovs makes	01:02
lifeless	derekh: if I may suggest an alternative implementation, how about we build a little pip mirror with the snapshots we want to install and set that as primary index, the pypi.o.o mirror as a secondary index	01:03
lifeless	derekh: the snapshot versions will be higher than than the released ones in the pypy.o.o mirror	01:03
lifeless	derekh: and when this is active we can then choose a policy about what we put in the little pip mirror	01:04
SpamapS	no vm running	01:04
lifeless	SpamapS: yes, it needs to be stopped and started after a distruptive deploy - see _overcloud at theend	01:04
SpamapS	tripleo-cd stopped	01:04
SpamapS	lifeless: I think we got past the stop/start	01:04
lifeless	derekh: one policy would be 'only things with a non-master ZUUL ref', which would test 'changed things against releases'	01:04
lifeless	derekh: another policy would be 'put trunk of everything into the mirror' which would test 'changed things against git' (which is what devstack tests)	01:05
lifeless	derekh: what do you think?	01:05
derekh	lifeless: ahh I see what your saying now, so this could be used to bring the three os-*-config elements to the same level as the other elements, we could then add the systematic approach on top of that?	01:05
derekh	lifeless: but I get what your talking about	01:06
derekh	lifeless: whch could be done without any of my patches	01:06
lifeless	yup	01:06
lifeless	less code to maintain, less special cases	01:06
lifeless	the os-*-config elements would DTRT from the get go	01:06
lifeless	the nova/neutron/etc things using source-repositories probably do need an option to say 'just pip install' to work in this mode, but OTOH they already install from git trunk/ZUUL_REF, so that can really wait for us doing an explicit against-release scenario for them	01:07
lifeless	(and also note that nova etc aren't in pypi so we do need to keep doing them differently for now anyhow)	01:08
derekh	lifeless: ok, so I can take a stab at that, I still think the pip support in source-repositories (maybe I'm biased) so that people and use it to install from git if developing	01:08
*** akuznetsov has quit IRC		01:09
derekh	lifeless: but what your saying does lower its priority for CI	01:09
lifeless	derekh: I think it's adding complexity we don't need; I'd like to see a use case for it first (one that a systemic approach doesn't cover)	01:09
SpamapS	hm	01:11
SpamapS	dunno what went wrong	01:11
derekh	usecase : I'm developing a feature in os-apply-config and can't currently run devtest from my local git repo as it downloads the release, with the change I could DIB_REPOTYPE_os_apply_config=git DIB_REPOLOCATION_os_apply_config=/path/to/o-a-c	01:11
SpamapS	we need to do a better job in wait_for of printing the command that failed	01:11
lifeless	derekh: I don't have the incantation to hand but its trivial to push that into a local pip repo	01:12
derekh	lifeless: ok	01:13
lifeless	derekh: pypi-mirror has stuff to aid with this, and I'm sure a couple of extra tweaks will be well appreciate	01:14
derekh	lifeless: ok, will take a look	01:14
lifeless	cool, thanks	01:15
lifeless	and sorry to be raining on your s-r parade :)	01:15
derekh	lifeless: no prob, remind me to wait until your asleep next time :-)	01:16
lifeless	AHAHAH that is not the way it works	01:16
lifeless	derekh: as an alternative, you might try chatting about what you're doing, here or list :)	01:17
lifeless	(if you're worried about rework, that is)	01:17
*** CaptTofu has joined #tripleo		01:19
derekh	lifeless: only messing, yup will do , all abandoned https://review.openstack.org/#/q/status:abandoned+branch:master+topic:pip-type,n,z	01:19
SpamapS	derekh: I wear my abandoned patches like badges of honor	01:20
lifeless	my cat is cray cray :)	01:20
lifeless	lying upside down on my lap purrin	01:20
*** CaptTofu has quit IRC		01:21
derekh	SpamapS: will print them out for my biography	01:21
*** CaptTofu has joined #tripleo		01:21
openstackgerrit	A change was merged to openstack/tripleo-incubator: Stop sourcing devtest_seed.sh. https://review.openstack.org/69286	01:24
lifeless	neutron dhcp agent on ci-overcloud naffed again	01:27
lifeless	(fixed)	01:34
*** derekh has quit IRC		01:36
*** nosnos has joined #tripleo		01:37
*** cwolferh has quit IRC		01:54
*** cd-undercloud has joined #tripleo		01:59
cd-undercloud	************ overcloud complete status=0 **********	01:59
*** cd-undercloud has quit IRC		01:59
lifeless	\o/	02:00
lifeless	StevenK: so hi	02:13
lifeless	and dhcp agent dead again	02:21
*** spzala has quit IRC		02:31
StevenK	lifeless: O hai	02:35
lifeless	StevenK: so wassup? can I offload these nova patches?	02:36
StevenK	lifeless: Sure	02:36
StevenK	lifeless: I am about to disappear to the court house, but I'm happy to look over them before I head off	02:36
lifeless	StevenK: court!	02:37
StevenK	lifeless: Not for court, for a JP	02:37
StevenK	lifeless: I need a notarized copy of my primary ID for the amex	02:38
lifeless	StevenK: StevenK https://review.openstack.org/#/c/69060/ and https://review.openstack.org/#/c/69108/	02:38
lifeless	StevenK: yay paperwork	02:38
StevenK	lifeless: Okay, and you want to me to make jenkins love them and push them through?	02:40
lifeless	StevenK: make them into good patches - tests etc	02:40
lifeless	StevenK: I'm debugging ops problems atm that only -admins can do	02:40
lifeless	so I can offer advice	02:41
lifeless	just doing it myself would be bottlenecking	02:41
StevenK	lifeless: I've pulled the latest nova, snarfed the patch, I'm going to read through git show while becoming shoes-enabled	02:42
lifeless	StevenK: ther are two patches ;)	02:43
StevenK	Sure, but the second depends on the first	02:44
lifeless	yah	02:44
lifeless	not semantically IIRC, just a convenience when making it	02:45
*** cd-undercloud has joined #tripleo		02:46
cd-undercloud	************ overcloud complete status=0 **********	02:46
*** cd-undercloud has quit IRC		02:46
StevenK	Hm, is that two in a row?	02:47
lifeless	yup	02:48
lifeless	back to sunshine and happiness	02:48
StevenK	SpamapS: Your champagne is now due for drinking	02:48
lifeless	StevenK: that was monday actually :)	02:48
lifeless	wow, the dnsmasq agent restart is expensive	02:54
lifeless	and badly sequenced	02:54
lifeless	it kills dnsmasq	02:58
lifeless	then makes new config files	02:58
pleia2	lifeless: have some time for a quick chat? trying to figure out if I have the right testing methodology now	02:58
lifeless	sure	02:58
lifeless	here? voice?	02:58
pleia2	voice should make it quick	02:59
*** CaptTofu has quit IRC		03:00
lifeless	so in theory kill -HUP is meant to reload dnsmasq	03:16
lifeless	and yay its bust again	03:17
lifeless	lets see if -HUP fixes it	03:17
*** vkozhukalov has joined #tripleo		03:17
lifeless	I wonder i this is a bug in dnsmasq	03:22
*** weshay has quit IRC		03:29
*** matsuhashi has quit IRC		03:30
*** matsuhashi has joined #tripleo		03:31
lifeless	also yay	03:33
lifeless	[620235.560911] WARNING: CPU: 5 PID: 0 at /build/buildd/linux-3.11.0/net/sched/sch_generic.c:260 dev_watchdog+0x	03:33
lifeless	26a/0x280()	03:33
lifeless	[620235.560915] NETDEV WATCHDOG: eth2 (mlx4_core): transmit queue 15 timed out	03:33
lifeless	[620235.830394] mlx4_en: eth2: Link Down	03:33
lifeless	[620235.950597] mlx4_en: eth2: Link Up	03:33
lifeless	thats not super good	03:33
clarkb	whats the story behind mellanox? can you not pop in some $otherchipset devices and move on?	03:35
*** cd-undercloud has joined #tripleo		03:35
cd-undercloud	************ overcloud complete status=0 **********	03:35
*** cd-undercloud has quit IRC		03:35
*** matsuhashi has quit IRC		03:35
clarkb	seems like that hardware has caused a lot of pain	03:36
lifeless	AHHA	03:36
lifeless	I've fixed out the dhcp agent bug	03:36
lifeless	clarkb: maybe	03:36
lifeless	clarkb: however, I don't know if we're running on ethernet or IB emulation cray cray	03:37
*** jcooley_ has quit IRC		03:41
pleia2	lifeless: doing something network-wise? got my image uploaded and testing and getting: BadRequest: Network 874b5d72-39db-4350-809e-cd45f6dd4049 could not be found.	03:48
lifeless	pleia2: thats the bare metal network you don't ahve access to	03:48
lifeless	pleia2: remove the networks section from your yaml config file	03:48
pleia2	ok, thanks	03:48
lifeless	pleia2: or	03:48
lifeless	pleia2: create a new private network of your own and reference it instead of 87...	03:49
* pleia2 nods		03:49
pleia2	lifeless: so I applied your patch and I'm not seeing a difference when I run nodepool, still spinning at INFO:paramiko.transport:Authentication (publickey) failed.	04:05
lifeless	pleia2: thats probably using the infra key	04:05
pleia2	bah	04:05
lifeless	pleia2: have you put keypair: '<some keypair name>' in your yaml?	04:05
lifeless	I added that (in providers section) as a debug aid	04:05
pleia2	yeah, it's looking at /home/elizabeth/.ssh/id_rsa	04:05
pleia2	private-key: /home/elizabeth/.ssh/id_rsa	04:06
lifeless	pleia2: ah no, I mean	04:06
lifeless	nova keypair-create --pub-key ~/.ssh/id_rsa.pub moi	04:06
lifeless	then	04:06
lifeless	providers:	04:06
lifeless	- name: ci-overcloud	04:06
pleia2	ah!	04:06
lifeless	keypair: moi	04:06
pleia2	some day I'll learn this openstack thing :x	04:06
lifeless	...	04:06
lifeless	and it will use the ssh agent to log in, but spawn the server with that keypair	04:07
pleia2	got it, thanks	04:07
lifeless	rather than the dynamically created pairs it normally uses to bootstrap, which then install the infra pubkey from puppet	04:07
lifeless	the private-key thing, AIUI tweaks the puppet rules	04:07
lifeless	but	04:07
lifeless	you have to get that far first for it to be useful	04:07
clarkb	lifeless: that depends on a passphraseless keys or an ssh-agent right?	04:08
lifeless	yes	04:08
lifeless	its a testing feature	04:08
lifeless	though a similar thing is probably what I'll do to reduce keypair churn, when I get time to hack on nodepool again	04:10
*** CaptTofu has joined #tripleo		04:16
openstackgerrit	A change was merged to openstack/tripleo-image-elements: Add a debian-mirror element https://review.openstack.org/68308	04:23
*** cd-undercloud has joined #tripleo		04:23
cd-undercloud	************ overcloud complete status=0 **********	04:23
*** cd-undercloud has quit IRC		04:23
pleia2	need to pick this up in the morning, fixed up permissions on my ssh key but them bumped into other auth issues I'll need to continue sorting out	04:33
*** edmund1 has quit IRC		04:44
lifeless	slagle: thanks!	04:46
lifeless	SpamapS: and uhm ng: https://review.openstack.org/#/c/67343/10	04:46
openstackgerrit	A change was merged to openstack/tripleo-image-elements: Updates permissions on 97-mysql-iptables https://review.openstack.org/69639	04:47
jamezpolley	lifeless: I've been watching the list but haven't seen any more discussion about timing for the meetup, except for ironic wanting to join in. I did see someone pop in here and say that march3 works for them. Have I missed anything else?	04:50
lifeless	nope	04:50
lifeless	I'm going to nag markmc tonight	04:51
*** jcoufal-mobile has joined #tripleo		04:52
*** matsuhashi has joined #tripleo		04:55
*** markmcclain has quit IRC		04:58
jamezpolley	Aside from Stevenk and I, do we have any other regular contributors around Sydney?	04:59
StevenK	jamezpolley: You're keen :-P	05:01
lifeless	jamezpolley: in openstack as a whole, I think so, tripleo specifically, not AFAIK	05:01
StevenK	lifeless: Which list is jamezpolley talking about, -dev?	05:07
lifeless	think so	05:07
StevenK	I should probably subscribe	05:07
StevenK	lifeless: I've subscribed, but I'm sure you'll tell me where I need to be.	05:13
lifeless	you want me to tell you where to go?	05:14
*** cd-undercloud has joined #tripleo		05:14
cd-undercloud	************ overcloud complete status=0 **********	05:14
*** cd-undercloud has quit IRC		05:14
StevenK	lifeless: I didn't phrase it like that for a very good reason. :-)	05:14
lifeless	:>	05:14
StevenK	Hm, nova are you supposed to run 8 concurrent test workers ...	05:15
lifeless	StevenK: testr...	05:16
lifeless	StevenK: check .testr.conf, it may be overriding concurrency	05:16
lifeless	lol, I have a cat snoring by my ear	05:16
lifeless	its a tad distractamundo	05:16
StevenK	lifeless: It does not override	05:17
lifeless	StevenK: how many CPUs do you have?	05:17
StevenK	8	05:17
lifeless	then yes :)	05:17
StevenK	Well, 4 cores, ht	05:17
*** jcooley_ has joined #tripleo		05:22
*** CaptTofu has quit IRC		05:22
*** jcoufal-mobile_ has joined #tripleo		05:28
*** jcoufal-mobile has quit IRC		05:28
StevenK	Blah, pdb no worky	05:31
*** jcoufal-mobile_ has quit IRC		05:32
*** jcoufal-mobile_ has joined #tripleo		05:33
*** matsuhashi has quit IRC		05:34
lifeless	StevenK: within testr? not so much	05:35
*** matsuhashi has joined #tripleo		05:35
lifeless	StevenK: python -m testtools.run <same args as the testr subunit.run calls> --load-list <file with tesetname in it>	05:35
*** jcoufal-mobile_ has quit IRC		05:38
*** nosnos has quit IRC		05:38
*** nosnos has joined #tripleo		05:41
*** jcoufal-mobile_ has joined #tripleo		05:44
*** jcoufal-mobile_ has quit IRC		05:49
*** jcoufal-mobile_ has joined #tripleo		05:49
*** jcooley_ has quit IRC		05:50
*** jcoufal-mobile_ has quit IRC		05:56
openstackgerrit	lifeless proposed a change to openstack/tripleo-incubator: Stop sourcing devtest_undercloud. https://review.openstack.org/69797	06:01
*** cd-undercloud has joined #tripleo		06:02
cd-undercloud	************ overcloud complete status=0 **********	06:02
*** cd-undercloud has quit IRC		06:02
StevenK	lifeless: Aren't you missing the actual variables in undercloudrc?	06:04
lifeless	ahahahahah	06:04
lifeless	thanks	06:04
lifeless	thanks	06:05
openstackgerrit	lifeless proposed a change to openstack/tripleo-incubator: Stop sourcing devtest_undercloud. https://review.openstack.org/69797	06:05
*** jcooley_ has joined #tripleo		06:06
*** vkozhukalov has quit IRC		06:13
*** matsuhashi has quit IRC		06:23
*** matsuhas_ has joined #tripleo		06:25
*** rlandy has joined #tripleo		06:40
*** e0ne has joined #tripleo		06:41
*** akuznetsov has joined #tripleo		06:42
*** e0ne has quit IRC		06:47
*** cd-undercloud has joined #tripleo		06:53
cd-undercloud	************ overcloud complete status=0 **********	06:53
*** cd-undercloud has quit IRC		06:53
*** jcooley_ has quit IRC		06:53
*** rdopieralski has joined #tripleo		06:53
*** jcooley_ has joined #tripleo		06:55
*** noslzzp has quit IRC		07:08
*** akuznetsov has quit IRC		07:08
*** jeremydei has quit IRC		07:11
*** sdake has quit IRC		07:14
*** coolsvap has joined #tripleo		07:19
*** jeremydei has joined #tripleo		07:20
*** pblaho has joined #tripleo		07:20
*** CaptTofu has joined #tripleo		07:23
*** CaptTofu has quit IRC		07:27
*** sdake has joined #tripleo		07:36
*** mrunge has joined #tripleo		07:38
*** mrunge has quit IRC		07:38
*** cd-undercloud has joined #tripleo		07:39
cd-undercloud	************ overcloud complete status=0 **********	07:39
*** cd-undercloud has quit IRC		07:39
*** mrunge has joined #tripleo		07:39
*** ifarkas has joined #tripleo		07:41
*** jistr has joined #tripleo		07:45
*** rwsu has quit IRC		07:46
*** rwsu has joined #tripleo		07:49
*** akuznetsov has joined #tripleo		07:50
*** jprovazn has joined #tripleo		07:54
*** vkozhukalov has joined #tripleo		07:55
openstackgerrit	A change was merged to openstack/tripleo-incubator: Ensure libvirtd is always configured on Red Hat https://review.openstack.org/69627	07:55
*** akuznets_ has joined #tripleo		08:14
*** akuznetsov has quit IRC		08:16
*** akuznets_ has quit IRC		08:16
*** jprovazn has quit IRC		08:20
*** akuznetsov has joined #tripleo		08:25
*** cd-undercloud has joined #tripleo		08:26
cd-undercloud	************ overcloud complete status=0 **********	08:26
*** cd-undercloud has quit IRC		08:26
*** jcooley_ has quit IRC		08:30
*** akuznetsov has quit IRC		08:41
*** akuznetsov has joined #tripleo		08:41
*** jprovazn has joined #tripleo		08:42
*** markmc has joined #tripleo		08:48
*** jtomasek has joined #tripleo		08:55
*** matsuhas_ has quit IRC		09:04
*** matsuhashi has joined #tripleo		09:04
*** athomas has joined #tripleo		09:11
*** cd-undercloud has joined #tripleo		09:11
cd-undercloud	************ overcloud complete status=0 **********	09:11
*** cd-undercloud has quit IRC		09:11
*** d0ugal has quit IRC		09:12
*** d0ugal has joined #tripleo		09:12
*** matsuhashi has quit IRC		09:17
*** derekh has joined #tripleo		09:18
*** matsuhashi has joined #tripleo		09:19
*** jcooley_ has joined #tripleo		09:21
*** viktors has joined #tripleo		09:21
*** CaptTofu has joined #tripleo		09:24
*** gcha has joined #tripleo		09:26
*** CaptTofu has quit IRC		09:28
*** e0ne has joined #tripleo		09:30
*** coolsvap_away has joined #tripleo		09:33
*** coolsvap has quit IRC		09:34
*** jcooley_ has quit IRC		09:37
openstackgerrit	Derek Higgins proposed a change to openstack-infra/tripleo-ci: Get logs for seed vm https://review.openstack.org/69825	09:38
*** coolsvap has joined #tripleo		09:45
*** coolsvap_away has quit IRC		09:48
*** matsuhashi has quit IRC		09:51
*** nosnos has quit IRC		09:51
*** e0ne_ has joined #tripleo		09:53
*** e0ne has quit IRC		09:56
*** cd-undercloud has joined #tripleo		09:58
cd-undercloud	************ overcloud complete status=0 **********	09:58
*** cd-undercloud has quit IRC		09:58
*** akuznetsov has quit IRC		10:01
*** akuznetsov has joined #tripleo		10:01
*** martyntaylor has joined #tripleo		10:01
lifeless	markmc: hi	10:09
markmc	lifeless, hey	10:09
lifeless	markmc: any word on the meetup thing? If not, I'm going to pull a date from my hat w/ mordred this week	10:09
lifeless	markmc: and we'll see what we see ;)	10:09
markmc	lifeless, ah, you missed what I said here yesterday	10:09
*** jcooley_ has joined #tripleo		10:09
markmc	lifeless, sorry, meant to come back and repeat	10:09
lifeless	markmc: apparently :)	10:09
markmc	lifeless, reckon 6ish from Red Hat could make it on the 3rd, short notice makes any more difficult	10:10
markmc	lifeless, but not much point in pushing it closer to the summit either	10:10
lifeless	markmc: ok, thanks	10:10
lifeless	yup	10:10
lifeless	Now, to hunt down mordred	10:10
*** akuznetsov has quit IRC		10:13
*** bauzas has joined #tripleo		10:14
*** jcooley_ has quit IRC		10:15
*** e0ne_ has quit IRC		10:15
*** e0ne has joined #tripleo		10:15
*** e0ne has quit IRC		10:18
*** e0ne has joined #tripleo		10:19
*** e0ne_ has joined #tripleo		10:26
*** e0ne has quit IRC		10:26
*** max_lobur_afk is now known as max_lobur		10:30
*** akuznetsov has joined #tripleo		10:37
openstackgerrit	Derek Higgins proposed a change to openstack/tripleo-incubator: Fix reference to no_proxy https://review.openstack.org/69832	10:39
*** akuznetsov has quit IRC		10:40
derekh	devtest regression ^^	10:40
*** akuznetsov has joined #tripleo		10:42
GheRivero	derekh: is no_proxy=,192.0.2.1 a valid option?	10:44
*** e0ne has joined #tripleo		10:44
*** cd-undercloud has joined #tripleo		10:45
cd-undercloud	************ overcloud complete status=0 **********	10:45
*** cd-undercloud has quit IRC		10:45
*** akuznetsov has quit IRC		10:46
*** akuznetsov has joined #tripleo		10:47
*** e0ne_ has quit IRC		10:48
derekh	GheRivero: yup, its whats being set in devtest_seed https://github.com/openstack/tripleo-incubator/blob/master/scripts/devtest_seed.sh#L57	10:48
*** akuznetsov has quit IRC		10:48
*** athomas has quit IRC		10:50
*** e0ne has quit IRC		10:51
pblaho	lifeless: I found ImportError: cannot import name all in subunit 0.17 ... did you see that?	10:52
*** athomas has joined #tripleo		10:56
*** akuznetsov has joined #tripleo		10:59
*** jistr has quit IRC		10:59
*** mrunge has quit IRC		11:00
*** lucasagomes has joined #tripleo		11:05
openstackgerrit	Derek Higgins proposed a change to openstack-infra/tripleo-ci: Get logs for seed vm https://review.openstack.org/69825	11:09
*** mrunge has joined #tripleo		11:14
*** akrivoka has joined #tripleo		11:15
*** coolsvap has quit IRC		11:15
*** lucasagomes has quit IRC		11:15
*** lucasagomes has joined #tripleo		11:15
*** jistr has joined #tripleo		11:19
*** athomas has quit IRC		11:22
derekh	hmm, the ci testenv seems to have lost its ip address , I can't get in to see whats happened	11:24
*** CaptTofu has joined #tripleo		11:25
*** CaptTofu has quit IRC		11:30
*** akuznetsov has quit IRC		11:30
*** athomas has joined #tripleo		11:31
*** gcha has quit IRC		11:32
*** cd-undercloud has joined #tripleo		11:32
cd-undercloud	************ overcloud complete status=0 **********	11:32
*** cd-undercloud has quit IRC		11:32
*** e0ne has joined #tripleo		11:39
*** e0ne has quit IRC		11:40
*** e0ne has joined #tripleo		11:41
*** e0ne has quit IRC		11:42
*** e0ne has joined #tripleo		11:42
*** e0ne has quit IRC		11:47
*** akuznetsov has joined #tripleo		11:49
*** CaptTofu has joined #tripleo		11:50
sdague	lifeless: on the hopes that you only popped off for dinner, and are coming back	11:52
jamezpolley	okay, that was weird.	11:53
*** akuznets_ has joined #tripleo		11:53
sdague	all of ci is wedged on the latest testtools release	11:53
jamezpolley	I saw sdague speak here, flipped my eyes across to stackaltyics, first word I saw was Sean Dague	11:53
sdague	heh	11:53
openstackgerrit	A change was merged to openstack/tripleo-image-elements: Ensure correct tftp directory ownership for Ironic https://review.openstack.org/69013	11:55
*** akuznetsov has quit IRC		11:55
*** jcooley_ has joined #tripleo		11:58
*** d0ugal has quit IRC		12:04
*** michchap has quit IRC		12:08
*** jcooley_ has quit IRC		12:08
*** cd-undercloud has joined #tripleo		12:19
cd-undercloud	************ overcloud complete status=0 **********	12:19
*** cd-undercloud has quit IRC		12:19
*** michchap has joined #tripleo		12:25
*** michchap has quit IRC		12:25
*** d0ugal has joined #tripleo		12:26
*** max_lobur is now known as max_lobur_afk		12:27
*** CaptTofu has quit IRC		12:28
*** e0ne has joined #tripleo		12:30
*** akuznets_ has quit IRC		12:35
*** e0ne has quit IRC		12:36
*** akuznetsov has joined #tripleo		12:37
*** e0ne has joined #tripleo		12:41
*** akuznetsov has quit IRC		12:44
*** e0ne has quit IRC		12:45
*** d0ugal has quit IRC		12:53
*** akuznetsov has joined #tripleo		12:54
*** jcooley_ has joined #tripleo		12:55
*** akuznetsov has quit IRC		12:57
*** CaptTofu has joined #tripleo		12:59
*** hewbrocca has joined #tripleo		13:02
*** noslzzp has joined #tripleo		13:02
openstackgerrit	Derek Higgins proposed a change to openstack/tripleo-image-elements: Install packages required to build clients https://review.openstack.org/69853	13:02
openstackgerrit	Derek Higgins proposed a change to openstack/tripleo-image-elements: Add dependency on source-repositories https://review.openstack.org/69854	13:02
*** d0ugal has joined #tripleo		13:05
*** cd-undercloud has joined #tripleo		13:06
cd-undercloud	************ overcloud complete status=0 **********	13:06
*** cd-undercloud has quit IRC		13:06
openstackgerrit	A change was merged to openstack/tripleo-image-elements: Add package install support to neutron https://review.openstack.org/67949	13:08
openstackgerrit	A change was merged to openstack/tripleo-image-elements: Add package install support to keystone https://review.openstack.org/67950	13:08
openstackgerrit	A change was merged to openstack/tripleo-image-elements: Install keystone-manage into /usr/local/bin https://review.openstack.org/67951	13:08
openstackgerrit	A change was merged to openstack/tripleo-image-elements: Add package install support to horizon https://review.openstack.org/67952	13:08
openstackgerrit	A change was merged to openstack/tripleo-image-elements: Add package install support to heat https://review.openstack.org/67953	13:08
openstackgerrit	A change was merged to openstack/tripleo-image-elements: Install heat-manage into /usr/local/bin https://review.openstack.org/67954	13:08
openstackgerrit	A change was merged to openstack/tripleo-image-elements: Add package install support to glance https://review.openstack.org/67955	13:08
openstackgerrit	A change was merged to openstack/tripleo-image-elements: Install glance-manage into /usr/local/bin https://review.openstack.org/67956	13:08
*** jdob has joined #tripleo		13:08
*** jcooley_ has quit IRC		13:09
*** akuznetsov has joined #tripleo		13:09
*** d0ugal has quit IRC		13:12
*** jcoufal-mobile has joined #tripleo		13:22
*** lblanchard has joined #tripleo		13:24
*** d0ugal has joined #tripleo		13:25
*** vkozhukalov has quit IRC		13:26
slagle	derekh: somehow i missed installing the openstack-neutron-ml2 package for the neutron package install. will submit a patch for that	13:29
*** jistr is now known as jistr\|english		13:30
*** e0ne has joined #tripleo		13:31
*** akuznetsov has quit IRC		13:35
*** vkozhukalov has joined #tripleo		13:39
*** akuznetsov has joined #tripleo		13:45
openstackgerrit	Dirk Mueller proposed a change to openstack/tripleo-image-elements: Fix editing of systemd files on openSUSE https://review.openstack.org/69863	13:48
*** jcooley_ has joined #tripleo		13:49
marios	guys why do i get [root@ibm-x3550m4-05 tripleo]# ROUTE_DEV=$(OS_CONFIG_FILES=$TE_DATAFILE os-apply-config --key seed-route-dev --type netdevice --key-default virbr0)	13:50
marios	[2014/01/29 08:35:05 AM] [ERROR] Could not parse metadata file: /root/tripleo/testenv.json	13:50
marios	its step 2 of devtest_seed @ http://docs.openstack.org/developer/tripleo-incubator/devtest_seed.html	13:50
openstackgerrit	Ralf Haferkamp proposed a change to openstack/tripleo-image-elements: Account for the differnt locations of systemd units on openSUSE https://review.openstack.org/69864	13:50
*** julim has joined #tripleo		13:52
Ng	morning	13:54
marios	o/	13:54
*** cd-undercloud has joined #tripleo		13:55
cd-undercloud	************ overcloud complete status=1 **********	13:55
*** cd-undercloud has quit IRC		13:55
*** jcooley_ has quit IRC		13:58
slagle	marios: you're doing a manual setup i assume?	14:03
*** dprince has joined #tripleo		14:03
slagle	TE_DATAFILE is defined at the bottom of devtest_variables	14:04
marios	slagle: yah, only ever had it running this way	14:04
slagle	marios: i don't see anything that creates TE_DATAFILE. looks like a bug in the documentation	14:08
slagle	oh, ok, at the bottom of devtest_testenv, there is a echo to $JSONFILE	14:09
slagle	instead of doing that, echo that to $TE_DATAFILE	14:09
slagle	we've been trying to get this fixed for a while. there is some objection that does not make sense to me	14:09
slagle	https://review.openstack.org/#/c/67535/	14:10
marios	slagle: the file is there but empty, thanks having a look now	14:11
marios	(though 'history' tells me i created it with 'touch' manually)	14:12
*** CaptTofu has quit IRC		14:12
slagle	marios: it's the empty file then that's causing the "could not parse" error	14:15
marios	slagle: u the man. thanks a bunch mate got it working now. made usre SEEDIP etc were set and did what you say: echo "{\"host-ip\":\"$HOSTIP\", \"seed-ip\":\"$SEEDIP\", \"node-macs\":\"$MACS\"}" > $TE_DATAFILE	14:17
*** jayg\|g0n3 is now known as jayg		14:19
*** matty_dubs\|gone is now known as matty_dubs		14:20
*** spzala has joined #tripleo		14:21
*** max_lobur_afk is now known as max_lobur		14:23
openstackgerrit	Marios Andreou proposed a change to openstack/tripleo-incubator: Fix typo in seed_ip and make sure testenv config file initialised https://review.openstack.org/69871	14:26
marios	slagle: ah damit sorry thats what the other review does right? abandoning now	14:27
slagle	yea :)	14:28
marios	bnemec: ping	14:28
*** sdake has quit IRC		14:28
marios	bnemec: unping	14:28
*** sdake has joined #tripleo		14:28
*** markmcclain has joined #tripleo		14:31
*** markmcclain1 has joined #tripleo		14:33
*** CaptTofu has joined #tripleo		14:34
*** markmcclain has quit IRC		14:35
*** gcha has joined #tripleo		14:37
openstackgerrit	Ryan Brady proposed a change to openstack/tripleo-image-elements: Fixes authentication for cinder https://review.openstack.org/69619	14:40
*** jistr\|english is now known as jistr		14:41
*** cd-undercloud has joined #tripleo		14:42
cd-undercloud	************ overcloud complete status=0 **********	14:42
*** cd-undercloud has quit IRC		14:42
*** e0ne_ has joined #tripleo		14:44
*** e0ne has quit IRC		14:46
*** vkozhukalov has quit IRC		14:49
*** mrunge has quit IRC		14:56
*** noslzzp has quit IRC		14:56
rlandy	derekh: hi ... https://review.openstack.org/#/c/69832/ .. there are another two occurrences (in devtest_undercloud.sh, devtest_overcloud.sh). Will that code be ok at these later points or should devtest_overcloud.sh and devtest_undercloud.sh also be changed?	14:57
derekh	rlandy: their ok as their both preceeded by set +u	14:59
*** vkozhukalov has joined #tripleo		15:03
*** bauzas has quit IRC		15:03
*** bauzas has joined #tripleo		15:04
*** jprovazn has quit IRC		15:05
*** boris-42 has quit IRC		15:05
*** CaptTofu has quit IRC		15:06
*** weshay has joined #tripleo		15:06
GheRivero	ls	15:09
derekh	bin boot dev etc home initrd.img lib lib64 lost+found media mnt opt proc root run sbin selinux srv sys tmp usr var vmlinuz	15:09
marios	ha	15:09
marios	beat me to it	15:09
derekh	:-)	15:09
*** ftcjeff has quit IRC		15:10
*** boris-42 has joined #tripleo		15:12
marios	if running devtest in a screen session, will my env be lost if i detach/re-attach to the socket?	15:13
openstackgerrit	James Slagle proposed a change to openstack/tripleo-image-elements: Install openstack-neutron-ml2 https://review.openstack.org/69880	15:14
*** markmcclain1 has quit IRC		15:18
*** e0ne_ has quit IRC		15:18
*** e0ne has joined #tripleo		15:18
*** CaptTofu has joined #tripleo		15:21
*** noslzzp has joined #tripleo		15:21
*** michchap has joined #tripleo		15:26
*** cd-undercloud has joined #tripleo		15:27
cd-undercloud	************ overcloud complete status=0 **********	15:27
*** cd-undercloud has quit IRC		15:27
*** ftcjeff has joined #tripleo		15:27
*** mestery has quit IRC		15:29
*** michchap has quit IRC		15:30
openstackgerrit	Derek Higgins proposed a change to openstack/tripleo-incubator: Fix seed domain name https://review.openstack.org/69679	15:37
*** michchap has joined #tripleo		15:40
*** bauzas has quit IRC		15:43
*** lucasagomes is now known as lucas-hungry		15:44
*** bauzas has joined #tripleo		15:45
*** michchap has quit IRC		15:45
*** edmund has joined #tripleo		15:46
jistr	jdob: hi, did you get a chance to look at the examples i posted on the review? do you like one of the options more?	15:48
*** coolsvap has joined #tripleo		15:49
jistr	or do you think we should go a different way?	15:49
jdob	jistr: i see both sides. i figured for a first pass that i'd leave it as is unless the UI goes to use it and its utterly unusable. i suspect for juno a lot of that will be up for discussion, so i didnt want to spend too much time rewriting it now	15:52
jdob	that work for you?	15:52
*** jcooley_ has joined #tripleo		15:53
*** akuznetsov has quit IRC		15:54
jistr	jdob: seems fine from work planning point of view, but i'm a bit worried about having to introduce backwards incompatibility for Juno. I'm not sure how it's in OpenStack rules, if we'd have to release a complete v2 API (to be consistent with how Glance handles properties) or not.	15:55
*** bauzas has quit IRC		15:55
jdob	ah, thats a good point	15:55
*** akrivoka has quit IRC		15:56
jdob	i favor the current approach since it allows for batch changes	15:56
jdob	if we pull things out to sub-collections, you're looking at a REST call per change	15:57
*** bauzas has joined #tripleo		15:57
jistr	well that's option no. 2, which both me and d0ugal considered less appealing afaik	15:57
jistr	with option no. 1 you can do batch updates as well as updating one	15:58
jdob	jistr: OH, I didnt see your second comment	15:58
jdob	ok, lemme digest that and I'll get back to you	15:59
jistr	ok :)	15:59
jistr	(actually option no 2 can do batch updates as well, but you'd have to update all attributes at once, so that's not so great)	16:00
jistr	s/no 2/no. 2/	16:00
jdob	ya, but its less update-semantics and more overwrite	16:01
jdob	which i agree, isnt great	16:01
jistr	yeah	16:01
marios	anyone else seeing heat not running on seed. meaning that step4 of devtest_undercloud fails (no heat-engine "ERROR: Timeout while waiting on RPC response")	16:03
*** markmcclain has joined #tripleo		16:03
*** akrivoka has joined #tripleo		16:03
jistr	marios: yesterday we saw nova timeouts, even on jenkins (see jenkins log on https://review.openstack.org/#/c/69547/)	16:04
jistr	marios: but i'm not sure if that's related	16:04
marios	tx	16:06
*** vkozhukalov has quit IRC		16:07
openstackgerrit	Ladislav Smola proposed a change to openstack/tuskar-ui: Adding nova-baremetal, heat and nova API https://review.openstack.org/68973	16:13
*** cd-undercloud has joined #tripleo		16:14
cd-undercloud	************ overcloud complete status=0 **********	16:14
*** cd-undercloud has quit IRC		16:14
openstackgerrit	A change was merged to openstack/tripleo-incubator: Fix seed domain name https://review.openstack.org/69679	16:16
dprince	Spamaps: are we using PUBLIC_INTERFACE_ROUTE ATM in the HP TripleO setup?	16:20
rlandy	GheRivero: https://bugs.launchpad.net/tripleo/+bug/1215298 ... since I also fell over this, I agree it should be doc'ed. Where do think we should out the info? in current-caveats-workarounds section or in the boot-seed-vm doc itself?	16:24
*** mestery has joined #tripleo		16:24
GheRivero	rlandy: wow! let me reread the bug. I reported it on August!	16:25
*** mestery has joined #tripleo		16:27
rlandy	dprince: hi - are you planning on having the ci meeting today?	16:31
dprince	rlandy: we can if everyone is interested. I'm pretty busy myself trying to fix some stuff ATM so I could punt this week	16:32
rlandy	dprince: ok with me - just want to plan my travel back time accordingly	16:32
dprince	rlandy: I'm skipping out this week I think so lets just say no meeting	16:33
NobodyCam	Morning TripleO	16:33
*** lucas-hungry is now known as lucasagomes		16:34
NobodyCam	I've got a question on moving things like the ssh power keys to /mnt/state	16:34
NobodyCam	currently keys are in /opt/stack/boot-stack	16:35
rlandy	GheRivero: going off line for a few hours - check in with you later	16:36
NobodyCam	with ironic getting closer to actually working	16:36
*** rlandy has quit IRC		16:36
*** jcooley_ has quit IRC		16:37
NobodyCam	on a review it was suggested I use /mnt/state/var/lib/boot-stack/	16:37
*** jcooley_ has joined #tripleo		16:37
NobodyCam	that seems odd to me?	16:37
NobodyCam	should it not be /mnt/state/opt/stack/boot-stack/????	16:38
*** ftcjeff has quit IRC		16:40
*** edmund has quit IRC		16:41
*** jcooley_ has quit IRC		16:42
*** ftcjeff has joined #tripleo		16:46
*** mestery has quit IRC		16:46
*** pblaho has quit IRC		16:48
*** matty_dubs is now known as matty_dubs\|lunch		16:56
*** bauzas has quit IRC		16:57
*** cwolferh has joined #tripleo		16:58
*** lblanchard has quit IRC		16:59
*** e0ne_ has joined #tripleo		17:01
*** e0ne has quit IRC		17:01
*** lblanchard has joined #tripleo		17:01
*** cd-undercloud has joined #tripleo		17:04
cd-undercloud	************ overcloud complete status=1 **********	17:04
*** cd-undercloud has quit IRC		17:04
*** jistr has quit IRC		17:05
*** gcha has quit IRC		17:05
*** viktors has left #tripleo		17:07
*** noslzzp has quit IRC		17:09
SpamapS	We seem to be timing out a lot waiting for nova or neutron	17:15
*** markmc has quit IRC		17:22
*** mestery_ has joined #tripleo		17:23
openstackgerrit	Chris Krelle proposed a change to openstack/tripleo-image-elements: Add nova-ironic element to replace nova-baremetal https://review.openstack.org/66461	17:24
openstackgerrit	Clint "SpamapS" Byrum proposed a change to openstack/tripleo-incubator: Make wait_for print out command on timeout. https://review.openstack.org/69915	17:25
*** michchap has joined #tripleo		17:29
openstackgerrit	Dan Prince proposed a change to openstack/tripleo-image-elements: Write out persistent interface/bridge scripts https://review.openstack.org/69918	17:30
dprince	Spamaps: I need your attention on ^^	17:30
SpamapS	dprince: on it	17:31
dprince	SpamapS: specifically with regard to line #52 (the lets talk about it comment)	17:31
*** michchap has quit IRC		17:33
SpamapS	dprince: ok is there a question how to do this with Debian/Ubuntu's ifupdown ?	17:34
dprince	SpamapS: The eni file (being just one big file) is a bit more problematic in this script. We could regenerate it from scratch... or do a more surgical operation w/ sed.	17:34
dprince	SpamapS: what we'll essentially want to do is outlined here: https://github.com/osrg/openvswitch/blob/master/debian/openvswitch-switch.README.Debian#L63	17:35
*** e0ne_ has quit IRC		17:35
openstackgerrit	Tzu-Mainn Chen proposed a change to openstack/tuskar-ui: Combined Instance and Node objects into Node https://review.openstack.org/69724	17:35
dprince	SpamapS: For me this fixes all issues w/ DHCP release/renewals	17:35
*** noslzzp has joined #tripleo		17:41
SpamapS	dprince: so as of Ubuntu 13.10 you can have /etc/network/interfaces.d	17:41
dprince	SpamapS: that sounds lovely?	17:41
dprince	SpamapS: tell me more!	17:42
SpamapS	dprince: just drop a file in there. It will be appended to the rest.	17:42
* dprince wonders if we can re-write dhcp-all-interfaces to use the same		17:42
SpamapS	dprince: I hope so	17:43
dprince	SpamapS: so, if we take that approach we'd need to make dhcp-all-interfaces do the same for sure because we have to overwrite that file	17:43
SpamapS	dprince: note that you have to ensure that 'source-directory interfaces.d' is in eni	17:43
SpamapS	dprince: it is in the default that gets shipped, but everybody is generating them so at worst.. if ! grep ; then echo "" >> ; fi :)	17:43
dprince	SpamapS: okay. so lets do this. You buy all this being persistent configs then?	17:45
*** matty_dubs\|lunch is now known as matty_dubs		17:45
SpamapS	dprince: persistent.. I'm not sure. Explain more to me. Right now it gets fully regenerated at boot...	17:47
dprince	SpamapS: previously only the OVS stuff would persist automatically on a reboot (in the OVS DB). All of our move the IP to the OVS bridge, then move the nic stuff would not. To me it makes a lot of sense to have it all be persistent...	17:48
dprince	SpamapS: and it solves these sort of problems too	17:48
dprince	SpamapS: like DHCP release/renew not working	17:48
*** cd-undercloud has joined #tripleo		17:49
cd-undercloud	************ overcloud complete status=0 **********	17:49
*** cd-undercloud has quit IRC		17:49
SpamapS	dprince: if we don't make it persistent, is it hard to just make sure nothing happens until dhcp-all-interfaces has run?	17:52
*** jcooley_ has joined #tripleo		17:53
dprince	SpamapS: well, if we don't persist this then we have to undo OVS stuff on reboot as well (like I've already started doing). The interactions between OVS initialization and DHCP are complicated. Distros have already solved it...	17:53
SpamapS	dprince: ok. Is there a way to detect when the physical interfaces have been changed (like, eth0 goes bad, so they plug eth1 in)	17:56
dprince	SpamapS: aha. So no. There isn't right now. We don't unconfigure things in dhcp-all-interfaces (for all distros)	17:56
*** cody-somerville has joined #tripleo		17:56
dprince	SpamapS: so we already don't do that...	17:57
dprince	SpamapS: That sort of problem may always require some sort of manual attention though. I can forsee cases where this would work fine w/ that... and others where you'd just have to jump in and fix it...	17:58
dprince	SpamapS: lifeless asked me this same question BTW and I told him something similar	17:58
*** derekh has quit IRC		18:01
SpamapS	dprince: manual not so much, but automation can be devised when the time comes. :)	18:02
SpamapS	dprince: Ok I am on board with it if for no other reason than "networking is hard, lets make it simpler"	18:03
dprince	SpamapS: sure. I will agree we can solve that more complex problem later.	18:03
SpamapS	dprince: thanks for explaining. :)	18:03
dprince	SpamapS: Okay, I'll push branches to switch Debian over to interfaces.d then here shortly	18:03
SpamapS	(and if you hadn't noticed, I am on a mission to seek and destroy all manual operator intervention. ;)	18:04
SpamapS	Not saying it is possible.. but.. the most glory comes from suicide missions anyway ;)	18:04
*** vkozhukalov has joined #tripleo		18:04
dprince	SpamapS: fair enough. Let's put em' out of a job then	18:05
*** lblanchard has quit IRC		18:07
*** lblanchard has joined #tripleo		18:08
*** mestery_ is now known as mestery		18:10
* Ng dinners		18:14
*** martyntaylor has left #tripleo		18:14
openstackgerrit	Dan Prince proposed a change to openstack/diskimage-builder: Update Debian to use /etc/network/interfaces.d https://review.openstack.org/69935	18:16
dprince	SpamapS: Try that ^^	18:16
openstackgerrit	Ana Krivokapic proposed a change to openstack/tuskar-ui: Add node detail view https://review.openstack.org/69462	18:17
openstackgerrit	Dan Prince proposed a change to openstack/tripleo-image-elements: Write out persistent interface/bridge scripts https://review.openstack.org/69918	18:23
NobodyCam	SpamapS: gots a second to take a quick peek at 66461 specifically: https://review.openstack.org/#/c/66461/3/elements/nova-ironic/os-refresh-config/configure.d/80-ironic-ssh-power-key	18:24
greghaynes	Do we usually put usernames with TODO's?	18:24
*** d0ugal has quit IRC		18:25
greghaynes	re, that file	18:25
NobodyCam	greghaynes: ack :) will do	18:26
greghaynes	oh, I legitimitely was asking, very new to this codebase :)	18:26
NobodyCam	greghaynes: yes we do!	18:27
greghaynes	ah, sweet	18:27
SpamapS	have a doctor's appt .. be back in a bit to review all the things	18:29
openstackgerrit	Chris Krelle proposed a change to openstack/tripleo-image-elements: Add nova-ironic element to replace nova-baremetal https://review.openstack.org/66461	18:29
NobodyCam	greghaynes: ^^	18:29
NobodyCam	:)	18:30
greghaynes	woot.	18:30
openstackgerrit	Dan Prince proposed a change to openstack/tripleo-image-elements: Write out persistent interface/bridge scripts https://review.openstack.org/69918	18:30
*** cd-undercloud has joined #tripleo		18:37
cd-undercloud	************ overcloud complete status=0 **********	18:37
*** cd-undercloud has quit IRC		18:37
*** lblanchard has quit IRC		18:37
*** lblanchard has joined #tripleo		18:38
*** jprovazn has joined #tripleo		18:41
openstackgerrit	Chris Krelle proposed a change to openstack/tripleo-image-elements: Add nova-ironic element to replace nova-baremetal https://review.openstack.org/66461	18:45
NobodyCam	greghaynes: line gone now! lol :-p	18:45
greghaynes	haha, always for removing todos :)	18:46
openstackgerrit	Dan Prince proposed a change to openstack/tripleo-image-elements: Stop using all ports option w/ neutron-ovs-cleanup https://review.openstack.org/69937	18:47
NobodyCam	your comment got me thinking... we cann't bail out as we don't know what power control ironic is using from nova	18:48
greghaynes	Still fine to generate the key though?	18:51
greghaynes	Just some wasted entropy :p	18:51
lifeless	morning	19:04
NobodyCam	greghaynes: unlike nova-bm where we could detect that it was using the ssh power driver, Ironic sets that per node so we may or may not need the key	19:04
NobodyCam	morning lifeless :)	19:04
greghaynes	good morning	19:05
lifeless	dprince: I don't think its true that distros have solved the ovs interactions really; they've just punted on 'statically configure this here<...>'	19:05
dprince	lifeless: well. the distro solutions I'm testing here seem to get us in the ball park, so I'm fine going with them	19:07
lifeless	dprince: thats fine :)	19:07
lifeless	dprince: just being a little ornery :>	19:07
lifeless	I am, that is	19:07
dprince	lifeless: to be fair I've only tested this stuff on Fedora. I've implemented the Debian solution too though	19:08
lifeless	cool	19:08
lifeless	note that interfaces.d isn't available on precise	19:08
lifeless	AFAIK, that is	19:08
dprince	lifeless: right, well it made my implementation easier (palatable even!) so I went for it immediately	19:09
dprince	lifeless: not having to slice and dice the stander interfaces file is highly desirable	19:09
dprince	standard	19:10
dprince	lifeless: but if we need to support older versions so be it	19:10
*** e0ne has joined #tripleo		19:11
lifeless	dprince: I think we do, because precise is the RHEL-equiv	19:12
*** pblaho has joined #tripleo		19:13
dprince	lifeless: Okay. What is your take on our best option then? We could just regenerate /etc/network/interfaces in ensure-bridge... or the other option is use a multi-line sed delete or something to cut out the parts we don't like	19:13
*** cody-somerville has quit IRC		19:15
lifeless	dprince: what are the operations we need to do ?	19:15
lifeless	add an interface, <...> ?	19:15
dprince	lifeless: we need to overwrite the config for the original DHCP interface which may already be there if dhcp-all-interfaces was used	19:16
lifeless	dprince: so ifdown; replace the defintion; ifup ?	19:16
dprince	lifeless: that is what the current script does yes	19:17
lifeless	kk	19:17
lifeless	uhm, awk is probably the tool for this	19:17
lifeless	but if you can wedge it into sed, sure.	19:17
*** michchap has joined #tripleo		19:17
dprince	lifeless: actually I re-write the script first, and then ifdown/ifup	19:17
lifeless	dprince: ah, so that will leave dhclient running	19:17
rbrady	having trouble with an element during image build (heat-cfntools). It hasn't changed in a long time and the source file causing the issue hasn't changed in a while either. Can someone let me know what I'm missing or point me in the right direction? http://paste.openstack.org/show/62129/	19:17
lifeless	dprince: and thus it will eventually drop off the network	19:17
lifeless	dprince: you have to ifdown with the config it had, then ifup with the new config	19:18
dprince	lifeless: hmmm. maybe. I'll tinker w/ it. In any case... I'm fine switching the order up	19:18
lifeless	which does introduce a lovely big race condition if the editing-of-config fails, but there you are	19:18
dprince	lifeless: right, its unlikely to fail though	19:19
*** michchap has quit IRC		19:22
*** cd-undercloud has joined #tripleo		19:22
cd-undercloud	************ overcloud complete status=0 **********	19:22
*** cd-undercloud has quit IRC		19:22
dprince	lifeless: do we need the PUBLIC_INTERFACE_ROUTE stuff now even? Wouldn't this always get pushed via DHCP in our cases?	19:23
*** athomas has quit IRC		19:24
lifeless	dprince: no	19:25
lifeless	dprince: thats used to bring up the vlans with external IPs on them	19:25
*** CaptTofu_ has joined #tripleo		19:25
dprince	lifeless: ah, okay. Well I left it alone anways	19:25
lifeless	dprince: since we don't [yet] have a working config for 'here, serve DHCP on a vlan but no this isn't a different interface so it has the same MAC address'	19:25
*** CaptTofu has quit IRC		19:27
*** pblaho has quit IRC		19:29
*** akuznetsov has joined #tripleo		19:30
*** peoplemerge has joined #tripleo		19:33
lifeless	SpamapS: Ng: reviews please - https://review.openstack.org/#/c/67343/	19:33
* peoplemerge is giving himself a crash course today on tripleo		19:36
lifeless	peoplemerge: cool	19:37
*** martyntaylor has joined #tripleo		19:38
*** martyntaylor has left #tripleo		19:38
peoplemerge	Thanks lifeless, it's a brave new world	19:39
*** CaptTofu_ has quit IRC		19:46
*** CaptTofu has joined #tripleo		19:51
*** bauzas has joined #tripleo		19:52
*** max_lobur is now known as max_lobur_afk		19:53
*** akuznetsov has quit IRC		19:55
*** vkozhukalov has quit IRC		19:56
Ng	lifeless: k	20:01
Ng	lifeless: unrelated question: Where would be a good place to stash this freecloud/jira guide? It's not really suitable for public dissemination	20:02
lifeless	Ng: uhm, wiki internally I guess	20:03
lifeless	Ng: or perhaps the internal etherpad?	20:04
SpamapS	lifeless: peoplemerge is a local friend here in LA who has tried unsuccessfully to bring hardware to my house to turn into a cloud... FYI :)	20:08
*** cd-undercloud has joined #tripleo		20:09
cd-undercloud	************ overcloud complete status=0 **********	20:09
*** cd-undercloud has quit IRC		20:09
lifeless	cool	20:12
peoplemerge	Hi lifeless. Yeah a sysadmin just did a POC on grizzly and everybody has had great feedback. Looking at commercial support options now. I gotta ramp up, and thought I'd get involved as a contributor.	20:14
*** jdob has quit IRC		20:15
*** ftcjeff has quit IRC		20:18
NobodyCam	peoplemerge: Awesome, :)	20:19
matty_dubs	LOL I thought I was the only one whose house bordered on 'cloud' status ;)	20:22
NobodyCam	does my RV count as a mobile cloud?	20:23
*** hewbrocca has quit IRC		20:23
matty_dubs	'mobile cloud' FTW!	20:23
NobodyCam	:-p	20:24
NobodyCam	I have to take my cloud to where the sun is	20:24
peoplemerge	Ya I gotta trunkful of laptops, servers, and desktops that meet the minimum specs. I wonder how many I could plug into that 500W car AC adapter before the cloud becomes a lightening storm	20:26
NobodyCam	peoplemerge: http://www.mini-box.com/Car-PC-Automotive-Computing-Solutions	20:29
*** dprince has quit IRC		20:31
* bnemec wonders how many servers he could run on the 30 amp service in his trailer.		20:32
lifeless	bauzas: one, if its big enough :)	20:34
lifeless	bah	20:34
lifeless	bnemec: ^	20:34
* NobodyCam notes that the link he pasted is for 12v computers, that do not draw off the rv/trailer's a/c... that said /me added a second 30a to his RV		20:34
*** e0ne has quit IRC		20:35
*** rdopieralski has quit IRC		20:36
*** lucasagomes has quit IRC		20:37
bnemec	Heh, true.	20:41
openstackgerrit	Ryan Brady proposed a change to openstack/tripleo-image-elements: Fixes heat-cfntools install https://review.openstack.org/69968	20:44
*** coolsvap is now known as coolsvap_away		20:46
*** gcha has joined #tripleo		20:53
*** cd-undercloud has joined #tripleo		20:55
cd-undercloud	************ overcloud complete status=0 **********	20:55
*** cd-undercloud has quit IRC		20:55
pleia2	lifeless: so, still not having luck with your ssh patch, I'm confident my ssh key works now (if I kill nodepoold right before it goes to delete the image, I can log in fine with my key as fedora@vm-ip), but paramiko ssh is still failing	21:05
*** michchap has joined #tripleo		21:06
*** michchap has quit IRC		21:11
*** jcooley_ has quit IRC		21:11
*** jcooley_ has joined #tripleo		21:12
*** jprovazn has quit IRC		21:38
*** pblaho has joined #tripleo		21:46
*** CaptTofu has quit IRC		21:46
*** e0ne has joined #tripleo		21:46
*** CaptTofu has joined #tripleo		21:47
* Ng ejects		21:47
*** e0ne has quit IRC		21:50
*** epim has joined #tripleo		21:58
*** cd-undercloud has joined #tripleo		22:04
cd-undercloud	************ overcloud complete status=1 **********	22:04
*** cd-undercloud has quit IRC		22:04
*** pblaho has quit IRC		22:06
*** gcha has quit IRC		22:10
*** cody-somerville has joined #tripleo		22:13
*** lblanchard has quit IRC		22:22
*** ftcjeff has joined #tripleo		22:22
*** olaph has joined #tripleo		22:22
*** cody-somerville has quit IRC		22:25
*** CaptTofu has quit IRC		22:32
*** julim has quit IRC		22:33
*** matty_dubs is now known as matty_dubs\|gone		22:38
*** jayg is now known as jayg\|g0n3		22:38
openstackgerrit	Tzu-Mainn Chen proposed a change to openstack/tuskar-ui: Combined Instance and Node objects into Node https://review.openstack.org/69724	22:58
*** noslzzp has quit IRC		23:00
ccrouch	lifeless: do you have any ideas about a date for the mid-cycle meetup?	23:04
SpamapS	ccrouch: I was thinking the same thing.. we need to choose like.. soon	23:05
*** epim has quit IRC		23:05
ccrouch	i think most folks will need at least a couple of weeks notice	23:06
lifeless	ccrouch: 3rd march	23:06
lifeless	ccrouch: just doing final approvals now	23:06
*** cd-undercloud has joined #tripleo		23:10
cd-undercloud	************ overcloud complete status=1 **********	23:10
*** cd-undercloud has quit IRC		23:10
ccrouch	lifeless: fantastic, let me see who we can line up to go	23:14
lifeless	ccrouch: cool; markmc reckoned you could send ~6 folk	23:31
*** weshay has quit IRC		23:31
*** epim has joined #tripleo		23:32
*** michchap has joined #tripleo		23:36
*** peoplemerge has quit IRC		23:40
*** bauzas has quit IRC		23:42
*** epim has quit IRC		23:48
*** cadenzajon has quit IRC		23:57
devananda	lifeless: mirantis will try to send a few folks as well for ironic	23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!