| *** boris-42 has quit IRC | 00:32 | |
| *** sdake has quit IRC | 00:34 | |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Make it possible to customise overcloud names. https://review.openstack.org/66467 | 00:38 |
|---|---|---|
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Use the JSON description to probe the seed VM. https://review.openstack.org/66248 | 00:38 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Use the test env JSON to detect remote VM hosts. https://review.openstack.org/66249 | 00:38 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Start to split devtest_setup out more. https://review.openstack.org/66165 | 00:38 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Populate devtest_variables from devtest_setup. https://review.openstack.org/66180 | 00:38 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Cleanup use of SEED_IP. https://review.openstack.org/66256 | 00:38 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Allow passed-in TE_DATAFILE environments. https://review.openstack.org/66243 | 00:38 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Create and use a JSON testenvironment file. https://review.openstack.org/66183 | 00:38 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: Fix typo. https://review.openstack.org/66245 | 00:38 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-incubator: No longer export DEPLOY_IMAGE_ELEMENT https://review.openstack.org/66157 | 00:38 |
| *** newell has quit IRC | 00:42 | |
| *** e0ne has quit IRC | 00:46 | |
| *** e0ne has joined #tripleo | 00:48 | |
| *** CaptTofu has quit IRC | 00:51 | |
| StevenK | lifeless: That was rebasing the remaining stack? | 01:01 |
| lifeless | StevenK: fixing the install-dependencies conflict | 01:02 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Update apache2 to use the os-apply-config dir. https://review.openstack.org/66470 | 01:07 |
| openstackgerrit | lifeless proposed a change to openstack/tripleo-image-elements: Update cinder to use the os-apply-config dir. https://review.openstack.org/66471 | 01:07 |
| lifeless | I wonder if I've found another bad machine | 01:10 |
| *** nosnos has joined #tripleo | 01:13 | |
| *** e0ne has joined #tripleo | 01:15 | |
| *** e0ne_ has joined #tripleo | 01:16 | |
| *** vkozhukalov has joined #tripleo | 01:17 | |
| *** e0ne_ has quit IRC | 01:18 | |
| *** e0ne has quit IRC | 01:20 | |
| lifeless | ah yes, it's marked bust by clint | 01:23 |
| lifeless | bnut it's still registered | 01:24 |
| *** xuhanp has joined #tripleo | 01:44 | |
| SpamapS | lifeless: perhaps it got re-reg'd somehow? | 01:45 |
| SpamapS | lifeless: the one I took out seemed to not be on the network | 01:46 |
| lifeless | SpamapS: dunno. I have changed the service host to 'disabled' which I hope will be self documenting | 01:46 |
| lifeless | SpamapS: better than unregistering | 01:46 |
| SpamapS | there's a disabled? | 01:46 |
| lifeless | update bm_nodes set service_host='disabled' where uuid='....'; | 01:46 |
| SpamapS | I recall looking for that and failing to find it | 01:46 |
| SpamapS | ah so not in the API | 01:46 |
| lifeless | if service_host doesn't match the nova-compute service hostname, then the node won't be advertised to the scheduler | 01:47 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Start to split devtest_setup out more. https://review.openstack.org/66165 | 01:49 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Populate devtest_variables from devtest_setup. https://review.openstack.org/66180 | 01:50 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: No longer export DEPLOY_IMAGE_ELEMENT https://review.openstack.org/66157 | 01:51 |
| StevenK | \o/ | 01:51 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Create and use a JSON testenvironment file. https://review.openstack.org/66183 | 01:51 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Allow passed-in TE_DATAFILE environments. https://review.openstack.org/66243 | 01:53 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Fix typo. https://review.openstack.org/66245 | 01:53 |
| lifeless | ctlplane=10.10.16.177, 10.10.16.177 | 01:54 |
| lifeless | odd | 01:54 |
| StevenK | lifeless: Is that the full stack that we did on Sunday? | 01:56 |
| lifeless | StevenK: I think so | 01:56 |
| StevenK | \o/ | 01:56 |
| SpamapS | https://review.openstack.org/#/c/66248/ <-- not sure why we need jq there | 02:01 |
| lifeless | SpamapS: to update the JSON | 02:02 |
| SpamapS | lifeless: oh jq is write? hm | 02:04 |
| lifeless | its both | 02:05 |
| lifeless | but the thing we needed was write | 02:05 |
| lifeless | that sets the attribute to the given value | 02:05 |
| lifeless | and outputs the entire JSON again | 02:05 |
| lifeless | which I then write to a file | 02:05 |
| SpamapS | that is not at all clear from the code.. but I didn't bother to read the manpage of jq either, so thats my bad. :) | 02:06 |
| SpamapS | well the "then write to file" is clear :) | 02:06 |
| lifeless | the -m stuff I'd like to leave, it's kinda meh | 02:06 |
| SpamapS | I was just noting it. :) | 02:08 |
| SpamapS | +2'd | 02:08 |
| lifeless | | 38ad3976-1c81-45b1-a840-97e52a5ff88a | ci-overcloud | CREATE_IN_PROGRESS | 2014-01-14T01:53:38Z | | 02:11 |
| lifeless | hurray up | 02:11 |
| *** sdake has joined #tripleo | 02:13 | |
| *** sdake has joined #tripleo | 02:13 | |
| lifeless | and tada | 02:21 |
| lifeless | a new cloud | 02:21 |
| SpamapS | h | 02:29 |
| SpamapS | hm | 02:29 |
| SpamapS | this haproxy element.. we really should be writing a Neutron LBaaS driver that configures it. | 02:29 |
| lifeless | maybe | 02:32 |
| lifeless | I think thats worth exploring, certainly | 02:32 |
| SpamapS | I think it would be a ton simpler | 02:33 |
| SpamapS | We have API's for these things for a reason :) | 02:33 |
| lifeless | One word: bootstrapcycles | 02:34 |
| SpamapS | We have neutron-dhcp-agent on the undercloud right? If we have that, we can have neutron-lbaas-agent too | 02:36 |
| SpamapS | lifeless: not sure what that word is? | 02:36 |
| lifeless | SpamapS: does neutron-lbaas depend on any APIs | 02:36 |
| SpamapS | neutron-server rpc certainly | 02:37 |
| SpamapS | and they've talked about having a driver that spins up VMs for haproxy | 02:37 |
| lifeless | remember that heat can't orchestrate the API for us yet | 02:37 |
| SpamapS | Hm, I think it can... this is 100% undercloud | 02:38 |
| lifeless | SpamapS: no, it's seed | 02:38 |
| SpamapS | seed? no | 02:38 |
| SpamapS | For the undercloud bootstrap, yes | 02:38 |
| lifeless | SpamapS: you want the haproxy in the undercloud to be configured by undercloud metadata, which the seed provides. | 02:38 |
| SpamapS | but that is the "scale out the undercloud" story that we have a ton of weirdness to overcome on. | 02:38 |
| lifeless | SpamapS: right, but we can either a) not have haproxy until we do that, or b) not have LBaaS until we do that. | 02:39 |
| SpamapS | the haproxy in the undercloud that I'm talking about will be proxying to overcloud machines behind it. | 02:39 |
| SpamapS | lifeless: I want HA in overcloud more than I want HA in undercloud. | 02:39 |
| SpamapS | and would rather not do it two different ways | 02:40 |
| lifeless | SpamapS: I get that; lets deconstruct it a little | 02:41 |
| lifeless | SpamapS: for HAproxy to do it's job, we need N HAProxies | 02:41 |
| lifeless | with VIP themselves | 02:41 |
| lifeless | so the LBaaS in neutron can lean on floating ip implementations | 02:42 |
| lifeless | we don't have one | 02:42 |
| lifeless | SpamapS: ahahahahaha 'unexpected poweroff event' on this server. | 02:50 |
| * lifeless STABS the hardware | 02:50 | |
| SpamapS | lifeless: right, but that seems like work we can do in Neutron the same way we made ports work. | 02:50 |
| lifeless | SpamapS: it took 7 months to get ports to work. | 02:50 |
| SpamapS | ow | 02:51 |
| lifeless | SpamapS: I'm super super worried about purity over pragmatism here. | 02:51 |
| SpamapS | This is pragmatism | 02:51 |
| SpamapS | It will almost "just work" for the overcloud. | 02:52 |
| SpamapS | anyway | 02:52 |
| SpamapS | I get that there are hurdles | 02:52 |
| lifeless | I don't think it will; I think floating ips are going to be a much bigger consideration than you do. | 02:52 |
| SpamapS | and I'm fine with bypassing them for a short term "get 'er done" .. :-P | 02:52 |
| *** coolsvap has quit IRC | 02:53 | |
| SpamapS | anyway.. kids.. family time.. bbl | 02:53 |
| lifeless | ciao | 02:54 |
| lifeless | 1611-Chassis Fan7 Failure | 02:54 |
| lifeless | 1611-Chassis Fan15 Failure | 02:54 |
| lifeless | Fan Solution Not Sufficient | 02:54 |
| lifeless | Critical Failure Detected - System Shutting Down in 1 Seconds! | 02:54 |
| lifeless | ahahahahahahahahahahahahahahahahahahahahahahahahahahahaha | 02:54 |
| StevenK | Hah | 02:56 |
| *** vipul is now known as vipul-away | 02:59 | |
| *** AaronGr is now known as AaronGr_Zzz | 03:02 | |
| *** vipul-away is now known as vipul | 03:19 | |
| lifeless | SpamapS: how do you get heat to use an existing floating IP ? | 03:22 |
| *** ccrouch has joined #tripleo | 03:35 | |
| *** weshay has quit IRC | 04:02 | |
| *** ramishra has joined #tripleo | 04:03 | |
| *** ramishra_ has joined #tripleo | 04:05 | |
| SpamapS | lifeless: you pass the ID in here http://docs.openstack.org/developer/heat/template_guide/openstack.html#OS::Neutron::FloatingIPAssociation | 04:08 |
| *** ramishra has quit IRC | 04:08 | |
| StevenK | lifeless, devananda: WRT what we discussed with mirror sizes at LCA: http://paste.openstack.org/show/61159/ | 04:09 |
| *** edmund has quit IRC | 04:32 | |
| *** coolsvap has joined #tripleo | 04:39 | |
| *** akuznetsov has joined #tripleo | 05:10 | |
| *** tserong has quit IRC | 05:17 | |
| *** tserong has joined #tripleo | 05:20 | |
| *** tserong has joined #tripleo | 05:20 | |
| *** ramishra_ has quit IRC | 05:36 | |
| *** hewbrocca has quit IRC | 05:49 | |
| *** hewbrocca has joined #tripleo | 05:49 | |
| *** ramishra has joined #tripleo | 05:57 | |
| *** coolsvap has quit IRC | 06:07 | |
| *** boris-42 has joined #tripleo | 06:18 | |
| *** rpodolyaka1 has joined #tripleo | 06:29 | |
| *** coolsvap has joined #tripleo | 06:32 | |
| *** rlandy has joined #tripleo | 06:46 | |
| *** akuznets_ has joined #tripleo | 06:49 | |
| *** akuznetsov has quit IRC | 06:49 | |
| *** mrunge has joined #tripleo | 06:51 | |
| *** vkozhukalov has quit IRC | 06:53 | |
| *** e0ne has joined #tripleo | 06:57 | |
| *** marun has quit IRC | 07:01 | |
| *** boris-42 has quit IRC | 07:04 | |
| *** akuznets_ has quit IRC | 07:32 | |
| *** akuznetsov has joined #tripleo | 07:33 | |
| *** ifarkas has joined #tripleo | 07:37 | |
| *** rpodolyaka1 has quit IRC | 07:41 | |
| *** jtomasek has joined #tripleo | 07:47 | |
| *** jcoufal has quit IRC | 07:48 | |
| *** ramishra has quit IRC | 07:54 | |
| *** jistr has joined #tripleo | 08:01 | |
| lifeless | Ng: so - new cloud is setup, creds for it are in ~/tripleo-ci-* and ~/os-asserted-users on the undercloud | 08:07 |
| lifeless | Ng: it has SSL | 08:07 |
| *** ifarkas has quit IRC | 08:07 | |
| lifeless | Ng: and the tripleo-bm-test network setup, in principle | 08:07 |
| *** xuhanp has quit IRC | 08:07 | |
| lifeless | Ng: infra should be able to spin up and build nodes on it from the nodepool tenant | 08:08 |
| * lifeless done for the day | 08:08 | |
| *** jcoufal has joined #tripleo | 08:08 | |
| lifeless | Ng: https://trello.com/c/Z0WFQaFg/64-dedicated-non-updated-overcloud-until-we-have-nodowntime-deploys | 08:08 |
| *** bauzas has joined #tripleo | 08:12 | |
| *** e0ne has quit IRC | 08:17 | |
| *** xuhanp has joined #tripleo | 08:19 | |
| rpodolyaka | morning tripleo | 08:20 |
| *** vkozhukalov has joined #tripleo | 08:27 | |
| *** rdopieralski has joined #tripleo | 08:35 | |
| *** kpavel has joined #tripleo | 08:42 | |
| *** kpavel_ has joined #tripleo | 08:43 | |
| *** jprovazn has joined #tripleo | 08:46 | |
| *** pblaho has joined #tripleo | 08:46 | |
| *** kpavel has quit IRC | 08:47 | |
| *** kpavel_ is now known as kpavel | 08:47 | |
| *** rushiagr has joined #tripleo | 08:53 | |
| *** xuhanp has quit IRC | 08:56 | |
| *** jrist has quit IRC | 09:00 | |
| Ng | morning | 09:03 |
| Ng | lifeless: nice! | 09:03 |
| openstackgerrit | Victor Sergeyev proposed a change to openstack/python-tuskarclient: Use HTTPClient from common apiclient code https://review.openstack.org/63749 | 09:04 |
| *** coolsvap has quit IRC | 09:05 | |
| *** e0ne has joined #tripleo | 09:07 | |
| *** e0ne_ has joined #tripleo | 09:08 | |
| *** e0ne has quit IRC | 09:12 | |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Add Nodes Overview page https://review.openstack.org/66522 | 09:14 |
| *** ifarkas has joined #tripleo | 09:18 | |
| *** ifarkas has quit IRC | 09:19 | |
| *** ifarkas has joined #tripleo | 09:19 | |
| *** xuhanp has joined #tripleo | 09:20 | |
| *** ifarkas has quit IRC | 09:31 | |
| *** athomas has joined #tripleo | 09:32 | |
| *** derekh has joined #tripleo | 09:36 | |
| *** tserong has quit IRC | 09:36 | |
| *** tserong has joined #tripleo | 09:38 | |
| *** coolsvap has joined #tripleo | 09:38 | |
| *** lsmola_ has quit IRC | 09:39 | |
| *** ifarkas has joined #tripleo | 09:42 | |
| *** markmc has joined #tripleo | 09:48 | |
| SpamapS | jprovazn: btw ignore my -1 on haproxy... lifeless set me straight.. reviewing it for real now | 09:49 |
| jprovazn | SpamapS: cool, thanks! | 09:50 |
| SpamapS | jprovazn: though I think I'm going to -1 the approach.. | 09:51 |
| *** lsmola_ has joined #tripleo | 09:51 | |
| SpamapS | jprovazn: I think it would be a bazillion times simpler to just have default ports (proxy and frontend) in the heat templates than in the images. | 09:51 |
| SpamapS | jprovazn: that way we can just configure things the way we usually do.. with Heat metadata. | 09:52 |
| jprovazn | SpamapS: well, I defined both port and proxy port in heat template in the previous patchset (check prev version), the problem is that knowledge of proxy port is needed on other places than haproxy config file too: when configuring iptables, it will also be needed when setting up custom endpoints | 09:55 |
| jprovazn | SpamapS: and I don't think that it's needed to have proxy port cinfgurable, choosing just some suitable set of ports should work fine | 09:56 |
| SpamapS | jprovazn: its not so much the need, but the fact that we're now using new config tools, and I'm not sure why. | 09:59 |
| SpamapS | jprovazn: all of those other places will have access to the ports.. | 09:59 |
| SpamapS | ? | 09:59 |
| jprovazn | SpamapS: what new config tools do you mean? save-proxy-setting script? | 10:00 |
| SpamapS | jprovazn: os-apply-config --key haproxy.services.keystone.port no? | 10:00 |
| SpamapS | jprovazn: And the python script to generate the template | 10:00 |
| jprovazn | SpamapS: usage of above commands comes from need to combine knowledge of metadata info and installed elements | 10:02 |
| jprovazn | SpamapS: so elements can say "hey, I'm able to by proxied" and if heat template says "proxy this service" then it's really proxied | 10:03 |
| SpamapS | jprovazn: and what about remote elements? | 10:03 |
| jprovazn | SpamapS: what do you mean by remote element? | 10:03 |
| SpamapS | what about load balancing from the controller to a dedicated set of nova-api boxes? | 10:04 |
| jprovazn | SpamapS: still not sure if I understand, but you define set of nodes by haproxy.nodes in heat metadata | 10:05 |
| jprovazn | I mean set of target real nodes | 10:06 |
| rlandy | derekh: thanks for your help - devtest completed :) | 10:07 |
| derekh | rlandy: cool, no problem | 10:07 |
| SpamapS | jprovazn: ok but nova-api wouldn't be on the controller at all at that point | 10:07 |
| jprovazn | SpamapS: you can define any set of nodes (doesn't have to be local) - https://review.openstack.org/#/c/61374/4/elements/haproxy/README.md | 10:08 |
| SpamapS | jprovazn: what I'm getting at is that "installed elements" has nothing to do with "services to provide" | 10:08 |
| jprovazn | ah, I see | 10:09 |
| SpamapS | jprovazn: if I just built an haproxy image how would I tell it to listen on all the keystone/etc. etc. ports? | 10:09 |
| jprovazn | mmt, let me check current patchset | 10:10 |
| *** phschwartz has quit IRC | 10:10 | |
| jprovazn | SpamapS: so https://review.openstack.org/#/c/61374/4/elements/haproxy/os-config-applier/etc/haproxy/haproxy.cfg would require an update to match this requirement | 10:11 |
| *** phschwartz has joined #tripleo | 10:12 | |
| jprovazn | SpamapS: is it real use case? | 10:12 |
| jprovazn | SpamapS: and https://review.openstack.org/#/c/61374/4/elements/haproxy/os-refresh-config/pre-configure.d/98-haproxy-iptables has to be updated too | 10:13 |
| *** jistr has quit IRC | 10:14 | |
| *** xuhanp has quit IRC | 10:16 | |
| SpamapS | jprovazn: being able to load balance to nodes that don't have the actual software on them is a real use case yes. It is not our use case today, but it is a pretty common setup. | 10:16 |
| SpamapS | jprovazn: It seems rather narrow minded to rule it out entirely. What was the problem with just passing the ports in via heat templates? (and I'm sorry if I was the one pushing back.. my perspective changes.. often..) | 10:17 |
| jprovazn | SpamapS: getting ports directly of heat template involves more complex iptables rule updating which I didn't like (though I have to admit that I didn't count with "having proxy nodes only w/o services" use case) | 10:20 |
| jprovazn | SpamapS: but there is no technical blocker I guess - os-apply-config --key can be used to get it from heat template | 10:20 |
| jprovazn | SpamapS: and if there will be more configuration of related service requried - all will have to be done in os-collect-config phase | 10:24 |
| jprovazn | SpamapS: e.g. current iptables rules (in boot-stack IIRC) should be then fixed - there is not reason to open default ports for proxy-only nodes | 10:25 |
| SpamapS | jprovazn: it doesn't have to be this 100% flexible, I just want to make sure it will be able to handle real changes in topology. | 10:26 |
| *** coolsvap has quit IRC | 10:27 | |
| jprovazn | SpamapS: another alternative to fetch ports from heat templat ewould be to create ports mapping file on image | 10:27 |
| SpamapS | jprovazn: Lets let others weigh in. I am very tired, and I may be asking too much. :p | 10:27 |
| jprovazn | SpamapS: one file installed as part of some basic element | 10:27 |
| jprovazn | SpamapS: sure | 10:27 |
| jprovazn | SpamapS: it's pretty late in your TZ | 10:27 |
| jprovazn | good night | 10:28 |
| SpamapS | jprovazn: I dislike having the ports for a remote application come from the local image. I think it may be too rigid, that is all. | 10:28 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Install traceroute on Fedora. https://review.openstack.org/66411 | 10:30 |
| jprovazn | SpamapS: well, I meant only proxy ports, not target ports of real application | 10:30 |
| *** akuznetsov has quit IRC | 10:31 | |
| *** akuznetsov has joined #tripleo | 10:32 | |
| *** lsmola_ has quit IRC | 10:40 | |
| *** max_lobur_afk is now known as max_lobur | 10:44 | |
| *** rpodolyaka has quit IRC | 10:52 | |
| *** lsmola_ has joined #tripleo | 10:53 | |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Add a wait condition to undercloud stack https://review.openstack.org/66323 | 10:54 |
| *** jistr has joined #tripleo | 10:57 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Install os-collect-config from release into a venv https://review.openstack.org/66351 | 10:58 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Install os-refresh-config from release into a venv https://review.openstack.org/66352 | 10:59 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Don't install python-pip https://review.openstack.org/66353 | 10:59 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Don't install git-core https://review.openstack.org/66354 | 10:59 |
| *** lucasagomes has joined #tripleo | 11:00 | |
| *** rpodolyaka has joined #tripleo | 11:07 | |
| *** boris-42 has joined #tripleo | 11:07 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Removing the config-applier element https://review.openstack.org/66337 | 11:09 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Retry link check up to 10 times. https://review.openstack.org/66397 | 11:12 |
| *** martyntaylor has joined #tripleo | 11:17 | |
| *** jcoufal has quit IRC | 11:47 | |
| *** jcoufal has joined #tripleo | 11:54 | |
| *** jtomasek has quit IRC | 12:10 | |
| *** morazi has joined #tripleo | 12:13 | |
| *** jtomasek has joined #tripleo | 12:13 | |
| *** coolsvap has joined #tripleo | 12:24 | |
| derekh | is it just me or is network access to the cd-undercloud gone flaky ? | 12:30 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Add package support to source-repositories https://review.openstack.org/65754 | 12:42 |
| *** coolsvap has quit IRC | 12:46 | |
| *** mrunge has quit IRC | 12:48 | |
| *** rushiagr has quit IRC | 12:56 | |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Add Nodes Overview page https://review.openstack.org/66522 | 13:02 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Add new modprobe-blacklist element. https://review.openstack.org/66425 | 13:05 |
| openstackgerrit | Petr Blaho proposed a change to openstack/python-tuskarclient: [WIP] Adds help for subcommands https://review.openstack.org/56257 | 13:07 |
| lsmola_ | SpamapS: hello | 13:08 |
| *** jtomasek has quit IRC | 13:19 | |
| *** jtomasek has joined #tripleo | 13:20 | |
| lsmola_ | SpamapS: can I ask when we are planning to get rid of merge.py and make it via HOT nested resources ? | 13:22 |
| lsmola_ | SpamapS: or are there any blockers with this approach? | 13:23 |
| *** jdob has joined #tripleo | 13:30 | |
| *** e0ne has joined #tripleo | 13:37 | |
| *** pblaho has quit IRC | 13:40 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates cinder element https://review.openstack.org/62693 | 13:40 |
| *** e0ne_ has quit IRC | 13:41 | |
| *** jayg|g0n3 is now known as jayg | 13:41 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates iptables rule for MySQL element https://review.openstack.org/66569 | 13:47 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates boot-stack iptables file https://review.openstack.org/66571 | 13:51 |
| *** weshay has joined #tripleo | 13:51 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates gearman iptables file https://review.openstack.org/66572 | 13:53 |
| *** akuznetsov has quit IRC | 13:55 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates iptables rule for MySQL element https://review.openstack.org/66569 | 13:57 |
| *** rushiagr has joined #tripleo | 14:01 | |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/tripleo-heat-templates: Avoid echoing comment from makefile recipe https://review.openstack.org/66575 | 14:03 |
| openstackgerrit | Mark McLoughlin proposed a change to openstack/tripleo-heat-templates: Add all and clean targets to makefile https://review.openstack.org/66576 | 14:03 |
| *** dprince has joined #tripleo | 14:04 | |
| rbrady | Looking for additional reviews for https://review.openstack.org/#/c/62735/ | 14:04 |
| *** lblanchard has joined #tripleo | 14:05 | |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Add Nodes Overview page https://review.openstack.org/66522 | 14:11 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates boot-stack iptables file https://review.openstack.org/66571 | 14:12 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates iptables rule for MySQL element https://review.openstack.org/66569 | 14:15 |
| *** ftcjeff has joined #tripleo | 14:18 | |
| *** julim has joined #tripleo | 14:22 | |
| *** ftcjeff has quit IRC | 14:30 | |
| *** ftcjeff has joined #tripleo | 14:30 | |
| derekh | Is anybody else having trouble with nova client in images? http://paste.openstack.org/show/61191/ | 14:31 |
| derekh | looks like the system version of six is being used instead of the venv one | 14:32 |
| *** jprovazn has quit IRC | 14:33 | |
| *** spzala has joined #tripleo | 14:33 | |
| *** matty_dubs|gone is now known as matty_dubs | 14:48 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Updates nova notty file https://review.openstack.org/66593 | 14:49 |
| *** nosnos has quit IRC | 14:50 | |
| *** jrist has joined #tripleo | 14:51 | |
| *** e0ne_ has joined #tripleo | 14:59 | |
| *** slagle has quit IRC | 15:02 | |
| *** e0ne has quit IRC | 15:02 | |
| *** sdake has quit IRC | 15:03 | |
| *** coolsvap has joined #tripleo | 15:04 | |
| *** vkozhukalov has quit IRC | 15:04 | |
| openstackgerrit | Mark McLoughlin proposed a change to openstack-infra/tripleo-ci: Fix some nitpicks in testenv-work element README https://review.openstack.org/66595 | 15:06 |
| *** sdake has joined #tripleo | 15:15 | |
| openstackgerrit | Roman Podoliaka proposed a change to openstack/tripleo-incubator: Use wait condition when waiting for undercloud https://review.openstack.org/66600 | 15:18 |
| openstackgerrit | Roman Podoliaka proposed a change to openstack/tripleo-incubator: Preserve USE_CACHE value in devtest_setup.sh https://review.openstack.org/66601 | 15:18 |
| openstackgerrit | Roman Podoliaka proposed a change to openstack/tripleo-incubator: Cleanup docs https://review.openstack.org/66602 | 15:18 |
| *** lucasagomes is now known as lucas-hungry | 15:27 | |
| *** rushiagr is now known as rushiagr_away | 15:28 | |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Move templates for nodes.overview https://review.openstack.org/66606 | 15:34 |
| *** kpavel has quit IRC | 15:41 | |
| *** akrivoka has joined #tripleo | 15:44 | |
| openstackgerrit | Roman Podoliaka proposed a change to openstack/tripleo-incubator: Cleanup docs https://review.openstack.org/66602 | 15:44 |
| *** rushiagr_away is now known as rushiagr | 15:46 | |
| rpodolyaka | derekh: I can confirm six version issue (ubuntu saucy, image built 1 hour ago) | 15:48 |
| derekh | rpodolyaka: looks like nova client isn't using the python-six installed into the venv instead using the globally installed one | 15:49 |
| derekh | rpodolyaka: this used to be ok because os-refresh-config was installing globally | 15:49 |
| derekh | rpodolyaka: but yesterday I moved it to a venv, so python-six is no longer upgraded by the install of os-refresh-config (or may its os-collect-config) | 15:50 |
| rpodolyaka | derekh: hmm, but doesn't this mean that novaclient has six version specified incorrectly in its requirements.txt? or it's just installed without -U? | 15:52 |
| rpodolyaka | so it must be missing -U option, requirements.txt has six>=1.4.1 | 15:54 |
| rpodolyaka | as well as site-packages=True | 15:54 |
| *** newell has joined #tripleo | 15:55 | |
| openstackgerrit | Imre Farkas proposed a change to openstack/tripleo-image-elements: Add iptables rule for Ironic in Fedora https://review.openstack.org/66610 | 15:55 |
| derekh | rpodolyaka: well it's installing 1.4.1 into the venv | 15:56 |
| derekh | but using the global one http://paste.openstack.org/show/61193/ | 15:56 |
| rpodolyaka | oh, nice | 15:56 |
| *** mrunge has joined #tripleo | 15:59 | |
| *** akuznetsov has joined #tripleo | 16:14 | |
| *** AaronGr_Zzz is now known as AaronGr | 16:16 | |
| *** jcoufal has quit IRC | 16:16 | |
| *** e0ne_ has quit IRC | 16:19 | |
| SpamapS | derekh: network on cd-undercloud is definitely flaky | 16:20 |
| *** e0ne has joined #tripleo | 16:20 | |
| SpamapS | lsmola_: we are not going to use nested resources in HOT, because that is not composable enough. | 16:20 |
| SpamapS | lsmola_: we are going to use some of that, but we will also use the software config stuff which will let us combine multiple metadata sections in one server. | 16:21 |
| SpamapS | lsmola_: it will also free us of explicit user/accesskey/policy definition | 16:21 |
| *** boris-42 has quit IRC | 16:21 | |
| SpamapS | lsmola_: oh and waitcondition definition too :) | 16:21 |
| *** lucas-hungry is now known as lucasagomes | 16:23 | |
| *** jistr has quit IRC | 16:23 | |
| derekh | SpamapS: I was kinda hoping it was a problem where ever their hosted but if any of us have changed anything in there recently its probably more likely | 16:23 |
| *** e0ne has quit IRC | 16:24 | |
| SpamapS | derekh: that box has a history of flaky network | 16:24 |
| derekh | SpamapS: ahh yes, didn't something need to be restarted periodically | 16:26 |
| SpamapS | derekh: we actually have to rmmod/insmod the network driver. | 16:28 |
| SpamapS | derekh: but lately we haven't needed to so I was hoping maybe it was handled elsewhere. :-P | 16:28 |
| *** bauzas has quit IRC | 16:36 | |
| derekh | SpamapS: hmm, I can't ssh in anymore | 16:36 |
| SpamapS | derekh: I'll ssh to our ilo bastion host and see if we can resolve there | 16:38 |
| *** mrunge has quit IRC | 16:38 | |
| derekh | SpamapS: ok, thanks, no hurry (for me at least), I'll pick up on the ci stuff tomorrow | 16:39 |
| * derekh just got ssh'd in for a moment but lost it again | 16:41 | |
| SpamapS | derekh: ok I kicked the mellanox driver .. should be more reliable now | 16:42 |
| SpamapS | though inevitably I'll break something else. :-P | 16:43 |
| *** slagle has joined #tripleo | 16:44 | |
| derekh | SpamapS: thanks, still feels a little like packets are doing a lap of the moon but I think its better | 16:44 |
| *** viktors has left #tripleo | 16:45 | |
| SpamapS | yeah it does | 16:46 |
| SpamapS | I think this may be something else | 16:46 |
| *** rdopieralski has quit IRC | 16:47 | |
| derekh | SpamapS: must be something else, terminal unresponsive again | 16:49 |
| SpamapS | no I did that.. I think something is bong on the bridge setup | 16:49 |
| SpamapS | ok actually that was the local bridges needing to flush cache | 16:50 |
| *** beekneemech is now known as bnemec | 16:50 | |
| *** blamar has quit IRC | 16:53 | |
| *** gothicmindfood has joined #tripleo | 16:54 | |
| *** blamar has joined #tripleo | 16:55 | |
| *** gothicmindfood has left #tripleo | 16:57 | |
| *** gothicmindfood_ has joined #tripleo | 17:01 | |
| *** vkozhukalov has joined #tripleo | 17:06 | |
| *** akuznetsov has quit IRC | 17:06 | |
| *** matty_dubs is now known as matty_dubs|lunch | 17:09 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Updates cinder element https://review.openstack.org/62693 | 17:09 |
| openstackgerrit | James Slagle proposed a change to openstack/diskimage-builder: Wait for tgtd socket to be available https://review.openstack.org/66639 | 17:11 |
| SpamapS | hmmmm | 17:12 |
| SpamapS | looking at https://review.openstack.org/62693 ... | 17:12 |
| SpamapS | I wonder if /var/run is correct.. | 17:12 |
| SpamapS | I _think_ that means on every reboot the iptables rules will be duplicated. | 17:12 |
| * SpamapS should think before writing those comments.. doh | 17:12 | |
| anteaya | lifeless mikal for when you are awake, please note there is another 3rd party testing meeting scheduled: http://lists.openstack.org/pipermail/openstack-dev/2014-January/024297.html | 17:16 |
| anteaya | would be great if you could attend and share your thoughts | 17:16 |
| anteaya | sdague too ^^ | 17:17 |
| anteaya | plus anyone else who wants to attend | 17:17 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Updates nova notty file https://review.openstack.org/66593 | 17:17 |
| *** akuznetsov has joined #tripleo | 17:18 | |
| sdague | anteaya: just remind me in the room, I'll join it :) | 17:18 |
| Ng | SpamapS: hmm, that's an interesting point actually | 17:18 |
| slagle | SpamapS: hmm, good point. /var/run is tmpfs mounted | 17:18 |
| slagle | so it should be something under /mnt/state | 17:19 |
| Ng | yeah, I suspect a fedora-iptables element is needed, to collect all these .ok files | 17:19 |
| Ng | does fedora not have some kind of idempotent service-level firewall tool like ufw? | 17:20 |
| SpamapS | I think it would make sense to simplify that logic into a little tool | 17:21 |
| Ng | yeah, that also makes sense | 17:22 |
| SpamapS | Ng: oh.. ufw.. why aren't we using that? | 17:22 |
| SpamapS | I sometimes forget it exists :) | 17:22 |
| Ng | SpamapS: does that even on fedora? | 17:22 |
| SpamapS | Ng: I see no reason why it wouldn't. | 17:22 |
| SpamapS | Other than politics. | 17:22 |
| slagle | fedora has firewalld, but not in the cloud images | 17:22 |
| slagle | too big a footprint | 17:22 |
| Ng | I don't see any immediate google juice for ufw being available on fedora | 17:24 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Updates gearman iptables file https://review.openstack.org/66572 | 17:25 |
| Ng | dammit | 17:25 |
| Ng | too late to unapprove that one | 17:25 |
| anteaya | sdague: #openstack-meeting-alt and thank you | 17:25 |
| *** boris-42 has joined #tripleo | 17:26 | |
| slagle | actually there is a iptables-services rpm for fedora that has this same save/restore logic | 17:27 |
| slagle | so we should probably just use that. have each element just open the ports it needs, then iptables-services that runs save at the very end | 17:28 |
| *** max_lobur is now known as max_lobur_afk | 17:28 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Update apache2 to use the os-apply-config dir. https://review.openstack.org/66470 | 17:31 |
| derekh | rlandy: ^ see iptables conversation above, related to the iptables script I mentioned earlier. | 17:37 |
| *** digambar_ has joined #tripleo | 17:42 | |
| *** matty_dubs|lunch is now known as matty_dubs | 17:44 | |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Avoid echoing comment from makefile recipe https://review.openstack.org/66575 | 17:44 |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Add all and clean targets to makefile https://review.openstack.org/66576 | 17:44 |
| *** akuznetsov has quit IRC | 17:46 | |
| * SpamapS is always happy and horrified to see merge.py related things land | 17:48 | |
| Ng | I am never happy to review makefiles ;) | 17:49 |
| *** akuznetsov has joined #tripleo | 17:49 | |
| Ng | it's 2014! | 17:49 |
| rlandy | derekh: ok - I was about to ask what happened to service firewalld but I see slagle answered that question. | 17:49 |
| SpamapS | Ng: all debian packages are make files | 17:49 |
| SpamapS | So... | 17:49 |
| SpamapS | such is life | 17:49 |
| Ng | SpamapS: sure, but debian is the 1990s making the best of what the 1980s left them with. perl and make. | 17:50 |
| *** akuznetsov has quit IRC | 17:50 | |
| digambar_ | Hi | 17:50 |
| digambar_ | Getting error | 17:50 |
| digambar_ | ronic node-create -d fake_ssh -i ssh_address=192.168.122.1 -i ssh_username=root -i ssh_password=openstack -i ssh_virt_type=virsh | 17:50 |
| digambar_ | HTTPForbidden (HTTP 403) | 17:50 |
| digambar_ | This is authentication error | 17:51 |
| SpamapS | Ng: that _might_ end up on twitter if I can navigate the distraction maze to get there | 17:51 |
| Ng | SpamapS: I will retoot it, if you do :) | 17:51 |
| openstackgerrit | James Slagle proposed a change to openstack/diskimage-builder: Wait for tgtd socket to be available https://review.openstack.org/66639 | 17:53 |
| *** jcoufal-m has joined #tripleo | 17:53 | |
| *** derekh has quit IRC | 17:54 | |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-incubator: Update write-tripleorc https://review.openstack.org/66655 | 18:02 |
| *** bauzas has joined #tripleo | 18:04 | |
| openstackgerrit | Jay Dobies proposed a change to openstack/tuskar: Tuskar domain model updates for wireframes https://review.openstack.org/66062 | 18:05 |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Add undercloud-vm-ironic-source yaml file https://review.openstack.org/66115 | 18:06 |
| * Ng breaks for dinner | 18:08 | |
| *** jcoufal has joined #tripleo | 18:09 | |
| *** mrunge has joined #tripleo | 18:10 | |
| *** rushiagr has quit IRC | 18:15 | |
| *** spzala_ has joined #tripleo | 18:19 | |
| *** spzala has quit IRC | 18:21 | |
| *** spzala_ is now known as spzala | 18:21 | |
| *** bauzas has quit IRC | 18:21 | |
| *** akrivoka has quit IRC | 18:23 | |
| lifeless | morning | 18:25 |
| *** martyntaylor has quit IRC | 18:28 | |
| *** mrunge has quit IRC | 18:29 | |
| lifeless | SpamapS: FWIW I'm in complete agreement that haproxy shouldn't care whats installed locally | 18:32 |
| SpamapS | lifeless: Yeah I am worried that I'm just not understanding why the heat template isn't the place to configure this. | 18:34 |
| lifeless | anteaya: sdague: I would, but I have a medical thing on then, sorry. | 18:35 |
| *** vipul is now known as vipul-away | 18:35 | |
| lifeless | slagle: Ng: SpamapS: FWIW, something that asserts iptables is likely better than save-restore. IMNSHO. | 18:36 |
| lifeless | SpamapS: [haproxy] it was, in an earlier iteration. | 18:37 |
| SpamapS | lifeless: ufw basically works that way. Packages ship service definitions to make it easy to poke holes in the firewall. | 18:38 |
| SpamapS | and then ufw runs and asserts that those holes are there, and that the default rules are in place | 18:38 |
| *** rushiagr has joined #tripleo | 18:40 | |
| lifeless | right | 18:41 |
| lifeless | better than incremental editing | 18:41 |
| SpamapS | 293 packets transmitted, 134 received, 54% packet loss, time 293170ms | 18:45 |
| SpamapS | something is not right with cd-undercloud | 18:45 |
| *** ccrouch has quit IRC | 18:45 | |
| *** martyntaylor has joined #tripleo | 18:45 | |
| SpamapS | 10 packets transmitted, 10 received, 0% packet loss, time 9011ms | 18:45 |
| mordred | lifeless, SpamapS: speaking of iptables - any ideas on how to start supporting nftables from 3.13? | 18:45 |
| SpamapS | bastion host is fine | 18:45 |
| SpamapS | mordred: looks like just start replacign all the other ones with nftables calls. | 18:46 |
| SpamapS | replacing even | 18:46 |
| SpamapS | " The nftables project provides a backward compatibility layer that allows you run iptables/ip6tables (using the same syntax) over the nftables infrastructure" | 18:46 |
| *** ccrouch has joined #tripleo | 18:47 | |
| SpamapS | mordred: so it can be incremental looksl ike | 18:47 |
| mordred | excellent | 18:47 |
| *** ccrouch has quit IRC | 18:48 | |
| lifeless | mordred: I thought there was a compat layer | 18:50 |
| lifeless | mordred: so, step a) don't, until b) we don't have iptables on a supported platform | 18:50 |
| mordred | lifeless: just forward looking to see if there was anything needed doing | 18:51 |
| *** jcoufal-m has quit IRC | 18:51 | |
| lifeless | mordred: ah, so with kernel stuff you can rely on backwards compat from userspace :) | 18:52 |
| lifeless | mordred: however a bigger question is when we want to take advantage of it's new features | 18:52 |
| mordred | lifeless: yah | 18:53 |
| *** vipul-away is now known as vipul | 18:53 | |
| *** rpodolyaka1 has joined #tripleo | 18:56 | |
| *** jprovazn has joined #tripleo | 18:57 | |
| *** e0ne has joined #tripleo | 18:57 | |
| *** gothicmindfood_ has left #tripleo | 18:58 | |
| lifeless | meeting time y'all | 18:59 |
| *** jistr has joined #tripleo | 18:59 | |
| *** pblaho has joined #tripleo | 19:00 | |
| *** ccrouch has joined #tripleo | 19:05 | |
| anteaya | lifeless: okay thanks | 19:08 |
| *** rushiagr has quit IRC | 19:10 | |
| *** spzala has quit IRC | 19:18 | |
| *** digambar_ has quit IRC | 19:21 | |
| *** bauzas has joined #tripleo | 19:23 | |
| *** vkozhukalov has quit IRC | 19:29 | |
| *** athomas has quit IRC | 19:34 | |
| *** athomas has joined #tripleo | 19:48 | |
| openstackgerrit | Lucas Alvares Gomes proposed a change to openstack/diskimage-builder: Fix deploy-ironic init script https://review.openstack.org/66677 | 19:48 |
| *** vipul is now known as vipul-away | 19:52 | |
| *** vipul-away is now known as vipul | 19:52 | |
| *** CaptTofu has joined #tripleo | 19:56 | |
| *** lucasagomes has quit IRC | 19:57 | |
| *** jcoufal has quit IRC | 20:00 | |
| SpamapS | and after that we need to fix Heat to not delete "the last server added" when a scaling group is scaled down | 20:00 |
| *** sdake has quit IRC | 20:01 | |
| SpamapS | otherwise we'll have to push merge.py harder to generate templates and then manually delete things from said templates. | 20:01 |
| *** pblaho has quit IRC | 20:01 | |
| *** sdake has joined #tripleo | 20:01 | |
| *** sdake has joined #tripleo | 20:01 | |
| rlandy | dprince: could I also get on that CI/testing call tomorrow? | 20:02 |
| *** jcoufal has joined #tripleo | 20:02 | |
| dprince | rlandy: sure. | 20:03 |
| ccrouch | (2:00:01 PM) lsmola: SpamapS, so we are keepig merge.py in I ? | 20:03 |
| ccrouch | is a good question | 20:03 |
| SpamapS | lifeless: anyway, my point is, there seems to be this "only SpamapS can do Heat" thing that is not going to scale well. | 20:03 |
| *** e0ne has quit IRC | 20:03 | |
| SpamapS | ccrouch: I have no idea. | 20:03 |
| shadower | SpamapS: I'm going to be more involved with Heat if that helps | 20:04 |
| *** e0ne has joined #tripleo | 20:04 | |
| ccrouch | SpamapS: fair enough. its not a goal i'm aware we've articulated | 20:04 |
| SpamapS | I'd love to get rid of it. It would make our dev more smooth to just write templates and not have to shove them through merge.py. But we have something that works so meh if it stays. | 20:04 |
| SpamapS | "bigger fish to fry" | 20:04 |
| SpamapS | I think I'm more worried about scaling out now | 20:05 |
| SpamapS | the retry thing and updatable wait condition thing are important, but we need to switch to using instancegroup or resource group and that is going to cause issues. | 20:05 |
| *** e0ne has quit IRC | 20:06 | |
| SpamapS | the other option is to just scale out by making a stack-per-compute. | 20:06 |
| *** e0ne has joined #tripleo | 20:06 | |
| lifeless | SpamapS: so, I think the spamaps-only-heat is a result of surfacing work | 20:06 |
| lifeless | SpamapS: till now most of our work hasn't been heat, and now - because we've finally pushed hard on update etc, it will be for a while | 20:06 |
| lifeless | SpamapS: so I think what you're doing - surfacing that there is heat work to do - is good | 20:07 |
| lifeless | SpamapS: but lets not lose focus at the same time; so keep the surfaced work fairly narrow, and collaborate on patches | 20:07 |
| lifeless | SpamapS: our current story, which AIUI we haven't delivered yet, is being able to do an upgrade, with downtime. | 20:07 |
| lifeless | SpamapS: we're close right? but not -there- | 20:08 |
| *** vipul is now known as vipul-away | 20:08 | |
| *** notq has joined #tripleo | 20:08 | |
| lifeless | did we have a heat template for running up the broker? | 20:10 |
| SpamapS | lifeless: I think we're there but we're not waiting long enough because the wait conditions don't wait. | 20:10 |
| lifeless | SpamapS: so we're not there, but the fix is in heat :) | 20:10 |
| lifeless | SpamapS: so once thats done, I think the next thing is graceful deploys w/out downtime - so HA control and HA compute, live migration and rolling upgrades. | 20:11 |
| SpamapS | yeah, so I'll whip up a second attempt at that patch today | 20:11 |
| lifeless | thats a bunch of fodder and we may want to have a bit of divergence on what folk tackle there, because it's still fairly narrow. | 20:11 |
| SpamapS | yeah, HA means "n" things.. and we don't have "n" things possible right now, we have static servers defined | 20:11 |
| *** CaptTofu has quit IRC | 20:12 | |
| SpamapS | anyway, in all of that we're also going to need retrying of updates to work | 20:12 |
| SpamapS | because OpenStack is non-deterministic :) | 20:12 |
| *** vipul-away is now known as vipul | 20:12 | |
| ftcjeff | So I'm trying to run devtest.sh, but I get this: | 20:17 |
| ftcjeff | distutils.errors.DistutilsError: Could not find suitable distribution for Requirement.parse('pbr') | 20:17 |
| lifeless | SpamapS: right, but I guess I'm saying lets be guided by what breaks first | 20:17 |
| ftcjeff | Anyone know what mess I've created for myself? I've seen this before doing horizon code, but I can't remember how I got out of it | 20:18 |
| lifeless | SpamapS: e.g. lets get 3 control plane and 2 compute | 20:18 |
| lifeless | SpamapS: HA != any-N | 20:18 |
| lifeless | ftcjeff: thats a cascade errror, the actual failure will be higher up | 20:18 |
| lifeless | ftcjeff: e.g. a SSL error or some such | 20:18 |
| ftcjeff | Download error on https://pypi.python.org/simple/pbr/: [Errno 1] _ssl.c:509: error:140770FC:SSL routines:SSL23_GET_SERVER_HELLO:unknown protocol -- Some packages may not be found! | 20:19 |
| ftcjeff | ? | 20:19 |
| lifeless | yup | 20:19 |
| lifeless | caused by http proxies with older pip IIRC | 20:19 |
| ftcjeff | okay, I'll try a different proxy | 20:19 |
| ftcjeff | thanks :-) | 20:19 |
| *** rlandy has quit IRC | 20:20 | |
| *** gothicmindfood has joined #tripleo | 20:21 | |
| *** gothicmindfood has left #tripleo | 20:21 | |
| *** vipul is now known as vipul-away | 20:27 | |
| *** jtomasek has quit IRC | 20:39 | |
| *** jtomasek has joined #tripleo | 20:41 | |
| *** jistr has quit IRC | 20:42 | |
| lifeless | we don't have /var/log/upstart persisted yet either | 20:43 |
| lifeless | maybe we shouldn't; just pull straight into logstash. | 20:43 |
| lifeless | thought for another day | 20:44 |
| SpamapS | lifeless: Ok well what has broken already meany times is update fails for abitrary reason and we just have to kick it to retry. | 20:45 |
| lifeless | SpamapS: so maybe that should be next | 20:45 |
| SpamapS | it is 33% done | 20:46 |
| SpamapS | CREATE FAILED works fine.. we delete/create | 20:46 |
| lifeless | SpamapS: either in MVP or the decreasefragility | 20:46 |
| SpamapS | DELETE FAILED just needs to retry DELETE | 20:46 |
| SpamapS | and UPDATE FAILED needs to retry UPDATE | 20:46 |
| SpamapS | lifeless: I don't think this is fragility. It is part of the story to keep it running "somehow" | 20:47 |
| SpamapS | fragility is it breaks on every update | 20:47 |
| lifeless | SpamapS: I think this is splitting hairs; we both want the work done | 20:47 |
| SpamapS | Right. Anyway, I need to run some errands. bbiab | 20:48 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Fix broken iptables command in cinder https://review.openstack.org/66684 | 20:50 |
| openstackgerrit | Jay Dobies proposed a change to openstack/tuskar: Tuskar domain model updates for wireframes https://review.openstack.org/66062 | 20:52 |
| rbrady | dprince: ping | 20:57 |
| *** lblanchard has quit IRC | 20:58 | |
| dprince | rbrady: hi | 20:59 |
| dprince | rbrady: gotta run in a minute but what is up? | 21:00 |
| rbrady | re: your cinder change. I just posted the comments | 21:01 |
| lifeless | yay ova secgroup-list | 21:01 |
| lifeless | ERROR: The server has either erred or is incapable of performing the requested operation. (HTTP 500) (Request-ID: req-3d2dd2a7-8b23-416d-bd96-975c3b586919) | 21:01 |
| rbrady | dprince: I'm not sure which way is the "only one obvious way" to do this | 21:02 |
| *** lblanchard has joined #tripleo | 21:03 | |
| *** lblanchard has quit IRC | 21:03 | |
| *** AaronGr is now known as AaronGr_afk | 21:03 | |
| rbrady | dprince: or if it matters after Spamaps review | 21:03 |
| lifeless | hmm, working now | 21:04 |
| dprince | rbrady: I would rather land this quick fix and then move towards using a common iptables element for this. | 21:04 |
| lifeless | and holy cow, 3 seconds to live. Thats a cloud. | 21:04 |
| rbrady | dprince: rgr | 21:04 |
| dprince | rbrady: agree that we should create /var/run/cinder in install.d... but I'm not sure that will be the eventual location we use for iptables stuff so lets get this in and move it to a common place | 21:05 |
| rbrady | dprince: updated review | 21:05 |
| dprince | rbrady: feel free to overwrite my patch if you have a better solution though, just trying to keep things working quickly | 21:05 |
| dprince | rbrady: gotta go. thanks for looking into this | 21:06 |
| *** dprince has quit IRC | 21:06 | |
| *** julim has quit IRC | 21:06 | |
| *** rpodolyaka1 has quit IRC | 21:09 | |
| ftcjeff | I got a little further this time: | 21:22 |
| ftcjeff | Waiting for keystone to initialise... | 21:22 |
| ftcjeff | <attribute 'message' of 'exceptions.BaseException' objects> (HTTP Unable to establish connection to http://192.0.2.1:35357/v2.0/tenants) | 21:22 |
| ftcjeff | and then it finished with: | 21:23 |
| ftcjeff | /home/jwalls/tripleo/tripleo-incubator/scripts/devtest_seed.sh: line 95: os-apply-config: command not found | 21:23 |
| *** jprovazn has quit IRC | 21:23 | |
| *** e0ne has quit IRC | 21:24 | |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Move templates for nodes.overview https://review.openstack.org/66606 | 21:26 |
| *** vipul-away is now known as vipul | 21:26 | |
| *** newell_ has joined #tripleo | 21:33 | |
| *** newell has quit IRC | 21:36 | |
| *** ifarkas has quit IRC | 21:45 | |
| *** markmc has quit IRC | 21:49 | |
| *** mrunge has joined #tripleo | 21:50 | |
| *** AaronGr_afk is now known as AaronGr | 21:53 | |
| *** jdob has quit IRC | 21:54 | |
| *** athomas has quit IRC | 21:59 | |
| *** lblanchard has joined #tripleo | 21:59 | |
| *** jtomasek has quit IRC | 22:01 | |
| *** jtomasek has joined #tripleo | 22:01 | |
| *** modi_vnc has joined #tripleo | 22:17 | |
| *** julim has joined #tripleo | 22:29 | |
| *** e0ne has joined #tripleo | 22:35 | |
| openstackgerrit | James Slagle proposed a change to openstack/diskimage-builder: Symlink correct element install type https://review.openstack.org/66724 | 22:38 |
| openstackgerrit | James Slagle proposed a change to openstack/tripleo-image-elements: Install nova based on source installation type https://review.openstack.org/65806 | 22:38 |
| *** e0ne has quit IRC | 22:39 | |
| SpamapS | ftcjeff: os-apply-config is installed in a virtualenv and then symlinked in tripleo-incubator/scripts ... so you should have it on your path if scripts is on your path | 22:40 |
| SpamapS | ftcjeff: devtest_variables.sh should handle htat | 22:41 |
| SpamapS | that | 22:41 |
| ftcjeff | spamaps: I don't see os-apply-config in tripleo-incubator/scripts :( I'll keep looking for what might have gone wrong | 22:43 |
| *** jcoufal has quit IRC | 22:43 | |
| *** mrunge has quit IRC | 22:44 | |
| *** CaptTofu has joined #tripleo | 22:49 | |
| SpamapS | ftcjeff: the symlink is setup in setup-clienttools | 22:51 |
| SpamapS | ftcjeff: which is called in install-dependencies and refresh-env | 22:51 |
| SpamapS | ftcjeff: so devtest_setup.sh should have handled that (it is run from devtest.sh) | 22:52 |
| *** julim has quit IRC | 22:52 | |
| ftcjeff | spamaps: so it's linked from tripleo-incubator/openstack-tools/bin; but it doesn't exist there either | 22:53 |
| *** lblanchard has quit IRC | 22:55 | |
| SpamapS | ftcjeff: I wonder if we're not detecting a failure to install that | 22:56 |
| ftcjeff | spamaps: unfortunately, my scroll buffer doesn't go back very far :( | 22:57 |
| SpamapS | ftcjeff: ah-hah! | 22:57 |
| SpamapS | bug | 22:57 |
| SpamapS | http://paste.openstack.org/show/61220 | 22:57 |
| SpamapS | ftcjeff: I don't have os-apply-config either.. | 22:57 |
| SpamapS | ftcjeff: the dir is never cleared out | 22:58 |
| SpamapS | ftcjeff: so we skip if we've already installed before.. but new tools have been added | 22:58 |
| SpamapS | ftcjeff: I know you were looking for each bugs but this is a one liner. :) | 22:59 |
| ftcjeff | spamaps: cool | 22:59 |
| SpamapS | ftcjeff: you can work around it by rm -r'ing openstack-tools | 23:00 |
| ftcjeff | spamaps: want me to try to fix that? :-) | 23:02 |
| openstackgerrit | Clint "SpamapS" Byrum proposed a change to openstack/tripleo-incubator: Refresh openstack-tools virtualenv every time https://review.openstack.org/66727 | 23:02 |
| SpamapS | ftcjeff: too late :) | 23:02 |
| ftcjeff | darn! | 23:02 |
| *** CaptTofu has quit IRC | 23:04 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Fix broken iptables command in cinder https://review.openstack.org/66684 | 23:04 |
| *** ccrouch1 has joined #tripleo | 23:09 | |
| *** CaptTofu has joined #tripleo | 23:10 | |
| *** ccrouch has quit IRC | 23:10 | |
| *** jtomasek has quit IRC | 23:10 | |
| *** modi_vnc has quit IRC | 23:12 | |
| *** CaptTofu has quit IRC | 23:18 | |
| *** CaptTofu has joined #tripleo | 23:19 | |
| *** CaptTofu has quit IRC | 23:19 | |
| *** ftcjeff has quit IRC | 23:25 | |
| *** CaptTofu has joined #tripleo | 23:36 | |
| *** ccrouch1 has quit IRC | 23:45 | |
| *** martyntaylor has quit IRC | 23:46 | |
| StevenK | lifeless: Oh, we should bump the 64bit docs in _variables, now that I remember. Double everything, you think? | 23:49 |
| lifeless | StevenK: I'm not sure; the difference really is the python footprint | 23:50 |
| StevenK | Right, it's about 30% bigger or so | 23:50 |
| lifeless | StevenK: data would be a good thing | 23:50 |
| *** spzala has joined #tripleo | 23:53 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!