*** DimGR2 is now known as DimGR | 00:00 | |
*** Jack_Iv has joined #openstack-ansible | 00:14 | |
*** Jack_Iv has quit IRC | 00:19 | |
*** woodard_ has joined #openstack-ansible | 00:20 | |
*** woodard has quit IRC | 00:20 | |
*** gouthamr has joined #openstack-ansible | 00:27 | |
*** dxiri has quit IRC | 00:32 | |
*** adrian_otto has joined #openstack-ansible | 00:33 | |
*** Jack_Iv has joined #openstack-ansible | 00:35 | |
*** Jack_Iv has quit IRC | 00:40 | |
*** h16mara_ has quit IRC | 00:41 | |
*** agrebennikov has quit IRC | 00:42 | |
*** cuongnv has joined #openstack-ansible | 00:48 | |
*** h16mara_ has joined #openstack-ansible | 00:49 | |
*** rmelero has quit IRC | 00:49 | |
*** SerenaFeng has joined #openstack-ansible | 00:56 | |
*** DimGR has quit IRC | 00:58 | |
*** adrian_otto has quit IRC | 00:59 | |
*** esberglu has quit IRC | 01:13 | |
*** gouthamr has quit IRC | 01:16 | |
*** MasterOfBugs has quit IRC | 01:23 | |
*** SerenaFeng has quit IRC | 01:24 | |
*** japestinho has joined #openstack-ansible | 01:32 | |
*** Jack_Iv has joined #openstack-ansible | 01:36 | |
*** SerenaFeng has joined #openstack-ansible | 01:40 | |
*** pmannidi_ has joined #openstack-ansible | 01:40 | |
*** Jack_Iv has quit IRC | 01:40 | |
*** weezS has quit IRC | 01:43 | |
*** pmannidi has quit IRC | 01:43 | |
*** Andrew_jedi has joined #openstack-ansible | 01:44 | |
*** mpotdar has joined #openstack-ansible | 01:44 | |
*** ianychoi has quit IRC | 01:48 | |
*** Andrew_jedi has quit IRC | 01:50 | |
*** weezS has joined #openstack-ansible | 01:51 | |
*** ianychoi has joined #openstack-ansible | 01:54 | |
*** SerenaFeng has quit IRC | 01:58 | |
openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible master: [WIP] Support cidr_networks in L3 network environments https://review.openstack.org/448850 | 02:07 |
---|---|---|
*** NostawRm has quit IRC | 02:23 | |
*** NostawRm has joined #openstack-ansible | 02:23 | |
*** agrebennikov_ has joined #openstack-ansible | 02:27 | |
*** Jack_Iv has joined #openstack-ansible | 02:36 | |
*** Jack_Iv has quit IRC | 02:41 | |
*** Mahe has quit IRC | 02:42 | |
*** Mahe has joined #openstack-ansible | 02:43 | |
*** manheim has joined #openstack-ansible | 02:45 | |
*** manheim has quit IRC | 02:50 | |
*** agrebennikov has joined #openstack-ansible | 02:58 | |
*** gouthamr has joined #openstack-ansible | 02:59 | |
*** sanfern has joined #openstack-ansible | 03:00 | |
*** lihi has quit IRC | 03:06 | |
*** lihi has joined #openstack-ansible | 03:07 | |
*** acormier has joined #openstack-ansible | 03:14 | |
openstackgerrit | Ravi Kumar Boyapati proposed openstack/openstack-ansible-lxc_hosts master: Fix the lock type variable https://review.openstack.org/448872 | 03:18 |
*** gouthamr has quit IRC | 03:19 | |
openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible master: Support cidr_networks in L3 network environments https://review.openstack.org/448850 | 03:33 |
*** acormier has quit IRC | 03:36 | |
*** Jack_Iv has joined #openstack-ansible | 03:37 | |
*** Jack_Iv has quit IRC | 03:42 | |
*** h16mara_ has quit IRC | 03:53 | |
*** h16mara_ has joined #openstack-ansible | 03:58 | |
*** udesale has joined #openstack-ansible | 04:07 | |
*** Amit82 has joined #openstack-ansible | 04:11 | |
Amit82 | Hi All, can we create multiple FLAT tenant networks? | 04:11 |
sc` | it would seem systemd-resolve comes about in newer releases of systemd, which centos 7 does not have. fun. | 04:11 |
*** pmannidi_ has quit IRC | 04:11 | |
*** NikhilS has joined #openstack-ansible | 04:19 | |
*** pmannidi_ has joined #openstack-ansible | 04:28 | |
*** manheim has joined #openstack-ansible | 04:30 | |
*** manheim has quit IRC | 04:34 | |
*** jrobinson has quit IRC | 04:36 | |
*** pmannidi_ has quit IRC | 04:43 | |
*** MasterOfBugs has joined #openstack-ansible | 04:44 | |
*** agrebennikov has quit IRC | 04:46 | |
*** agrebennikov_ has quit IRC | 04:46 | |
*** Jeffrey4l has joined #openstack-ansible | 04:55 | |
*** Jeffrey4l_ has quit IRC | 04:55 | |
*** pmannidi_ has joined #openstack-ansible | 05:00 | |
*** Jack_Iv has joined #openstack-ansible | 05:02 | |
*** shashank_t_ has joined #openstack-ansible | 05:03 | |
*** Jack_Iv has quit IRC | 05:06 | |
*** markvoelker has quit IRC | 05:08 | |
*** markvoelker has joined #openstack-ansible | 05:09 | |
*** markvoelker has quit IRC | 05:13 | |
*** jrobinson has joined #openstack-ansible | 05:14 | |
*** Jack_Iv has joined #openstack-ansible | 05:23 | |
*** cuongnv has quit IRC | 05:25 | |
*** Jack_Iv has quit IRC | 05:28 | |
*** cuongnv has joined #openstack-ansible | 05:32 | |
*** asmaa has joined #openstack-ansible | 05:41 | |
*** SerenaFeng has joined #openstack-ansible | 05:53 | |
*** arif-ali has quit IRC | 05:56 | |
*** shashank_t_ has quit IRC | 05:57 | |
*** Jack_Iv has joined #openstack-ansible | 05:58 | |
*** arif-ali has joined #openstack-ansible | 05:58 | |
*** fxpester has joined #openstack-ansible | 06:00 | |
*** qiliang28 has joined #openstack-ansible | 06:04 | |
*** askb_ has joined #openstack-ansible | 06:08 | |
*** bpetit has quit IRC | 06:08 | |
*** bpetit has joined #openstack-ansible | 06:08 | |
*** markvoelker has joined #openstack-ansible | 06:09 | |
*** Jack_Iv has quit IRC | 06:10 | |
*** jrobinson has quit IRC | 06:11 | |
*** lkoranda_ has joined #openstack-ansible | 06:12 | |
*** Jack_Iv has joined #openstack-ansible | 06:12 | |
*** fedruantine has joined #openstack-ansible | 06:12 | |
*** lkoranda has quit IRC | 06:13 | |
*** askb has quit IRC | 06:13 | |
*** fedruantine_ has quit IRC | 06:13 | |
*** qiliang27 has quit IRC | 06:13 | |
*** lkoranda_ is now known as lkoranda | 06:13 | |
*** qiliang28 is now known as qiliang27 | 06:13 | |
*** markvoelker has quit IRC | 06:13 | |
*** john51 has quit IRC | 06:14 | |
*** jrobinson has joined #openstack-ansible | 06:15 | |
*** john51 has joined #openstack-ansible | 06:15 | |
*** tonytan4ever has quit IRC | 06:26 | |
*** tonytan4ever has joined #openstack-ansible | 06:26 | |
*** weezS has quit IRC | 06:27 | |
*** tonytan4ever has quit IRC | 06:31 | |
*** weezS has joined #openstack-ansible | 06:37 | |
*** shashank_t_ has joined #openstack-ansible | 06:47 | |
*** shashank_t_ has quit IRC | 06:50 | |
*** udesale__ has joined #openstack-ansible | 06:53 | |
*** udesale has quit IRC | 06:56 | |
openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible master: Support cidr_networks in L3 network environments https://review.openstack.org/448850 | 06:59 |
*** weezS has quit IRC | 07:01 | |
*** weezS has joined #openstack-ansible | 07:01 | |
*** z- has quit IRC | 07:02 | |
*** z- has joined #openstack-ansible | 07:03 | |
*** jasondotstar has quit IRC | 07:03 | |
*** dmellado has quit IRC | 07:03 | |
*** bpetit has quit IRC | 07:03 | |
*** admin0- has quit IRC | 07:04 | |
*** bpetit has joined #openstack-ansible | 07:05 | |
*** jasondotstar has joined #openstack-ansible | 07:08 | |
*** dmellado has joined #openstack-ansible | 07:08 | |
*** admin0 has joined #openstack-ansible | 07:09 | |
*** Jack_Iv has quit IRC | 07:10 | |
*** Jack_Iv has joined #openstack-ansible | 07:10 | |
*** jrobinson has quit IRC | 07:11 | |
openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-galera_server master: Add support for the openSUSE Leap distributions https://review.openstack.org/445248 | 07:12 |
*** udesale has joined #openstack-ansible | 07:19 | |
*** udesale__ has quit IRC | 07:23 | |
*** cuongnv has quit IRC | 07:29 | |
*** manheim has joined #openstack-ansible | 07:33 | |
*** manheim has quit IRC | 07:36 | |
*** manheim has joined #openstack-ansible | 07:36 | |
*** Jack_Iv has quit IRC | 07:40 | |
*** Jack_Iv has joined #openstack-ansible | 07:40 | |
*** askb_ has quit IRC | 07:43 | |
openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible master: Support cidr_networks in L3 network environments https://review.openstack.org/448850 | 07:49 |
*** Amit82 has quit IRC | 07:58 | |
*** cuongnv has joined #openstack-ansible | 07:58 | |
*** Mahe has quit IRC | 08:00 | |
*** askb has joined #openstack-ansible | 08:00 | |
openstackgerrit | Andy McCrae proposed openstack/openstack-ansible master: SHA Bump on Master as at 22 March 2017 https://review.openstack.org/448674 | 08:02 |
andymccr | FYI ^ worked - but we just need the dependent patch to merge and to adjust the SHA | 08:02 |
*** Mahe has joined #openstack-ansible | 08:02 | |
*** askb has quit IRC | 08:02 | |
*** askb has joined #openstack-ansible | 08:04 | |
*** shardy has joined #openstack-ansible | 08:05 | |
*** pcaruana has joined #openstack-ansible | 08:06 | |
*** vnogin has quit IRC | 08:09 | |
openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-tests stable/ocata: Bump Tempest SHA for stable/ocata https://review.openstack.org/448984 | 08:09 |
*** markvoelker has joined #openstack-ansible | 08:10 | |
*** pmannidi_ has quit IRC | 08:13 | |
*** markvoelker has quit IRC | 08:14 | |
*** manheim has quit IRC | 08:21 | |
*** Andrew_jedi has joined #openstack-ansible | 08:21 | |
*** manheim has joined #openstack-ansible | 08:21 | |
*** tonytan4ever has joined #openstack-ansible | 08:27 | |
*** DimGR has joined #openstack-ansible | 08:28 | |
*** vnogin has joined #openstack-ansible | 08:52 | |
*** qiliang28 has joined #openstack-ansible | 08:56 | |
*** SerenaFeng has quit IRC | 08:56 | |
*** SerenaFeng has joined #openstack-ansible | 08:57 | |
*** shardy has quit IRC | 08:58 | |
*** qiliang27 has quit IRC | 08:59 | |
*** qiliang28 is now known as qiliang27 | 08:59 | |
*** shardy has joined #openstack-ansible | 09:00 | |
*** openstackgerrit has quit IRC | 09:03 | |
*** lemroid has joined #openstack-ansible | 09:08 | |
*** david-lyle has quit IRC | 09:11 | |
*** david-lyle has joined #openstack-ansible | 09:13 | |
*** h16mara_ has quit IRC | 09:15 | |
*** h16mara_ has joined #openstack-ansible | 09:16 | |
*** esberglu has joined #openstack-ansible | 09:16 | |
*** esberglu has quit IRC | 09:20 | |
*** lemroid has left #openstack-ansible | 09:22 | |
*** Mahe has quit IRC | 09:24 | |
*** david-lyle_ has joined #openstack-ansible | 09:27 | |
*** Mahe has joined #openstack-ansible | 09:27 | |
*** david-lyle has quit IRC | 09:27 | |
*** tonytan4ever has quit IRC | 09:27 | |
*** manheim has quit IRC | 09:37 | |
*** manheim has joined #openstack-ansible | 09:37 | |
*** electrofelix has joined #openstack-ansible | 09:39 | |
*** karimb has joined #openstack-ansible | 09:39 | |
*** SerenaFeng has quit IRC | 09:46 | |
*** udesale has quit IRC | 09:48 | |
*** ildikov has joined #openstack-ansible | 09:49 | |
*** MasterOfBugs has quit IRC | 09:59 | |
*** rgogunskiy has joined #openstack-ansible | 10:04 | |
*** Andrew_jedi has quit IRC | 10:09 | |
*** markvoelker has joined #openstack-ansible | 10:11 | |
*** Jack_Iv has quit IRC | 10:11 | |
*** sanfern has quit IRC | 10:16 | |
*** markvoelker has quit IRC | 10:16 | |
*** cuongnv has quit IRC | 10:18 | |
pjm6 | Good morning guys :) | 10:24 |
pjm6 | andymccr: do you have the ETA of Ocata 15.0.1 ? | 10:24 |
pjm6 | or you know if it's possible to update only nova (to solve that quota problem) | 10:25 |
admin0 | morning pjm6 | 10:33 |
odyssey4me | pjm6 15.1.0 will release next week if I remember correctly | 10:34 |
*** manheim has quit IRC | 10:35 | |
*** Jack_Iv has joined #openstack-ansible | 10:35 | |
pjm6 | admin0: odyssey4me o/ | 10:49 |
pjm6 | thank you odyssey4me :) | 10:49 |
*** rabel has joined #openstack-ansible | 10:49 | |
*** esberglu has joined #openstack-ansible | 11:05 | |
*** esberglu has quit IRC | 11:10 | |
*** manheim has joined #openstack-ansible | 11:21 | |
*** stuartgr has joined #openstack-ansible | 11:23 | |
*** tonytan4ever has joined #openstack-ansible | 11:28 | |
*** retreved has joined #openstack-ansible | 11:30 | |
*** mpotdar has quit IRC | 11:36 | |
*** Jack_Iv has quit IRC | 11:41 | |
*** Jack_Iv has joined #openstack-ansible | 11:41 | |
pjm6 | with a server of 32GB, running XenServer it's possible to make a "good" testbed with OSA? using like 3 controller nodes, 1 log node and 1 compute node | 11:45 |
pjm6 | probably the recommended were 64GB !? | 11:45 |
*** Jack_Iv has quit IRC | 11:45 | |
pjm6 | i'm trying to figure out what's the best way to make a testbed for OSA deployment | 11:46 |
pjm6 | or it's more preferable to have bare metals (even if they had lower hardware requirements) | 11:54 |
*** Jack_Iv has joined #openstack-ansible | 11:58 | |
*** esberglu has joined #openstack-ansible | 11:59 | |
admin0 | pjm6: what tests do you want to do ? | 12:02 |
*** esberglu has quit IRC | 12:04 | |
*** askb has quit IRC | 12:05 | |
*** karimb has quit IRC | 12:06 | |
*** shardy is now known as shardy_lunch | 12:08 | |
pjm6 | admin0: testing upgrades and see if the minimal things works | 12:09 |
mgariepy | FAILURE in 5h 00m 48s that's a long time to run | 12:09 |
pjm6 | like networking, create VMs, etc | 12:09 |
pjm6 | just to test if a upgrade wents good | 12:09 |
admin0 | pjm6: i use this , which I documented | 12:09 |
admin0 | http://www.openstackfaq.com/openstack-dev-server-setup-ubuntu/ | 12:09 |
admin0 | for every cluster i have, I also have one server where the same ip/network/setup is identical .. so once it passes this closed system, its applied in the real one .. | 12:10 |
admin0 | this acts as acceptance environment | 12:10 |
pjm6 | yes that's something i want | 12:10 |
admin0 | and because those are multiple VMS but under a vyos router, i can have real ips inside this .. so zero changes to playbooks | 12:11 |
*** markvoelker has joined #openstack-ansible | 12:13 | |
*** sanfern has joined #openstack-ansible | 12:13 | |
admin0 | in smaller ones, i have min of 4 vms .. vyos, controller, storage, compute .. bigger ones have the full virtual setup, including ceph | 12:13 |
*** openstackgerrit has joined #openstack-ansible | 12:13 | |
openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-tests master: Stop removing epel https://review.openstack.org/449105 | 12:13 |
mgariepy | odyssey4me, do you mind if I restore : https://review.openstack.org/#/c/443108/ and then backport https://review.openstack.org/44910 in ocata, just to sync up both branches. | 12:16 |
*** markvoelker has quit IRC | 12:17 | |
*** mrtenio has quit IRC | 12:18 | |
mgariepy | i would also need another vote if a core have time : https://review.openstack.org/#/c/448756/ | 12:27 |
mgariepy | it would be needed in ocata as well. | 12:27 |
evrardjp | mgariepy: could you ping me them all? I've kinda missed what's happening on the chan | 12:28 |
evrardjp | is there a topic? | 12:28 |
*** tonytan4ever has quit IRC | 12:28 | |
mgariepy | evrardjp, there is only a few patches ;) https://review.openstack.org/#/c/448756/ | 12:29 |
mgariepy | and i would like to restore https://review.openstack.org/#/c/443108/ to sync up ocata with master for the epel stuff. | 12:29 |
evrardjp | so your goal is RDO release pkg and epel removed by default for nodepool | 12:30 |
evrardjp | because | 12:30 |
evrardjp | https://review.openstack.org/#/c/449105/ | 12:30 |
evrardjp | shouldn't we just do a combined backport? | 12:30 |
mgariepy | ok i'll do a combined backport. | 12:31 |
evrardjp | sorry it's because I'm easily lost :) | 12:31 |
mgariepy | haha well it's not that bad ;) | 12:31 |
mgariepy | I don't have much time for OSA this week. and probably next week won't be much better. | 12:32 |
openstackgerrit | Merged openstack/openstack-ansible-pip_install master: yum-utils is needed to configure repos https://review.openstack.org/448756 | 12:32 |
evrardjp | it happens to everyone, I've been quite away too recently | 12:32 |
openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-tests stable/ocata: Replace EPEL with RDO https://review.openstack.org/443108 | 12:32 |
evrardjp | do we need to take over some of the work for you mgariepy? | 12:33 |
evrardjp | if we can | 12:33 |
*** tonytan4ever has joined #openstack-ansible | 12:33 | |
mgariepy | it's kinda close since : https://review.openstack.org/#/c/448499/ passed :D | 12:34 |
mgariepy | i would like to cleanup as much as possible epel stuff. | 12:34 |
*** woodard_ has quit IRC | 12:35 | |
evrardjp | well it makes sense, it's been a while we've been discussing this | 12:35 |
evrardjp | :p | 12:35 |
evrardjp | thanks for the work there btw | 12:35 |
evrardjp | sorry for not stepping up | 12:35 |
*** woodard has joined #openstack-ansible | 12:35 | |
evrardjp | I leave you to it because I know RDO is in good hands with you :p | 12:36 |
evrardjp | (because EPEL can't be in good hands of everyone :p) | 12:36 |
evrardjp | #terriblejokepoorlyexecuted | 12:36 |
mgariepy | lol | 12:37 |
openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-tests stable/ocata: Install RDO, stop removing EPEL. https://review.openstack.org/443108 | 12:38 |
mgariepy | ^^ like this :D | 12:38 |
openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-tests stable/ocata: Install RDO, stop removing EPEL. https://review.openstack.org/443108 | 12:40 |
*** qiliang27 has quit IRC | 12:41 | |
*** Andrew_jedi has joined #openstack-ansible | 12:43 | |
*** markvoelker has joined #openstack-ansible | 12:46 | |
*** klamath has joined #openstack-ansible | 12:46 | |
*** klamath has quit IRC | 12:47 | |
*** klamath has joined #openstack-ansible | 12:47 | |
*** schwicht has joined #openstack-ansible | 12:52 | |
*** h16mara_ is now known as hachi | 12:53 | |
*** DimGR has quit IRC | 12:54 | |
*** shardy_lunch is now known as shardy | 12:57 | |
*** udesale has joined #openstack-ansible | 13:01 | |
hachi | Hi, | 13:04 |
hachi | I have a SSL problem within HAProxy | 13:05 |
hachi | I added ceph object storage to the existing ceph related things but RGW only responds to the request coming from the internal VIP. The RGW through the external VIP is not working, I have this error message in haproxy.log: | 13:06 |
hachi | Mar 23 12:48:59 localhost haproxy[39014]: 10.1.4.10:37290 [23/Mar/2017:12:48:59.400] ceph-front-1/1: SSL handshake failure | 13:07 |
hachi | Any idea how to fix this ? | 13:07 |
*** esberglu has joined #openstack-ansible | 13:15 | |
fdegir | all the attempts I made to deploy using openstack-ansible from master failed | 13:16 |
fdegir | I appreciate if someone can pass me a specific version that works | 13:17 |
fdegir | trying the ones with sha bumps but they fail too | 13:17 |
*** jwitko has joined #openstack-ansible | 13:18 | |
*** esberglu has quit IRC | 13:19 | |
*** gouthamr has joined #openstack-ansible | 13:20 | |
Andrew_jedi | fdegir: Have you tried the stable/ocata branch ? | 13:26 |
fdegir | Andrew_jedi: that works | 13:27 |
fdegir | Andrew_jedi: but I want master :) | 13:27 |
ildikov | Andrew_jedi: the idea here would be to deploy master as part of a bigger stack | 13:27 |
Andrew_jedi | ildikov: ohhhh okay! | 13:28 |
*** kstev has joined #openstack-ansible | 13:30 | |
ildikov | Andrew_jedi: there's an initiative in OPNFV (another open source community focusing on NFV) to deploy a full stack from the latest master of open source components, where OpenStack provides the cloud infrastructure | 13:30 |
ildikov | Andrew_jedi: this would help end-to-end testing of new features along with the integration of the components | 13:30 |
odyssey4me | mgariepy shouldn't we ensure that EPEL is actually added? | 13:31 |
odyssey4me | I know it's already there in the gating as infra puts it there, but perhaps we should check if it's there and add it if it is not? | 13:31 |
Andrew_jedi | ildikov: sounds really great ... | 13:31 |
ildikov | Andrew_jedi: so we are looking for ways to deploy OpenStack from master | 13:31 |
odyssey4me | fdegir master is broken right now due to some issues with the upstream service repositories | 13:32 |
odyssey4me | the fix is in progress, but we're stuck waiting for them | 13:32 |
ildikov | Andrew_jedi: that would help much with feedback as OPNFV deploys OpenStack from stable, but if we start testing features at that point than we are loosing time | 13:32 |
sc` | odyssey4me: i may have accidentally gotten a centos Vagrantfile functioning | 13:32 |
*** DimGR has joined #openstack-ansible | 13:32 | |
fdegir | odyssey4me: any earlier version you might know: | 13:32 |
fdegir | ? | 13:32 |
odyssey4me | sc` oh really? I was about to push a patch with a working change to the vagrantfile | 13:32 |
sc` | osa still fails because of systemd-resolve not existing | 13:32 |
odyssey4me | fdegir yes, use a stable branch | 13:32 |
odyssey4me | it seems to work fine for me so far | 13:33 |
fdegir | odyssey4me: yes, that works | 13:33 |
odyssey4me | but I'm not all that far along | 13:33 |
fdegir | in case if anyone is interested: https://build.opnfv.org/ci/job/xci-deploy-virtual-xenial-daily-ocata/20/consoleFull | 13:33 |
fdegir | but we are after master... | 13:33 |
odyssey4me | fdegir alternatively, see if you can figure out the problem that's causing tempest to fail | 13:33 |
sc` | odyssey4me: i probably went about it a different way :D | 13:33 |
fdegir | odyssey4me: it is not tempest failure as I excluded tempest from the playbook | 13:34 |
odyssey4me | sc` lemme push this patch up, then perhaps you can review? | 13:34 |
sc` | odyssey4me: for sure | 13:34 |
openstackgerrit | Ravi Kumar Boyapati proposed openstack/openstack-ansible-lxc_hosts master: Add lxc_net_manage_iptables variable https://review.openstack.org/449149 | 13:34 |
fdegir | odyssey4me: keystone error: https://build.opnfv.org/ci/job/xci-deploy-virtual-xenial-daily-master/3/consoleFull | 13:35 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Update Vagrantfile to include CentOS7 https://review.openstack.org/449150 | 13:35 |
odyssey4me | fdegir so have you done any debugging to figure out the cause? | 13:36 |
fdegir | I'm looking | 13:36 |
fdegir | but before going further, I jsut wanted to check if anyone has a known good version which I could try | 13:37 |
odyssey4me | sc` boo, it failed for me too - but it's not a vagrantfile problem, rather it's some missing dep somewhere as you say | 13:37 |
*** cathrichardson has joined #openstack-ansible | 13:37 | |
sc` | systemd-resolve comes in systemd-230 i think | 13:37 |
sc` | c7 has 219 | 13:37 |
sc` | i chased the ubuntu side of things last night | 13:37 |
sc` | ~surprise!~ | 13:38 |
*** esberglu has joined #openstack-ansible | 13:38 | |
odyssey4me | sc` ah, you're tracking a red herring I think | 13:39 |
odyssey4me | anything after exit_fail in the dialog is diagnostic info and has nothing to do with the failure itself | 13:39 |
mgariepy | odyssey4me, on gates epel is installed, and it will be added by either pip_install or openstack_hosts | 13:39 |
odyssey4me | mgariepy ok then why do we do https://review.openstack.org/443108 when it's already done by other roles? | 13:40 |
odyssey4me | unless we're not always executing the other roles? | 13:40 |
*** cathrich_ has joined #openstack-ansible | 13:41 | |
odyssey4me | sc` I think you might be seeing the output of https://github.com/openstack/openstack-ansible/blob/master/scripts/scripts-library.sh#L191 and thinking that's where the failure is | 13:41 |
*** cathrichardson has quit IRC | 13:41 | |
odyssey4me | lemme pastebin my bits and show you where the real error is | 13:41 |
odyssey4me | sc` this is the last portion of my failure output: http://pastebin.com/9t2jVHBi | 13:42 |
sc` | could be. i've been working with osa for but a short time | 13:42 |
odyssey4me | line 39 onwards is all diagnostic, where we try to output a bunch of stuff to help figure out what went wrong | 13:42 |
sc` | oh! that one was dealt with by my vagrantfile | 13:42 |
odyssey4me | line 3 is the thing that actually failed | 13:43 |
cloudnull | mornings | 13:43 |
mgariepy | morning cloudnull | 13:43 |
cloudnull | hows everything on this fine day ? | 13:44 |
odyssey4me | sc` blast, I see there's no growroot in the centos7 image :/ | 13:44 |
sc` | odyssey4me: :D | 13:44 |
sc` | odyssey4me: i have that working | 13:44 |
mgariepy | odyssey4me, i want to sync up tests on both master and ocata for the epel stuff. | 13:44 |
odyssey4me | sc` can you paste your vagrantfile? | 13:44 |
sc` | let me dedup the unnecessary comments | 13:44 |
*** agrebennikov_ has joined #openstack-ansible | 13:45 | |
sc` | you know what... it's not on this machine | 13:45 |
sc` | it's on another workstation | 13:45 |
*** manheim has quit IRC | 13:45 | |
sc` | i'll have to get that later this morning | 13:45 |
sc` | tl;dr create a second disk, then shell out to a helper script that shoves it in the lvm and extends it | 13:46 |
*** manheim has joined #openstack-ansible | 13:46 | |
odyssey4me | sc` yeah ok, that makes sense | 13:47 |
odyssey4me | lemme adjust and try again | 13:47 |
*** fguillot has joined #openstack-ansible | 13:50 | |
sc` | it's xfs, so no resize2fs | 13:51 |
*** NikhilS has quit IRC | 13:59 | |
*** schwicht has quit IRC | 13:59 | |
*** DimGR2 has joined #openstack-ansible | 13:59 | |
*** DimGR has quit IRC | 14:00 | |
*** DimGR2 is now known as DimGR | 14:00 | |
*** cjloader_ has joined #openstack-ansible | 14:01 | |
*** phalmos has joined #openstack-ansible | 14:05 | |
*** rabel_b1 has joined #openstack-ansible | 14:05 | |
*** cjloader_ has quit IRC | 14:06 | |
*** Jack_Iv has quit IRC | 14:07 | |
*** jamesdenton has joined #openstack-ansible | 14:07 | |
*** Jack_Iv has joined #openstack-ansible | 14:07 | |
*** lucasxu has joined #openstack-ansible | 14:08 | |
*** rabel has quit IRC | 14:08 | |
*** acormier has joined #openstack-ansible | 14:11 | |
*** Jack_Iv has quit IRC | 14:12 | |
*** karimb has joined #openstack-ansible | 14:13 | |
*** Jack_Iv has joined #openstack-ansible | 14:18 | |
*** Jack_Iv has quit IRC | 14:24 | |
*** fguillot has quit IRC | 14:25 | |
*** shashank_t_ has joined #openstack-ansible | 14:28 | |
*** Andrew_jedi has quit IRC | 14:34 | |
*** galstrom_zzz is now known as galstrom | 14:34 | |
*** shashank_t_ has quit IRC | 14:36 | |
*** shashank_t_ has joined #openstack-ansible | 14:36 | |
openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible master: Restructure lib directory to install via pip https://review.openstack.org/418076 | 14:39 |
*** Andrew_jedi has joined #openstack-ansible | 14:41 | |
*** rgogunskiy has quit IRC | 14:41 | |
*** fandi has joined #openstack-ansible | 14:42 | |
*** gouthamr has quit IRC | 14:44 | |
*** phalmos_ has joined #openstack-ansible | 14:45 | |
*** phalmos has quit IRC | 14:48 | |
*** shashank_t_ has quit IRC | 14:49 | |
*** schwicht has joined #openstack-ansible | 14:50 | |
*** vnogin has quit IRC | 14:51 | |
*** vnogin has joined #openstack-ansible | 14:52 | |
*** Andrew_jedi has quit IRC | 14:53 | |
*** manheim has quit IRC | 14:54 | |
*** fandi has quit IRC | 14:58 | |
openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-memcached_server master: Add SUSE support https://review.openstack.org/449183 | 14:59 |
*** DimGR2 has joined #openstack-ansible | 14:59 | |
*** DimGR has quit IRC | 15:00 | |
*** cathrich_ is now known as cathrichardson | 15:00 | |
*** marst has joined #openstack-ansible | 15:03 | |
*** adrian_otto has joined #openstack-ansible | 15:03 | |
*** DimGR2 has quit IRC | 15:03 | |
*** Andrew_jedi has joined #openstack-ansible | 15:09 | |
openstackgerrit | German Eichberger proposed openstack/openstack-ansible-os_octavia master: Adds iptables rules to protect octavia server container https://review.openstack.org/447151 | 15:10 |
*** weezS has quit IRC | 15:18 | |
*** muxdaemon has joined #openstack-ansible | 15:18 | |
*** shashank_t_ has joined #openstack-ansible | 15:19 | |
*** DanyC has joined #openstack-ansible | 15:22 | |
*** woodard has quit IRC | 15:24 | |
*** manheim has joined #openstack-ansible | 15:25 | |
openstackgerrit | Logan V proposed openstack/openstack-ansible master: [DOCS] Add checklist item for ceph_stable_release https://review.openstack.org/449195 | 15:30 |
*** woodard_ has joined #openstack-ansible | 15:36 | |
*** jascott1 has joined #openstack-ansible | 15:37 | |
*** manheim has quit IRC | 15:38 | |
*** electrofelix has quit IRC | 15:38 | |
*** schwicht has quit IRC | 15:39 | |
*** mpotdar has joined #openstack-ansible | 15:40 | |
*** electrofelix has joined #openstack-ansible | 15:41 | |
*** Spokozord has joined #openstack-ansible | 15:42 | |
Spokozord | Hi! Pip installation is breaking everything! Halp! :L | 15:42 |
*** weezS has joined #openstack-ansible | 15:46 | |
*** shasha_t_ has joined #openstack-ansible | 15:48 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Update Vagrantfile to include CentOS7 https://review.openstack.org/449150 | 15:49 |
*** karimb has quit IRC | 15:50 | |
odyssey4me | sc` ^ updated to include resizing the root LV | 15:50 |
*** shashank_t_ has quit IRC | 15:52 | |
odyssey4me | fdegir a while ago I suggested to yolanda that you guys implement a pin for the roles in your implementation which gets updated once in a while so that the deployment will be more stable, is that done yet? | 15:52 |
*** karimb has joined #openstack-ansible | 15:53 | |
odyssey4me | my suggestion was that the role pins in ansible-role-requirements get updated every time anything changes in https://github.com/openstack/openstack-ansible/tree/master/playbooks/defaults/repo_packages | 15:53 |
*** foutatoro has joined #openstack-ansible | 15:54 | |
fdegir | odyssey4me: right | 15:54 |
fdegir | odyssey4me: if we can catch a working version, we will go that way | 15:54 |
*** shashank_t_ has joined #openstack-ansible | 15:54 | |
fdegir | hwoarang: yolanda: ^ | 15:55 |
*** rmelero has joined #openstack-ansible | 15:55 | |
fdegir | odyssey4me: you also mentioned you test stuff | 15:56 |
fdegir | odyssey4me: did you mean usual openstack ci | 15:56 |
sc` | odyssey4me: lgtm. doesn't work on mac, but linux doesn't have a problem with it i think | 15:56 |
fdegir | odyssey4me: or you have another ci system where you test stuff and then you bump openstack component versions? | 15:57 |
*** david-lyle_ is now known as david-lyle | 15:58 | |
odyssey4me | sc` I'm using it on a Mac | 15:58 |
odyssey4me | sc` I have a Macbook Pro and the execution is going through now - busy with the security role implementation | 15:59 |
sc` | odyssey4me: peculiar. i seem to recall the controller needing to be "IDE" instead of "IDE Controller". i'll check it out in a bit | 15:59 |
odyssey4me | sc` I'm using recent versions of VirtualBox and vagrant, if that helps | 15:59 |
* sc` => commute(); | 15:59 | |
*** jmckind has joined #openstack-ansible | 16:00 | |
evrardjp | cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, mhayden, scarlisle, luckyinva, ntt, javeriak, automagically, spotz, vdo, jmccrory, alextricity25, jasondotstar, admin0, michaelgugino, ametts, v1k0d3n, severion, bgm | 16:02 |
evrardjp | ccollum, darrenc, JRobinson__, asettle, colinmcnamara, thorst, adreznec, eil397, qwang,nishpatwa_, cathrichardson, drifterza, sc68cal | 16:02 |
evrardjp | community meeting in #openstack-meeting-4 | 16:03 |
*** foutatoro_ has joined #openstack-ansible | 16:03 | |
*** foutatoro has quit IRC | 16:03 | |
*** foutatoro has joined #openstack-ansible | 16:03 | |
*** Spokozord has quit IRC | 16:04 | |
*** DanyC has quit IRC | 16:05 | |
*** bgmccollum_ is now known as bgmccollum | 16:07 | |
foutatoro | Hi , I would like to create 2 flat providers network for OS. how to configure the openstack_user_config to do that | 16:09 |
foutatoro | I tried this config http://paste.openstack.org/show/603953/ but I doesn't work | 16:09 |
xdfil | foutatoro: I think you will need another bridge defined if you want a second flat network | 16:10 |
*** Andrew_jedi has quit IRC | 16:11 | |
xdfil | you have both flat networks on the same bridge | 16:11 |
*** udesale has quit IRC | 16:12 | |
foutatoro | xdfil: I define a new veth (eth15) to avoid confite with the other flat net | 16:14 |
xdfil | foutatoro: on your host you are adding which phisical interface to br-vlan? you can use a vlan tag off that same device like eth0.15 or whatever vlan makes sense and then ad that interface to the br-public bridge | 16:15 |
*** manheim has joined #openstack-ansible | 16:16 | |
xdfil | use a vlan outside the range: "1:1" | 16:16 |
xdfil | do you have VLAN support on your network? | 16:16 |
*** acormier has quit IRC | 16:19 | |
*** jlockwood has joined #openstack-ansible | 16:19 | |
sc` | odyssey4me: here's how i did it - like i said, a different way than you did http://paste.openstack.org/show/603958/ | 16:20 |
sc` | virtualbox 5.1.18 and vagrant 1.9.2 | 16:20 |
foutatoro | xdfil: I don't have vlan support on my net but I will try your suggestion to add new bridge | 16:21 |
foutatoro | xdfil: so I wonder if we are already in prod and we want to add one more provider net. how could we do that ? | 16:23 |
*** jamesden_ has joined #openstack-ansible | 16:23 | |
sc` | odyssey4me: with your Vagrantfile, i get VBoxManage: error: Could not find a controller named 'IDE Controller' | 16:27 |
sc` | odyssey4me: in newer releases, the name gets an abbreviation. the Controller word gets dropped | 16:28 |
*** jamesden_ has quit IRC | 16:28 | |
*** jamesden_ has joined #openstack-ansible | 16:29 | |
*** tonytan4ever has quit IRC | 16:31 | |
*** jamesden_ has quit IRC | 16:31 | |
*** tonytan4ever has joined #openstack-ansible | 16:31 | |
sc` | odyssey4me: other than that, lgtm | 16:32 |
*** lucasxu has quit IRC | 16:32 | |
*** SerenaFeng has joined #openstack-ansible | 16:32 | |
*** Andrew_jedi has joined #openstack-ansible | 16:33 | |
*** SerenaFeng has quit IRC | 16:33 | |
jamesdenton | Hey jmccrory - have a sec to look at https://github.com/openstack/openstack-ansible-openstack_openrc/commit/6848f7d1b7312b296beec5deba85399937af11dc ? | 16:39 |
jmccrory | jamesdenton sure, what's up? | 16:40 |
jamesdenton | Running the os-neutron-install playbook, and I'm hitting an error during the [os_neutron : Ensure neutron user] task. Turns out, by removing the single quotes from the openrc.j2 file, authentication is successful | 16:42 |
jamesdenton | https://gist.github.com/busterswt/538558eda8a6cdf661063848a57a16e5 | 16:42 |
jamesdenton | https://gist.github.com/busterswt/538558eda8a6cdf661063848a57a16e5#file-gistfile1-txt-L941 | 16:42 |
jamesdenton | I confirmed in the other containers running older master branch, there are no quotes around the password | 16:43 |
jmccrory | there was a corresponding change needed in the plugins repo https://review.openstack.org/#/c/433991/ | 16:43 |
jamesdenton | oye, ok | 16:44 |
jamesdenton | thanks! | 16:44 |
jmccrory | no prob | 16:46 |
*** adrian_otto1 has joined #openstack-ansible | 16:46 | |
agrebennikov_ | folks, please, somebody vote for https://review.openstack.org/#/c/448743/2 and https://review.openstack.org/#/c/425997/29 | 16:48 |
*** acormier has joined #openstack-ansible | 16:48 | |
*** foutatoro has quit IRC | 16:49 | |
*** adrian_otto has quit IRC | 16:49 | |
evrardjp | cloudnull: due to the light shed by jmccrory on 2.3 connection plugins, maybe we should spend some time to make our lxc connection plugin more robust for 2.3 or maybe spend time on alternatives... what do you think? | 16:59 |
evrardjp | maybe refactor the "super" of the existing 2.3 connection plugin | 16:59 |
evrardjp | class | 16:59 |
cloudnull | evrardjp: jmccrory: looking at the error it looks like the issue is contained within only the ssh plugin? is that right? are seeing this in the lxc_container module too ? | 17:00 |
*** DimGR has joined #openstack-ansible | 17:02 | |
evrardjp | I didn't have a look at it, merely just refreshing memories about what 2.3 changed. | 17:02 |
*** schwicht has joined #openstack-ansible | 17:06 | |
cloudnull | I'll have a look at it to see what's what. | 17:08 |
*** pbandark has joined #openstack-ansible | 17:12 | |
*** pcaruana has quit IRC | 17:14 | |
openstackgerrit | Merged openstack/openstack-ansible-os_rally stable/newton: Update missing database exception text https://review.openstack.org/445334 | 17:17 |
jmccrory | cloudnull: oh sorry , right. it's just in the OSA ssh connection plugin | 17:17 |
*** lostrhino has joined #openstack-ansible | 17:21 | |
luzC | odyssey4me just a heads up, earlier today we had some issues with our BME environment, as soon as we have it back online we will try the upgrade flow (with galera patch) N to O... | 17:24 |
cloudnull | ++ thanks luzC | 17:24 |
luzC | cloudnull any time ;) | 17:26 |
lostrhino | I was wondering if someone could help me - I’ve having difficulty with the openstack ansible play setup_hosts, we have our own apt-cache repo, I’ve been trying to test. I’ve added the line user_external_repo_keys_list: | 17:26 |
lostrhino | url: https://fqdn/my-key.asc | 17:26 |
lostrhino | to the user_variables.yml file | 17:26 |
lostrhino | and I have been unsuccessful - is there anyone who can give me some guidance | 17:26 |
lostrhino | I feel like I’ve tried everything | 17:26 |
lostrhino | to no avail | 17:26 |
*** manheim has quit IRC | 17:27 | |
lostrhino | I get the following error failed: [server1] (item=url) => {"failed": true, "item": "url", "msg": "needed a URL but was not specified"} | 17:28 |
lostrhino | even though I’ve specified it | 17:28 |
*** lucasxu has joined #openstack-ansible | 17:32 | |
odyssey4me | lostrhino so, unfortunately evrardjp has not updated the defaults documentation to match the other change - and we didn't pick it up in review | 17:33 |
evrardjp | try with another - | 17:33 |
cloudnull | lostrhino: do you know which task is failing? | 17:33 |
odyssey4me | change 'url: https://fqdn/my-key.asc' to '- url: https://fqdn/my-key.asc' and it should work | 17:34 |
evrardjp | cloudnull: I think we need a list, and it looks like it's only a dict | 17:34 |
lostrhino | yes - TASK [pip_install : Install external repo key manually (apt)] ****************** | 17:34 |
evrardjp | odyssey4me: where is the docs? | 17:34 |
odyssey4me | shame on us for not picking that up in review :( | 17:34 |
odyssey4me | evrardjp make fix please :) | 17:34 |
evrardjp | odyssey4me: nobody got the time to review this | 17:34 |
evrardjp | it was in a hurry during ptg IIRC | 17:35 |
lostrhino | if that is it ….. oh man…. | 17:35 |
evrardjp | it's written list on it! | 17:35 |
lostrhino | I’m runnin the lay again with another change i madetesting - I’l update in 3 minutes | 17:35 |
evrardjp | not my fault! | 17:35 |
lostrhino | play* | 17:35 |
evrardjp | :p | 17:35 |
odyssey4me | lostrhino if you're using user_external_repos_list then also make sure it's a list | 17:35 |
lostrhino | I used a asc file since I got a deprecation warning and needed it… I thought (key word thought) it would be ok | 17:36 |
odyssey4me | mgariepy mhayden sc` so the good news is that https://review.openstack.org/449150 appears to be working quite well, my deployment is busy completing the designate service which is quite far down the list | 17:38 |
sc` | odyssey4me: mine is on rabbit | 17:38 |
sc` | i had to adapt it to newer virtualbox, but it's humming along | 17:39 |
*** dxiri has joined #openstack-ansible | 17:39 | |
evrardjp | for a rabbit it seems slower. | 17:39 |
evrardjp | #terriblejoke | 17:39 |
odyssey4me | sc` you have a newer virtualbox and something changed? | 17:39 |
sc` | odyssey4me: i'm running 5.1.18 | 17:39 |
sc` | "IDE Controller" changes to "IDE", ditto for other controllers | 17:40 |
odyssey4me | the thing should probably do more detection before adjusting things | 17:40 |
odyssey4me | ie it should detect the default controller, then re-use that later | 17:40 |
sc` | ya | 17:40 |
mgariepy | odyssey4me, cool :) | 17:40 |
odyssey4me | if you could work up a patch for that it'd be awesome :) | 17:41 |
odyssey4me | I guess it'd be nice to also be able to give it some CLI options, so you could test alternative scenarios or actions | 17:41 |
odyssey4me | scenario = aio/whatever, action = deploy/upgrade | 17:42 |
odyssey4me | that would then mean you could do exactly what we do in gate testing | 17:42 |
lostrhino | so who do I buy a beer for in boston and whom do I yell at for not updating the docs??? | 17:42 |
odyssey4me | lostrhino :) not sure I'll be in boston, but I'm sure that andymccr will take the beer ;) | 17:43 |
odyssey4me | you can yell at us all for not updating docs, because that's all our responsibilities to check before approving a patch | 17:43 |
sc` | odyssey4me: looks like maybe a bug in 5.1 | 17:43 |
sc` | https://www.virtualbox.org/ticket/16387 | 17:43 |
sc` | if it's not intentional, it's a regression, natch | 17:44 |
odyssey4me | sc` if you don't get to it first then I'll push a patch some time to do the detection and consumption | 17:45 |
*** messy has quit IRC | 17:45 | |
odyssey4me | that'll make it work regardless | 17:45 |
odyssey4me | I'd like to also add optional caching through a file that's durable and re-usable across multiple VM's | 17:45 |
sc` | word. given the newness of that ticket, i'd be willing to believe it be fixed in 5.1 | 17:45 |
cloudnull | jmccrory: evrardjp: this is what breaks us w/ 2.3 https://github.com/ansible/ansible/blob/devel/lib/ansible/plugins/connection/ssh.py#L756-L782 | 17:45 |
odyssey4me | it'll help cut down the time to deploy | 17:45 |
*** retreved has quit IRC | 17:46 | |
cloudnull | its running the "piped" method first | 17:46 |
*** karimb has quit IRC | 17:46 | |
*** retreved has joined #openstack-ansible | 17:46 | |
cloudnull | before it was https://github.com/ansible/ansible/blob/v2.2.1.0-1/lib/ansible/plugins/connection/ssh.py#L626-L635 | 17:46 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-pip_install master: Fix documentation for external repos and keys https://review.openstack.org/449250 | 17:46 |
odyssey4me | lostrhino ^ your review of that would be nice :) | 17:47 |
lostrhino | how would I go about that - I’m new to reviewing …. or you just kidding? | 17:48 |
evrardjp | lostrhino: no I think he was serious :) | 17:49 |
evrardjp | it's fine, we are not monsters | 17:49 |
odyssey4me | lostrhino if you've registered and have the ability to login to gerrit, then go to https://review.openstack.org/449250 | 17:49 |
evrardjp | it's basically saying what you think of the change :) | 17:49 |
*** vnogin has quit IRC | 17:50 | |
odyssey4me | lostrhino https://docs.openstack.org/infra/manual/developers.html may be handy | 17:50 |
evrardjp | cloudnull: what is piped? | 17:50 |
cloudnull | its new | 17:50 |
evrardjp | Thanks captain | 17:50 |
evrardjp | p | 17:51 |
evrardjp | :p * | 17:51 |
cloudnull | lol | 17:51 |
*** deadnull has joined #openstack-ansible | 17:51 | |
evrardjp | I mean maybe we should improve our support of this new shiny thing. | 17:51 |
cloudnull | or kill it. | 17:51 |
cloudnull | :) | 17:51 |
evrardjp | "runs faster on my machine" | 17:51 |
evrardjp | haha | 17:51 |
*** adrian_otto1 has quit IRC | 17:52 | |
odyssey4me | mgariepy sc` unfortunately it failed to download the cirros image when executing the tempest role - retrying now | 17:54 |
jmccrory | cloudnull evrardjp this change too https://github.com/ansible/ansible/commit/1fe67f9f436595003f7951dd88159731e6d82498, OSA plugin is overriding _exec_command, it's removed there | 17:55 |
*** electrofelix has left #openstack-ansible | 17:55 | |
evrardjp | I think it's time for a connection plugin refactor | 17:55 |
lostrhino | +1’d | 17:57 |
odyssey4me | lostrhino :) thanks | 17:57 |
*** adrian_otto has joined #openstack-ansible | 17:57 | |
sc` | odyssey4me: womp womp. i'm not to that point yet | 17:58 |
jrosser_ | i liked that proposed change for the container ready check to use ping module | 17:58 |
evrardjp | woot! | 17:58 |
jrosser_ | currently it doesnt work through an ssh bastion | 17:58 |
*** messy has joined #openstack-ansible | 17:58 | |
odyssey4me | jrosser_ the current discussion around refactoring the connection plugins also relates to the similar thing | 17:58 |
jrosser_ | which imho would be good thing to aim at | 17:58 |
odyssey4me | sc` mgariepy well, I had to retry 3 times but it eventually pulled it down | 17:59 |
mgariepy | weird | 17:59 |
openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-pip_install stable/ocata: yum-utils is needed to configure repos https://review.openstack.org/449256 | 18:01 |
*** MasterOfBugs has joined #openstack-ansible | 18:02 | |
odyssey4me | mgariepy sc` it looks like we could do with some retries here: https://github.com/openstack/openstack-ansible-os_tempest/blob/master/tasks/tempest_post_install.yml#L46 | 18:03 |
*** Andrew_jedi has quit IRC | 18:03 | |
odyssey4me | I'll patch it up now | 18:03 |
odyssey4me | so the tempest tests failed, but the same as they do in master right now - so I think that Vagrantfile is good as-is | 18:04 |
odyssey4me | mgariepy I'll add CentOS deployments to the periodics in infra now too FYI | 18:05 |
mgariepy | cool | 18:05 |
mgariepy | the periodics will be ran every day ? | 18:05 |
mgariepy | or week ? | 18:05 |
odyssey4me | every day | 18:05 |
mgariepy | how do i check the result ? | 18:06 |
*** muxdaemon has quit IRC | 18:06 | |
odyssey4me | mgariepy all periodics post their log results here: http://logs.openstack.org/periodic/ | 18:07 |
odyssey4me | you just need to find the job name, then look in there | 18:07 |
odyssey4me | for example: http://logs.openstack.org/periodic/periodic-openstack-ansible-upgrade-aio-master-ubuntu-xenial/ | 18:07 |
jamesdenton | Quick question.. what is used to distribute the fernet keys across containers? | 18:07 |
*** muxdaemon has joined #openstack-ansible | 18:07 | |
odyssey4me | jamesdenton magic! | 18:08 |
lbragstad | s/magic/voodoo magic/ | 18:08 |
*** muxdaemon has quit IRC | 18:08 | |
odyssey4me | jamesdenton actually, when the keys are rotated by the cron job on node 1, they're pushed via rsync to the other nodes | 18:08 |
mgariepy | odyssey4me, cool thanks | 18:08 |
evrardjp | haha voodoo magic. | 18:09 |
evrardjp | It's because you never practiced it. | 18:09 |
lbragstad | https://github.com/openstack/openstack-ansible-os_keystone/blob/master/tasks/keystone_fernet_keys_autorotate.yml | 18:09 |
jamesdenton | so at all times, the contents of /etc/keystone/fernet_keys should be consistent across nodes, right? | 18:09 |
evrardjp | <- voodoo 2 user. | 18:09 |
mgariepy | magic ^^ https://www.youtube.com/watch?v=w0JCw1dWIMA | 18:09 |
jamesdenton | thanks lbragstad - good blog btw! | 18:09 |
odyssey4me | jamesdenton supposed to be, yes | 18:10 |
lbragstad | jamesdenton some more magic lives here https://github.com/openstack/openstack-ansible-os_keystone/blob/master/templates/keystone-fernet-rotate.sh.j2 | 18:10 |
evrardjp | ahah mgariepy love your references. I could add a arrested development one, but then it's becoming... weird! | 18:10 |
lbragstad | jamesdenton thanks! on fernet key rotation? | 18:10 |
jamesdenton | odyssey4me you see where this is going, right? Trying to figure out why they're inconsistent in my env :P | 18:10 |
jamesdenton | yes | 18:10 |
lbragstad | jamesdenton if you find anything inconsistent/wrong let me know ;) | 18:11 |
lbragstad | or confusing | 18:11 |
lbragstad | jamesdenton a more evolved version of those posts is available here - https://docs.openstack.org/admin-guide/identity-fernet-token-faq.html | 18:12 |
jamesdenton | thanks! | 18:12 |
*** lostrhino has left #openstack-ansible | 18:12 | |
*** pbandark has quit IRC | 18:16 | |
*** adrian_otto has quit IRC | 18:16 | |
*** adrian_otto has joined #openstack-ansible | 18:16 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest master: Retry the cirros image download https://review.openstack.org/449260 | 18:20 |
odyssey4me | sc` mgariepy ^ that should help | 18:20 |
cloudnull | evrardjp: jmccrory: looks like changing over to the new execute method seems to fix it | 18:22 |
*** rboyapat has joined #openstack-ansible | 18:22 | |
mgariepy | odyssey4me, delay 5? | 18:22 |
odyssey4me | mgariepy I'm open to suggestion | 18:23 |
mgariepy | 25 sec instead of 50. | 18:23 |
odyssey4me | I took that from lines 27-28 | 18:23 |
cloudnull | evrardjp: jmccrory: http://cdn.pasteraw.com/fqd53i83fuaqiqpltzns7il73g5mwno | 18:24 |
odyssey4me | mgariepy yeah, I'm more a fan of delay 5, retries 12 | 18:24 |
odyssey4me | ie 60 secs total | 18:24 |
cloudnull | and it looks like that methdo will work in 2.1 | 18:24 |
mgariepy | yeah, issues if we start adding retry everywhere and wait 10 sec it quickly add up :) | 18:25 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest master: Retry the cirros image download https://review.openstack.org/449260 | 18:26 |
odyssey4me | mgariepy now 30 secs max | 18:26 |
evrardjp | cloudnull: ahah great. | 18:26 |
evrardjp | we can also use this new connection method | 18:26 |
evrardjp | maybe it's gonna be super fast | 18:26 |
mgariepy | nice | 18:27 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-plugins master: Updated connection plugin for ansible 2.3 support https://review.openstack.org/449263 | 18:27 |
cloudnull | evrardjp: jmccrory: if you have a moment to test this it'd be great to get some validation there. | 18:27 |
evrardjp | I'll test that at home soon | 18:28 |
evrardjp | do we have tests of our connection plugin? | 18:29 |
evrardjp | I am not sure we have | 18:29 |
evrardjp | else manual testing I guess | 18:29 |
odyssey4me | evrardjp every role and the integrated build test it ;) | 18:31 |
cloudnull | evrardjp: what odyssey4me said, | 18:33 |
*** manheim has joined #openstack-ansible | 18:36 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-plugins master: Add Packages file parsing lookup https://review.openstack.org/449267 | 18:36 |
sc` | odyssey4me: woo :D | 18:37 |
*** lucasxu has quit IRC | 18:38 | |
sc` | my build is still going, but i forgot to put it in a tmux so i could observe it from afar | 18:38 |
*** cjloader_ has joined #openstack-ansible | 18:39 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-plugins master: Add list of packages to install lookup https://review.openstack.org/449274 | 18:40 |
evrardjp | odyssey4me: the 2 new lookups have been uploaded ^ | 18:41 |
odyssey4me | mgariepy logan- andymccr FYI https://review.openstack.org/449272 (more periodics) | 18:42 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest master: Retry the cirros image download https://review.openstack.org/449260 | 18:44 |
odyssey4me | evrardjp ^ good catch, thanks | 18:44 |
openstackgerrit | Merged openstack/openstack-ansible master: Update Vagrantfile to include CentOS7 https://review.openstack.org/449150 | 18:45 |
openstackgerrit | Merged openstack/openstack-ansible-os_rally stable/newton: Bring consistency to tags https://review.openstack.org/423244 | 18:46 |
*** stuartgr has quit IRC | 18:50 | |
sc` | odyssey4me: i got lucky and the cirros download seems to have worked on the first try | 18:53 |
*** dxiri has quit IRC | 18:53 | |
openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_container_create master: Use ping module instead of wait_for to check container https://review.openstack.org/386835 | 18:55 |
evrardjp | logan-: interesting, I thought wait_for could wait for a socket open now | 18:55 |
evrardjp | ahah | 18:56 |
evrardjp | it's wait_for_connection | 18:56 |
logan- | oh really | 18:56 |
evrardjp | 2.3 :( | 18:56 |
evrardjp | sorry | 18:56 |
logan- | dang | 18:56 |
evrardjp | http://docs.ansible.com/ansible/wait_for_connection_module.html | 18:56 |
logan- | that's cool | 18:56 |
evrardjp | problem is python dependency on ping | 18:57 |
evrardjp | for hosts that don't have python yet | 18:57 |
evrardjp | (not really a problem for us I think, because the containers are auto bootstrapped with python) | 18:57 |
evrardjp | FYI :p | 18:57 |
logan- | yeah | 18:57 |
*** shashank_t_ has quit IRC | 18:59 | |
odyssey4me | mgariepy can you add your review to https://review.openstack.org/443108 please? | 19:03 |
mgariepy | odyssey4me, done | 19:05 |
*** dxiri has joined #openstack-ansible | 19:08 | |
*** deadnull has quit IRC | 19:09 | |
*** shashank_t_ has joined #openstack-ansible | 19:09 | |
openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-os_cinder master: Adjust cinder init restart times, provide variable https://review.openstack.org/448777 | 19:14 |
*** Jack_Iv has joined #openstack-ansible | 19:14 | |
*** shardy is now known as shardy_afk | 19:18 | |
*** Andrew_jedi has joined #openstack-ansible | 19:19 | |
*** shashank_t_ has quit IRC | 19:20 | |
*** shashank_t_ has joined #openstack-ansible | 19:20 | |
sc` | odyssey4me: is it worth respinning until 449260 is merged? | 19:21 |
*** fxpester has quit IRC | 19:21 | |
odyssey4me | sc` master is busted due to other reasons right now - something's causing tempest to fail | 19:22 |
sc` | fun | 19:22 |
sc` | i don't even want master, just a stable branch :D | 19:23 |
*** shashank_t_ has quit IRC | 19:23 | |
sc` | but, if i'm blocked on testing, then my focus switches back to chef | 19:23 |
odyssey4me | sc` there're a few patches going into ocata right now which will fix ocata after a role sha bump | 19:24 |
odyssey4me | once they merge we can implement that sha bump | 19:24 |
*** adrian_otto has quit IRC | 19:24 | |
sc` | word. sounds like i'll just get in the way | 19:24 |
odyssey4me | at least for now your master-based deploy gives you something to look at | 19:25 |
*** adrian_otto has joined #openstack-ansible | 19:26 | |
*** cmart has joined #openstack-ansible | 19:33 | |
*** adrian_otto has quit IRC | 19:34 | |
openstackgerrit | Merged openstack/openstack-ansible-tests master: Stop removing epel https://review.openstack.org/449105 | 19:44 |
openstackgerrit | Merged openstack/openstack-ansible-tests stable/ocata: Install RDO, stop removing EPEL. https://review.openstack.org/443108 | 19:44 |
odyssey4me | mgariepy should https://review.openstack.org/448756 also be included in openstack_hosts and lxc_hosts ? | 19:46 |
agrebennikov_ | odyssey4me, 2 minutes please for https://review.openstack.org/#/c/448743/2 ;( | 19:47 |
odyssey4me | agrebennikov_ ok, discussing with evrardjp quickly | 19:48 |
agrebennikov_ | sure | 19:48 |
agrebennikov_ | also seems andymccr is not in today | 19:49 |
agrebennikov_ | I just don't know what to do with https://review.openstack.org/#/c/425997/29 | 19:49 |
*** vnogin has joined #openstack-ansible | 19:50 | |
mgariepy | odyssey4me, it's already in lxc_host | 19:50 |
*** cmart has quit IRC | 19:50 | |
*** shashank_t_ has joined #openstack-ansible | 19:51 | |
mgariepy | wouldn't hurt to be in opesntack_host | 19:51 |
mgariepy | i'll patch up. | 19:51 |
odyssey4me | evrardjp https://review.openstack.org/425997 and https://review.openstack.org/448743 need to be reviewed together | 19:53 |
evrardjp | sure | 19:53 |
openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-openstack_hosts master: yum-utils is needed to configure repos https://review.openstack.org/449299 | 19:54 |
openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/ocata: CentOS pkg cleanup https://review.openstack.org/448578 | 19:54 |
*** Andrew_jedi has quit IRC | 19:57 | |
*** kstev has quit IRC | 20:03 | |
*** Jack_Iv has quit IRC | 20:03 | |
*** Jack_Iv has joined #openstack-ansible | 20:04 | |
*** rabel_b1 has quit IRC | 20:04 | |
*** Jack_Iv has quit IRC | 20:04 | |
openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible stable/newton: Bump Keepalived https://review.openstack.org/449305 | 20:07 |
odyssey4me | agrebennikov_ do you mind if we tweak these two patches a bit? | 20:08 |
odyssey4me | two things we'd prefer to change, apologies for the revision (again) but we're working out a pattern which we'd ideally like to replicate to other ssl things | 20:08 |
*** kstev has joined #openstack-ansible | 20:08 | |
odyssey4me | 1 - the cert generation should happen on the deploy node if none are provided | 20:09 |
odyssey4me | 2 - the certs are always distributed using the user_provided vars... no special vars for self signed stuff | 20:09 |
*** manheim has joined #openstack-ansible | 20:10 | |
openstackgerrit | Merged openstack/openstack-ansible-pip_install stable/ocata: yum-utils is needed to configure repos https://review.openstack.org/449256 | 20:12 |
*** askb has joined #openstack-ansible | 20:17 | |
agrebennikov_ | crap... seriously? :) | 20:20 |
rboyapat | core members please review - https://review.openstack.org/#/c/449149/ | 20:20 |
*** cjloader_ has quit IRC | 20:30 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-rabbitmq_server master: Reload systemd-daemon in all cases https://review.openstack.org/440463 | 20:37 |
evrardjp | rboyapat: there is an issue with an extra whitespace | 20:39 |
evrardjp | should I fix it? | 20:39 |
evrardjp | if you fix it, I can adapt my review to a positive one right now. | 20:40 |
*** karimb has joined #openstack-ansible | 20:40 | |
rboyapat | fixing it | 20:41 |
openstackgerrit | Ravi Kumar Boyapati proposed openstack/openstack-ansible-lxc_hosts master: Add lxc_net_manage_iptables variable https://review.openstack.org/449149 | 20:43 |
*** karimb has quit IRC | 20:43 | |
rboyapat | evrardjp: fixing it would you mind reviewing this https://review.openstack.org/#/c/448872/ | 20:43 |
*** jrobinson has joined #openstack-ansible | 20:44 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_nova master: Ensure the components are isolated from the system https://review.openstack.org/449315 | 20:45 |
odyssey4me | rboyapat I think cloudnull and logan- should be reviewing that | 20:45 |
odyssey4me | or jmccrory perhaps | 20:45 |
rboyapat | odyssey4me: ok. | 20:45 |
cloudnull | rboyapat: are we just changing the case of USE_IPTABLES_LOCK ? | 20:47 |
rboyapat | cloudnull: yep. updating the case actually allows to acquire the lock | 20:48 |
rboyapat | as the commands already have upper case USE_IPTABLES_LOCK | 20:49 |
rboyapat | but the variable where it set is in lower case | 20:49 |
cloudnull | evrardjp: odyssey4me: logan-: jmccrory: others-if-interested: https://review.openstack.org/#/c/449315/ -- this is a change i'd like to implement across the board if aproved. if you have a moment it'd be appreciated. | 20:49 |
cloudnull | rboyapat: okiedokie. makes sense . | 20:50 |
*** jwitko_ has joined #openstack-ansible | 20:51 | |
agrebennikov_ | odyssey4me, evrardjp so essentially you want me to not create the dir separately at all, and if user_cert is not specified by the user - just silently create the self-sign bundle in the default place and distribure? | 20:52 |
agrebennikov_ | *distribute | 20:52 |
*** manheim has quit IRC | 20:53 | |
*** jwitko has quit IRC | 20:55 | |
*** jwitko_ has quit IRC | 20:57 | |
*** adrian_otto has joined #openstack-ansible | 20:58 | |
*** jmckind has quit IRC | 20:59 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_nova master: Ensure the components are isolated from the system https://review.openstack.org/449315 | 20:59 |
*** galstrom is now known as galstrom_zzz | 21:01 | |
*** kstev has quit IRC | 21:02 | |
*** galstrom_zzz is now known as galstrom | 21:05 | |
cloudnull | jmccrory: odyssey4me: evrardjp: https://review.openstack.org/#/c/449263/ | 21:07 |
openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts master: yum-utils is needed to configure repos https://review.openstack.org/449299 | 21:08 |
cloudnull | seems to get through the gate, which is executing a playbook. so i suspect the change does not break anything in ansible <2.3 | 21:08 |
*** kstev has joined #openstack-ansible | 21:10 | |
*** retreved has quit IRC | 21:14 | |
lbragstad | o/ | 21:17 |
cloudnull | lbragstad: o/ | 21:18 |
lbragstad | curious if a few folks in here have their operator hats strapped on and would like to take a gander at the upgrade process we've got outlines for https://review.openstack.org/#/c/438701/13 | 21:18 |
lbragstad | outlines* | 21:18 |
cloudnull | new hashing algo ? | 21:18 |
lbragstad | ideally - it's a change that allows keystone to store password hashes that are better | 21:18 |
lbragstad | cloudnull yessir | 21:18 |
cloudnull | interesting. | 21:19 |
* cloudnull reading | 21:19 | |
lbragstad | very | 21:19 |
lbragstad | its a tricky one | 21:19 |
lbragstad | especially when considering rolling upgrades | 21:19 |
*** rboyapat has quit IRC | 21:20 | |
cloudnull | seems like at a min we could just set password_hash_algorithm to keep old pw hashing compatibility on older clouds. Flipping rolling_upgrade_password_hash_compat=true during an upgrade and false after seems like an entry point for pain. could this potentially be an option in a given sync command ? | 21:23 |
* cloudnull just spitballing. | 21:23 | |
openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Fix the lock type variable https://review.openstack.org/448872 | 21:23 |
lbragstad | right | 21:24 |
lbragstad | that's the part that kinda threw me for a loop because at some point there is going to have be a play/recipe to flip that bit | 21:24 |
lbragstad | back to false | 21:24 |
cloudnull | i mean we can make it go in an upgrade script. | 21:25 |
cloudnull | but in my mind it seems like it might be better suited as a cli switch | 21:25 |
cloudnull | others may have different thoughts on that though | 21:25 |
*** galstrom is now known as galstrom_zzz | 21:25 | |
lbragstad | cloudnull cli switch? like a bulk migration? | 21:25 |
cloudnull | yea. i guess that wouldn't do for a rolling upgrade | 21:26 |
lbragstad | currently we store a hash of the password | 21:26 |
*** galstrom_zzz is now known as galstrom | 21:26 | |
lbragstad | its an old hash, but a hash non-the-less | 21:26 |
lbragstad | in order to do a bulk migration, wouldn't we need the original value of the password? | 21:27 |
bgmccollum | was there any resolution on Horizon doing API calls to Glance directly from the browser, and it using the wrong API endpoint? | 21:27 |
bgmccollum | and the fact that you need to manually configure CORS in Glance if i recall correctly | 21:27 |
cloudnull | lbragstad: i believe so . we'd have to decrypt and reencrypt. so just ignore me :) | 21:28 |
cloudnull | bgmccollum: I'm unaware of such things | 21:28 |
cloudnull | is there a bug for that? | 21:28 |
bgmccollum | cloudnull have you tried uploading an image via horizon? | 21:28 |
cloudnull | in ocata? | 21:28 |
*** esberglu has quit IRC | 21:29 | |
bgmccollum | at some point, horizon stopped proxying glance API calls to Glance, and instead sends them directly to glance to upload images | 21:29 |
bgmccollum | i havent tried ocata | 21:29 |
cloudnull | hum. | 21:29 |
bgmccollum | im trying to find the old bug report | 21:29 |
bgmccollum | cloudnull https://bugs.launchpad.net/openstack-ansible/+bug/1639080 | 21:29 |
openstack | Launchpad bug 1639080 in openstack-ansible "Image uploads fail in Horizon if upload mode is set to direct if endpoint set to internal." [Low,Confirmed] | 21:29 |
bgmccollum | looks like it might be fixed in horizon... | 21:30 |
bgmccollum | hmm | 21:30 |
cloudnull | maybe it needs a cherry-pick to newton too ? | 21:30 |
bgmccollum | i need to see what the fix was, cause half was in horizon, half to fix CORS support in glance | 21:31 |
bgmccollum | https://bugs.launchpad.net/openstack-ansible/+bug/1639080/comments/4 | 21:31 |
openstack | Launchpad bug 1639080 in openstack-ansible "Image uploads fail in Horizon if upload mode is set to direct if endpoint set to internal." [Low,Confirmed] | 21:31 |
kencjohnston | Hi OSA team, I work with OSIC and our QA/CI team members ( dankolbrs ) identified this bug in our nightly N->O upgrade job. It is presently unassigned, can you take a look? | 21:31 |
kencjohnston | https://bugs.launchpad.net/openstack-ansible/+bug/1673889 | 21:31 |
openstack | Launchpad bug 1673889 in openstack-ansible "Nova services do not restart on N->O upgrade" [Critical,Confirmed] | 21:31 |
cloudnull | lbragstad: so the ident flag for backward compat hashes during rolling upgrade, would that have to be done for every upgrade ? | 21:32 |
cloudnull | or just the initial one from ocata to pike ? | 21:32 |
cloudnull | kencjohnston: lookin | 21:32 |
lbragstad | cloudnull that's a good question - i think it's there for the ocata -> pike upgrade though | 21:33 |
*** karimb has joined #openstack-ansible | 21:34 | |
*** pbandark has joined #openstack-ansible | 21:34 | |
cloudnull | lbragstad: looking at https://review.openstack.org/#/c/438701/13/keystone/identity/backends/sql_model.py looks like its only there for O>P | 21:37 |
cloudnull | so its not so bad, add the one key for hash compat roll forward to P and remove the compat option | 21:37 |
cloudnull | we can do that within the config_template | 21:38 |
cloudnull | we simply would have to detect an upgrade from O>P | 21:38 |
lbragstad | cloudnull but you can only remove the compat option once all your users have created new passwords with the new hash algo? | 21:38 |
*** lucasxu has joined #openstack-ansible | 21:40 | |
cloudnull | in the RENO -- https://review.openstack.org/#/c/438701/13/releasenotes/notes/bug_1543048_and_1668503-7ead4e15faaab778.yaml -- it seems like it only needs to be their for the initial upgrade. | 21:40 |
cloudnull | I guess we'd have to do a deplpoyment wide password reset ? | 21:40 |
cloudnull | or will the old hash continue to function and the new users (or folks who reset) will have a better hash algo | 21:41 |
cloudnull | ? | 21:41 |
cloudnull | kencjohnston: seems odd. | 21:41 |
cloudnull | going from N>O would create new files and drop new code. which should restart the init scripts (systemd unit files) | 21:41 |
cloudnull | the containers would not necessarily restart. | 21:42 |
cloudnull | in fact we do our best not to restart containers in an effort to maintain uptime. | 21:42 |
lbragstad | cloudnull well - the option would have to be True even if a single person hasn't reset their password | 21:45 |
openstackgerrit | Merged openstack/openstack-ansible-tests stable/ocata: Bump Tempest SHA for stable/ocata https://review.openstack.org/448984 | 21:45 |
cloudnull | kencjohnston: i commented in the bug. | 21:47 |
*** adrian_otto has quit IRC | 21:47 | |
kencjohnston | cloudnull thanks | 21:47 |
cloudnull | if you might be able to get the QE team to forward on the logs from that run or another one I'd be happy to see about digging into this | 21:48 |
cloudnull | lbragstad: so then my next question would be why does that need to be set at all and why does it need special handling? | 21:50 |
cloudnull | seems like that op would have to be there for the life time of any user running an old password ? | 21:51 |
lbragstad | cloudnull good question | 21:51 |
lbragstad | would it be a better/easier approach to just make it so all new passwords are hashed with bcrypt and then incrementally force people to change their passwords? | 21:51 |
cloudnull | it'd be easier on the endusers that for sure. | 21:52 |
lbragstad | cloudnull enduser as in operator, right? | 21:52 |
cloudnull | maybe there needs to be a notice in the reno stating that keystone requires all users of the cloud to reset their password within the PIKE cycle. | 21:52 |
lbragstad | yeah | 21:52 |
lbragstad | cloudnull you should leave that as a comment on the review | 21:53 |
cloudnull | also would this not be a CVE? | 21:53 |
lbragstad | cloudnull i think it was deemed a security hardening measure | 21:54 |
cloudnull | lbragstad: I was thinking of the cloud consumers. as an operator its easy for me to tell folks to reset their passwords. | 21:54 |
cloudnull | but to the conusmers this may mean dealing many things outside of the cloud | 21:54 |
dstanek | mass password resets are good times | 21:54 |
cloudnull | ^ | 21:55 |
cloudnull | I know i love them | 21:55 |
cloudnull | lbragstad: i ask a couple questions in the review | 21:55 |
cloudnull | thanks for passing this along | 21:55 |
cloudnull | :) | 21:55 |
*** galstrom is now known as galstrom_zzz | 21:55 | |
lbragstad | cloudnull anytime | 21:55 |
lbragstad | cloudnull what about something like a `keystone-manage password-hash` command that gave you a list of users that haven't changed their password? | 21:56 |
lbragstad | cloudnull would that be helpful? | 21:56 |
cloudnull | was just about to ask for that :) | 21:56 |
lbragstad | cloudnull https://review.openstack.org/#/c/438701/13/keystone/common/password_hashing.py | 21:57 |
*** shashank_t_ has quit IRC | 22:00 | |
cloudnull | lbragstad: would folks be open to backporting this to OCATA with the option rolling_upgrade_password_hash_compat set to true for that release? asking because if folks will be forced to reset their passwords before their able to go to QUEEN it may be of some benefit to give users more time? or maybe this option is removed in U instead of Q? | 22:01 |
cloudnull | sorry... | 22:02 |
lbragstad | cloudnull yeah - that's an idea, too | 22:02 |
cloudnull | ... .option is removed in R instead of Q? | 22:02 |
*** kstev has quit IRC | 22:15 | |
dmsimard | cloudnull, odyssey4me, mhayden, mgariepy: Just curious, have you come across any RDO repository instability in the past couple days ? | 22:15 |
*** pmannidi has joined #openstack-ansible | 22:15 | |
sc` | womp womp. build failed in tempest. some nigh decipherable blob of awful and sad times. probably the aforementioned broken that odyssey4me enlightened me to | 22:16 |
dmsimard | sc`: ohai | 22:17 |
dmsimard | sc`: my question above ^ you see anything out of the ordinary ? | 22:18 |
sc` | dmsimard: ohai. nothing thus far, no | 22:20 |
dmsimard | thank you sir | 22:20 |
sc` | as you can see, i picked up another shiny ;) | 22:20 |
dmsimard | sc`: you fleeing chef ? | 22:21 |
sc` | no | 22:21 |
sc` | broadening my horizons | 22:21 |
*** weezS has quit IRC | 22:24 | |
cloudnull | dmsimard: i've not. | 22:25 |
cloudnull | though i may not have been looking | 22:25 |
cloudnull | I did read your manifesto. I failed to comment on it as of yet. | 22:25 |
dmsimard | cloudnull: you would know, some users are experiencing problems since march 14th or so | 22:25 |
cloudnull | but it looks great! | 22:25 |
dmsimard | cloudnull: oh, no worries | 22:25 |
cloudnull | we've had a crazy dance of RDO >< EPEL these last few weeks. | 22:26 |
cloudnull | if there was an issue with the repo mgariepy would probably be the one in the know | 22:26 |
cloudnull | sc`: do you have a sample of the nasty tempest blob ? | 22:27 |
sc` | cloudnull: unfortunately, no. i had to restart iterm2 and i didn't have the run in a tmux | 22:27 |
cloudnull | though I'd suspect if odyssey4me said there was something broken it probably was :) | 22:28 |
sc` | i tried understanding it, but mandatory patchening put a stop to that | 22:28 |
sc` | my focus is on chef right now anyway. it's not like i don't have a todo list miles long | 22:31 |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_nova master: Ensure the components are isolated from the system https://review.openstack.org/449315 | 22:32 |
*** lucasxu has quit IRC | 22:39 | |
*** manheim has joined #openstack-ansible | 22:43 | |
*** marst has quit IRC | 22:50 | |
*** jamesden_ has joined #openstack-ansible | 22:50 | |
*** jamesden_ has quit IRC | 22:50 | |
openstackgerrit | Luz Cazares proposed openstack/openstack-ansible-os_tempest master: (WIP) Copy tempest subunit result files to log folder https://review.openstack.org/449352 | 22:59 |
logan- | cloudnull: was just thinking-- any reason we can't expose the unit file to ops thru config_template as ini file type? | 23:00 |
cloudnull | hum... | 23:01 |
cloudnull | checking | 23:01 |
logan- | i'm trying to think of situations where systemd might break from ini syntax and cant think of any offhand | 23:01 |
cloudnull | nope, seems like it'll work | 23:02 |
cloudnull | i just needed to make sure a unit file would allow for k = v instead of k=v | 23:03 |
cloudnull | and it does | 23:03 |
logan- | nice! | 23:05 |
cloudnull | so we could add these options, as well as others, and then change the template engine to config_template and anything would be possible | 23:07 |
logan- | yep. saves you from having to add a var for the slice too | 23:07 |
cloudnull | ++ | 23:07 |
*** manheim has quit IRC | 23:18 | |
*** MasterOfBugs has quit IRC | 23:19 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_nova master: Ensure the components are isolated from the system https://review.openstack.org/449315 | 23:23 |
cloudnull | logan-: ^ | 23:23 |
* cloudnull fingers crossed :) | 23:23 | |
*** klamath has quit IRC | 23:23 | |
openstackgerrit | Luz Cazares proposed openstack/openstack-ansible master: WIP do not merge https://review.openstack.org/449358 | 23:24 |
*** karimb has quit IRC | 23:27 | |
*** pbandark has quit IRC | 23:31 | |
*** adrian_otto has joined #openstack-ansible | 23:42 | |
*** vishwanathj has quit IRC | 23:43 | |
xdfil | so, I'm confused. I set up an LDAP domain in keystone and I think it's working but, | 23:43 |
xdfil | I'm not understanding how to "map" the LDAP user to a user in keystone | 23:44 |
cloudnull | lbragstad: dstanek: ^ | 23:45 |
*** DimGR2 has joined #openstack-ansible | 23:49 | |
*** DimGR has quit IRC | 23:51 | |
cloudnull | palendae: you around ? | 23:54 |
coolj | xdfil: there's an id_mapping table in keystone that correlates the ldap user id with the keystone user id iirc | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!