| *** mbound has quit IRC | 00:00 | |
| *** r-mibu has quit IRC | 00:05 | |
| *** diazjf has quit IRC | 00:05 | |
| *** weshay has joined #tripleo | 00:07 | |
| *** rbrady has quit IRC | 00:12 | |
| *** dmacpher has quit IRC | 00:13 | |
| *** r-mibu has joined #tripleo | 00:20 | |
| *** apetrich has joined #tripleo | 00:21 | |
| *** absubram has joined #tripleo | 00:30 | |
| openstackgerrit | Dan Sneddon proposed openstack/tripleo-heat-templates: Updates and fixes to NIC config template routes https://review.openstack.org/318389 | 00:34 |
|---|---|---|
| *** julim has joined #tripleo | 00:35 | |
| *** julim has quit IRC | 00:49 | |
| *** fragatina has quit IRC | 00:52 | |
| *** thrash is now known as thrash|g0ne | 00:59 | |
| *** Slower has quit IRC | 01:17 | |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: http://artifacts.<snip>/artifacts/rdo/images/ is a 404 atm https://review.openstack.org/318398 | 01:22 |
| weshay | larsks, ^ | 01:25 |
| openstackgerrit | Paul Belanger proposed openstack/diskimage-builder: Remove hardcoded ext4 values https://review.openstack.org/318400 | 01:27 |
| larsks | weshay: thanks. I don't know if the answer is patch quickstart or fix the server; I'm not all that familiar with the image building side of things. I defer to trown|outtypewww... | 01:29 |
| larsks | If it's still broken in the morning I will take a closer look. | 01:29 |
| weshay | larsks, k | 01:30 |
| *** pkunicki has joined #tripleo | 01:30 | |
| trozet | weshay: hi | 01:38 |
| trozet | weshay: do you know where trown moved his latest mitaka tripleO quickstart images to? https://ci.centos.org/artifacts/rdo/images/mitaka/delorean/stable//undercloud.qcow2 is gone | 01:38 |
| *** sthillma has quit IRC | 01:52 | |
| gbraad | http://buildlogs.centos.org/centos/7/cloud/x86_64/tripleo_images/${RELEASE}/delorean/undercloud.qcow2 | 01:54 |
| *** yamahata has joined #tripleo | 01:55 | |
| openstackgerrit | Gerard Braad proposed openstack/tripleo-quickstart: Change artifacts location in libvirt playbook https://review.openstack.org/318406 | 01:59 |
| *** pradk has joined #tripleo | 01:59 | |
| gbraad | weshay: ^ | 02:03 |
| *** pradk_ has joined #tripleo | 02:06 | |
| dmsimard | gbraad: duplicate of https://review.openstack.org/#/c/318398/ | 02:07 |
| dmsimard | ah, but you know that already :) | 02:08 |
| gbraad | duplciate and different. weshay forgot the libvirt playbook alteration | 02:10 |
| gbraad | I would suggest to merge them actually into a single change | 02:10 |
| *** pradk has quit IRC | 02:10 | |
| *** pradk_ is now known as pradk | 02:10 | |
| *** xinwu has joined #tripleo | 02:12 | |
| *** mbound has joined #tripleo | 02:18 | |
| *** mbound has quit IRC | 02:22 | |
| ayoung | dmsimard, network question for you. I am trying to attach a second VM to the network that connects the undercloud to the overcloud on a quickstart system. I think this is done by the overcloud deploy code...where do I look? | 02:22 |
| ayoung | IO think I need to attach the bridge device to the new vm (called identity) | 02:23 |
| ayoung | I have access to the undercloud vm, and can see the device ther | 02:23 |
| dmsimard | ayoung: I have absolutely no clue, you know more about tripleo and quickstart than I do :( | 02:23 |
| ayoung | dmsimard, talk me through it anyway | 02:24 |
| ayoung | I fake it all the time | 02:24 |
| ayoung | 7: br-ctlplane: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN | 02:24 |
| ayoung | I see that on one vm | 02:24 |
| ayoung | but not on the other | 02:24 |
| ayoung | the undercloud vm has this in the last line of the undercloud-post.sh | 02:25 |
| ayoung | meh..too many lines to put in here, but it creates a device called DEVICE=vlan10 | 02:25 |
| ayoung | and that is type ovs...which talks to that br-ctlplane: | 02:25 |
| ayoung | trown|outtypewww, larsks if either of you are not comatose, feel free to chime in | 02:26 |
| * ayoung guessess larsks actually knows this | 02:26 | |
| dmsimard | You want an interface on both ends on the same network so they can talk to each other ? | 02:26 |
| ayoung | dmsimard, right | 02:26 |
| ayoung | and I need to attach the bridge to the identity vm | 02:27 |
| ayoung | but I can't ID it from the host | 02:27 |
| ayoung | My guess right now is that I need to do something like use virsh to attach the br-ctlplane bridge to the identity domain | 02:28 |
| ayoung | but that is pure guesswork | 02:28 |
| dmsimard | Those are two VMs on the same host ? | 02:28 |
| weshay | ayoung, it's done in the undercloud-post | 02:28 |
| ayoung | weshay, but that depends on the bridge, I thought? | 02:29 |
| ayoung | weshay, http://paste.openstack.org/show/497635/ where? | 02:29 |
| *** myoung|remote has joined #tripleo | 02:30 | |
| weshay | line 77 is the network that connects the undercloud to the overcloud | 02:30 |
| ayoung | weshay, I see where the devcie vlan10 is created, but that depends on the OVS_BRIDGE=br-ctlplane | 02:30 |
| ayoung | weshay, something has to happen before that | 02:30 |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Gnocchi composable roles https://review.openstack.org/318413 | 02:31 |
| ayoung | weshay, I tried just executing that same code on identity, but it does not know about br-ctlplane | 02:31 |
| weshay | k.. sec | 02:31 |
| ayoung | $ sudo ifup ifcfg-vlan10 | 02:32 |
| ayoung | /sbin/ifup: configuration for br-ctlplane not found. | 02:32 |
| ayoung | Usage: ifup <configuration> | 02:32 |
| ayoung | ovs-vsctl: no bridge named br-ctlplane | 02:32 |
| *** yamahata has quit IRC | 02:34 | |
| *** links has joined #tripleo | 02:35 | |
| *** myoung|remote has quit IRC | 02:40 | |
| weshay | ayoung, has the undercloud install completed on the second vm? | 02:41 |
| ayoung | weshay, yes | 02:42 |
| *** ramishra has joined #tripleo | 02:42 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Gnocchi composable roles https://review.openstack.org/318413 | 02:42 |
| weshay | ayoung, no errors? | 02:42 |
| ayoung | weshay So, everything looks right, including overcloud | 02:42 |
| ayoung | I have two vms called undercloud and identity | 02:42 |
| weshay | k | 02:42 |
| ayoung | and the undercloud is set up right on undercloud | 02:42 |
| ayoung | identity is not an undercloud VM | 02:43 |
| weshay | both acting as underclouds? | 02:43 |
| ayoung | no, utility VM, going to run identity stuff on it | 02:43 |
| weshay | br-ctlplane is setup by the undercloud puppet | 02:43 |
| ayoung | how can I trace what that connects to? | 02:43 |
| weshay | so on the undercloud.. you should see 5 neutron networks | 02:44 |
| weshay | on the overcloud you should see atleast two I think and one should connect to the network that vlan10 ison | 02:45 |
| weshay | is on | 02:45 |
| ayoung | on undercloud I see 7a890318-4732-44ad-94c8-4c5b5f2dc25a | ctlplane | ec5b80b8-5406-4aae-a8b6-db95d08895c7 | 02:45 |
| ayoung | provider:physical_network | ctlplane | 02:47 |
| ayoung | weshay, ok, so I need to mimic what neutron is doing here, or is there a better way to connect a VM on brext and brovc to the controller? | 02:48 |
| weshay | ayoung, is ovs installed on the ident box? | 02:48 |
| ayoung | I think so | 02:48 |
| ayoung | it used the same image as the undercloud | 02:49 |
| ayoung | it might not be active, let me check | 02:49 |
| weshay | you need an ovs bridge.. and I think the key is the vlan tag.. | 02:49 |
| weshay | but I'm not sure.. | 02:49 |
| weshay | I get a little lost w/ ovs bridging.. native linux bridges make more sense to me | 02:49 |
| ayoung | $ systemctl status openvswitch.service | 02:50 |
| ayoung | ● openvswitch.service - Open vSwitch | 02:50 |
| ayoung | Loaded: loaded (/usr/lib/systemd/system/openvswitch.service; disabled; vendor preset: disabled) | 02:50 |
| ayoung | Active: active (exited) since Thu 2016-05-19 02:32:02 UTC; 18min ago | 02:50 |
| ayoung | Process: 10327 ExecStart=/bin/true (code=exited, status=0/SUCCESS) | 02:50 |
| ayoung | Main PID: 10327 (code=exited, status=0/SUCCESS) | 02:50 |
| weshay | trozet, did you see the cdn url for the image? | 02:50 |
| weshay | gbraad, so what shall we do w/ our reviews? | 02:50 |
| weshay | ayoung, ok.. the service is active but the bridge needs to be created | 02:51 |
| weshay | ovsbrctl show | 02:51 |
| weshay | or list | 02:51 |
| gbraad | weshay; is it confirmed this is the new location? | 02:52 |
| weshay | gbraad, no.. I would not say it's confirmed, I think what happened is that there was a plan to move to the cdn and that may have happened more quickly then trown thought | 02:53 |
| weshay | we did not get any emails about it, and he didn't mention it today.. so I think there was some kind of mix up, but I'm not sure exactly where | 02:54 |
| weshay | ayoung, try to setup the bridge in the same way as the undercloud using ovs.. then you can attach w/ a different ip on the same vlan. That's not much help though I'm sure | 02:56 |
| ayoung | weshay, that moves me along | 02:56 |
| ayoung | the command is | 02:56 |
| ayoung | sudo ovs-vsctl show | 02:56 |
| ayoung | right? | 02:56 |
| * weshay checks if I have a system running.. sec | 02:56 | |
| ayoung | Bridge br-ctlplane | 02:56 |
| openstackgerrit | Gerard Braad proposed openstack/tripleo-quickstart: Change artifacts location in libvirt playbook https://review.openstack.org/318406 | 02:57 |
| gbraad | weshay: I modified the changeset I had, Add you as co-author if you are OK with this | 02:57 |
| weshay | ya | 02:57 |
| weshay | gbraad, that's fine | 02:57 |
| weshay | gbraad, suggestions on the commit msg are welcome | 02:58 |
| openstackgerrit | Gerard Braad proposed openstack/tripleo-quickstart: Change artifacts location for the undercloud image https://review.openstack.org/318406 | 02:58 |
| ayoung | weshay, http://paste.openstack.org/show/497636/ is what my undercloud has. I assume I need to make that second on on identity | 02:58 |
| gbraad | weshay: The title needs to be more descriptive next time to explain more about the intent of the change ;-) | 02:58 |
| gbraad | ;-) | 02:59 |
| weshay | ayoung, http://docs.openstack.org/developer/tripleo-incubator/neutron_notes.html | 02:59 |
| weshay | add port | 02:59 |
| weshay | gbraad, the thing is.. tihs was unplanned.. or the timing was | 02:59 |
| weshay | but I can make it more descriptive.. I think | 03:00 |
| gbraad | I understand. I was also confused this morning when I noticed a test run failed | 03:00 |
| weshay | oh man.. it's been down all day? | 03:00 |
| weshay | we need a gate just for this | 03:00 |
| ayoung | weshay, so: sudo ovs-vsctl add-port br-ctlplane eth1 ; sudo ip addr add 192.0.2.33/29 dev br-ctlplane | 03:01 |
| weshay | liberty worked 5 hrs ago | 03:01 |
| ayoung | on the identity machine? | 03:01 |
| weshay | ayoung, the ovs looks good.. | 03:01 |
| ayoung | $ sudo ovs-vsctl add-port br-ctlplane eth1 | 03:01 |
| ayoung | ovs-vsctl: no bridge named br-ctlplane | 03:01 |
| weshay | but you might as well create a cfg in /etc/sysconfig/network-scripts | 03:02 |
| weshay | for the port itself | 03:02 |
| weshay | you can mimic what quickstart does there | 03:02 |
| weshay | bah | 03:02 |
| gbraad | weshay: the problem with the title is with the '404 atm' => betetr to explain "Fixes location of undercloud images" instead of saying "they result in a 404". The 404 information can be part of the description itself. The title should be a tl;dr for the description | 03:02 |
| ayoung | so I probably need to create the bridge. | 03:02 |
| ayoung | Let me see if that is in quickstart and I just missed doing it for this vm | 03:03 |
| weshay | gbraad, that's a good suggestion.. thanks | 03:03 |
| weshay | gbraad, going to add a depends-on to your patch | 03:05 |
| openstackgerrit | wes hayutin proposed openstack/tripleo-quickstart: Fixes location of undercloud images https://review.openstack.org/318398 | 03:07 |
| weshay | ayoung, add-br br-ctlplane ? | 03:09 |
| weshay | I need to play w/ ovs again .. geez | 03:09 |
| openstackgerrit | Gerard Braad proposed openstack/tripleo-quickstart: Change artifacts location for the undercloud image https://review.openstack.org/318406 | 03:09 |
| ayoung | weshay, htat looks like part of it | 03:10 |
| ayoung | weshay, can't ping yet, though...think I need this | 03:11 |
| ayoung | Port phy-br-ctlplane | 03:11 |
| ayoung | Interface phy-br-ctlplane | 03:11 |
| ayoung | type: patch | 03:11 |
| ayoung | options: {peer=int-br-ctlplane} | 03:11 |
| ayoung | somehow | 03:11 |
| ayoung | that is on the bridge in undercloud but not identity, yet | 03:11 |
| *** noslzzp_ has quit IRC | 03:12 | |
| *** dsariel has joined #tripleo | 03:13 | |
| *** coolsvap has joined #tripleo | 03:17 | |
| weshay | ayoung, ya.. that's how I would be walking through it as well.. the puppet is not doing anything special there you can't recreate by hand so I think you'll have it in short order | 03:20 |
| ayoung | sudo ip addr add 192.0.2.41/24 dev br-ctlplane | 03:20 |
| ayoung | the bridge needs an address... | 03:20 |
| weshay | ayoung, you may want to try w/o network isolation for your first pass, but you just about have it | 03:21 |
| ayoung | how do I get the Port phy-br-ctlplane? | 03:21 |
| weshay | so .. I think you need to first add the bridge, then the port | 03:22 |
| ayoung | I did... | 03:23 |
| weshay | neither has an ip afaik.. then you attach to the port ifcfg-vlan10 | 03:23 |
| ayoung | so undercloud has this vlan10: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN | 03:25 |
| ayoung | do I need to create the vlan? | 03:25 |
| ayoung | yes I do | 03:26 |
| weshay | yes.. via the interface in /etc/sysconfig/network-scripts | 03:26 |
| ayoung | yeah | 03:26 |
| ayoung | I had it down | 03:26 |
| ayoung | still missing a piece | 03:26 |
| weshay | maybe br-int.. | 03:26 |
| weshay | ovs-vsctl add-port <bridge> <interface> tag=<VLAN number> : Converts port to an access port on specified VLAN (by default all OVS ports are VLAN trunks). | 03:28 |
| weshay | I'm guessing atm | 03:29 |
| weshay | heh.. as if you couldn't tell | 03:29 |
| ayoung | weshay, at least you have enough context to make guesses | 03:29 |
| ayoung | thanks for the help | 03:29 |
| weshay | ayoung, if you are still stuck tomorrow.. thrash|g0ne is pretty dam good.. he's setup ovs for fun and profit | 03:30 |
| ayoung | doesn't this stuff need to be persisted somehwere on undercloud, so it survives a reboot? | 03:30 |
| ayoung | /etc/openvswitch/conf.db | 03:31 |
| weshay | ayoung, ovs is persisted although I'm not sure where the config is written.. and the connection is persisted w/ the ifcfg script | 03:31 |
| ayoung | weshay, goldmine | 03:33 |
| ayoung | its all in there, including the commands | 03:33 |
| weshay | perfect | 03:33 |
| * weshay notes that | 03:33 | |
| weshay | I'm going to sleep.. good luck!! | 03:34 |
| *** weshay has quit IRC | 03:35 | |
| *** MaxPC has joined #tripleo | 03:37 | |
| gbraad | weshay, take care | 03:44 |
| *** MaxPC has quit IRC | 03:47 | |
| *** tzumainn has quit IRC | 03:49 | |
| *** pkunicki has quit IRC | 03:59 | |
| *** links has quit IRC | 04:00 | |
| *** chlong has joined #tripleo | 04:04 | |
| *** coolsvap has quit IRC | 04:04 | |
| openstackgerrit | Jason E. Rist proposed openstack/tripleo-ui: Refactor notification for better display https://review.openstack.org/316874 | 04:05 |
| *** links has joined #tripleo | 04:16 | |
| *** gumbyblake has joined #tripleo | 04:17 | |
| *** masco_ has joined #tripleo | 04:21 | |
| *** anshul has joined #tripleo | 04:23 | |
| *** anshul is now known as Guest75979 | 04:23 | |
| *** coolsvap has joined #tripleo | 04:40 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Deploy Loadbalancer as a composable role https://review.openstack.org/313029 | 05:08 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Restart haproxy after configuring SSL certs https://review.openstack.org/300902 | 05:13 |
| *** jprovazn has joined #tripleo | 05:17 | |
| *** chlong has quit IRC | 05:17 | |
| *** bvandenh has quit IRC | 05:17 | |
| *** coolsvap has quit IRC | 05:20 | |
| *** coolsvap has joined #tripleo | 05:20 | |
| *** coolsvap has quit IRC | 05:21 | |
| *** coolsvap_ has joined #tripleo | 05:21 | |
| *** coolsvap_ is now known as coolsvap | 05:21 | |
| *** chlong has joined #tripleo | 05:30 | |
| *** bvandenh has joined #tripleo | 05:31 | |
| openstackgerrit | Jason E. Rist proposed openstack/tripleo-ui: Refactor notification for better display https://review.openstack.org/316874 | 05:42 |
| *** yamahata has joined #tripleo | 05:44 | |
| *** ramishra has quit IRC | 05:50 | |
| *** ramishra has joined #tripleo | 05:51 | |
| *** bvandenh has quit IRC | 05:57 | |
| *** pkovar has joined #tripleo | 05:58 | |
| *** Guest75979 has quit IRC | 06:02 | |
| *** anshul has joined #tripleo | 06:02 | |
| *** anshul is now known as anshul|bootcamp | 06:02 | |
| *** chlong has quit IRC | 06:05 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Avoid repetition in endpoint map https://review.openstack.org/308182 | 06:06 |
| *** rcernin has joined #tripleo | 06:12 | |
| *** masco_ is now known as masco | 06:16 | |
| *** chlong has joined #tripleo | 06:21 | |
| *** florianf has joined #tripleo | 06:22 | |
| *** pkovar has quit IRC | 06:25 | |
| *** pkovar has joined #tripleo | 06:25 | |
| *** pkovar has quit IRC | 06:27 | |
| *** pkovar has joined #tripleo | 06:27 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Add cloud-initramfs-growroot for Precise https://review.openstack.org/318461 | 06:38 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Tighten the access rules for galera https://review.openstack.org/316297 | 06:39 |
| *** bvandenh has joined #tripleo | 06:43 | |
| *** gfidente has joined #tripleo | 06:44 | |
| *** tremble has joined #tripleo | 06:45 | |
| *** tremble has joined #tripleo | 06:45 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Avoid repetition in endpoint map https://review.openstack.org/308182 | 06:49 |
| *** liverpooler has joined #tripleo | 06:51 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Remove hardcoded ext4 values https://review.openstack.org/318400 | 06:53 |
| *** chlong has quit IRC | 06:54 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Create keystone endpoints in keystone profile https://review.openstack.org/304125 | 06:59 |
| *** jaosorior has quit IRC | 06:59 | |
| *** jaosorior has joined #tripleo | 07:00 | |
| *** karthiks has joined #tripleo | 07:11 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Add cloud-initramfs-growroot for Precise https://review.openstack.org/318461 | 07:12 |
| *** mikelk has joined #tripleo | 07:16 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack-infra/tripleo-ci: Add IP addresses compliant with rfc5737 https://review.openstack.org/289250 | 07:17 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/instack-undercloud: Fix default IP addressed violating rfc5737 https://review.openstack.org/289221 | 07:17 |
| *** rasca has quit IRC | 07:19 | |
| *** ccamacho has joined #tripleo | 07:20 | |
| *** bvandenh has quit IRC | 07:20 | |
| *** paramite has joined #tripleo | 07:20 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Make default addresses comply with rfc5737 https://review.openstack.org/289234 | 07:20 |
| *** mcornea has joined #tripleo | 07:24 | |
| *** rasca has joined #tripleo | 07:24 | |
| *** ebarrera has joined #tripleo | 07:26 | |
| *** veteran has joined #tripleo | 07:29 | |
| *** veteran has quit IRC | 07:30 | |
| *** aufi has joined #tripleo | 07:32 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Export FS_TYPE and remove hardcoded ext4 values https://review.openstack.org/318400 | 07:34 |
| *** devvesa has joined #tripleo | 07:35 | |
| *** zoli_gone-proxy is now known as zoliXXL | 07:38 | |
| *** zoliXXL is now known as zoli|wfh | 07:39 | |
| *** gumbyblake has quit IRC | 07:39 | |
| *** jpena|off is now known as jpena | 07:42 | |
| zoli|wfh | good morning | 07:42 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Pass MysqlVirtualIP via EndpointMap https://review.openstack.org/305751 | 07:43 |
| *** ccamacho is now known as ccamacho|afk | 07:46 | |
| *** markmc``` is now known as markmc | 07:48 | |
| EmilienM | gfidente: can we get https://review.openstack.org/#/c/313026/ and https://review.openstack.org/#/c/313029/ please? | 07:49 |
| *** dbecker has joined #tripleo | 07:50 | |
| *** dmk0202 has joined #tripleo | 07:50 | |
| *** fzdarsky has joined #tripleo | 07:51 | |
| *** ifarkas has joined #tripleo | 07:56 | |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: WIP: move pacemaker resource to step 4 https://review.openstack.org/315938 | 08:00 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: deploy keystone at step 3 and 4 https://review.openstack.org/315943 | 08:00 |
| *** stendulker has joined #tripleo | 08:04 | |
| *** shardy has joined #tripleo | 08:05 | |
| openstackgerrit | Emilien Macchi proposed openstack/instack-undercloud: Deploy and Upgrade Ironic to run in mod_wsgi https://review.openstack.org/309140 | 08:07 |
| *** dtantsur|afk is now known as dtantsur | 08:08 | |
| openstackgerrit | Merged openstack/python-tripleoclient: Allow import command to set deploy image and local boot https://review.openstack.org/299362 | 08:09 |
| jaosorior | EmilienM: stupid question, where are the $::os_service_default defined? | 08:10 |
| EmilienM | jaosorior: its a fact in puppet-openstacklib, let me find the code | 08:10 |
| EmilienM | https://github.com/openstack/puppet-openstacklib/blob/104f2d3e5836dd4c39e4b954cd93e4c86798fae4/facts.d/os_service_default.txt | 08:11 |
| jaosorior | I see | 08:12 |
| jaosorior | I'm still missing what the <SERVICE DEFAULT> value is really :/ | 08:13 |
| EmilienM | jaosorior: http://docs.openstack.org/developer/puppet-openstack-guide/coding-style.html#config-file-defaults-and-parameters | 08:13 |
| *** ohamada has joined #tripleo | 08:16 | |
| *** yamahata has quit IRC | 08:17 | |
| shardy | Morning all | 08:18 |
| shardy | can someone please check out https://review.openstack.org/#/c/296578/ | 08:19 |
| shardy | it's a simple fix, been waiting ~2months | 08:19 |
| shardy | Also https://review.openstack.org/#/c/314260/ https://review.openstack.org/#/c/315504 https://review.openstack.org/#/c/301266/ https://review.openstack.org/#/c/314344/ | 08:23 |
| marios | shardy: i don't think the ci fails would even apply here - well possibly thte upgrades job but that is green (https://review.openstack.org/#/c/296578/1) | 08:24 |
| marios | shardy: s/apply/be relevant | 08:25 |
| shardy | marios: it passed all the jobs over the last two rechecks | 08:26 |
| *** electrofelix has joined #tripleo | 08:26 | |
| *** lucas-dinner is now known as lucasagomes | 08:26 | |
| shardy | marios: what bug reference are you looking for there? The override is part of the heat plugin API, basically the initial implementation was incomplete | 08:27 |
| shardy | I assumed the commit message bug reference w/git blame would be enough | 08:27 |
| marios | shardy: yeah but ... apr <--> may ... should we wait for the latest run since it got started ~7 mins ago | 08:28 |
| shardy | Sure, what's another two hours ;) | 08:28 |
| marios | shardy: yeah the launchpad bug... yeah as i said no biggie (if you wanted to you could find it) | 08:28 |
| EmilienM | shardy: done | 08:28 |
| shardy | thanks all | 08:29 |
| *** jistr has joined #tripleo | 08:30 | |
| shardy | https://review.openstack.org/#/c/296488/ could probably land too while we're about it :) | 08:30 |
| d0ugal | tripleo-common has become a rather strange collection of code | 08:30 |
| EmilienM | shardy: if you have time, I have haproxy roles ready for review: https://review.openstack.org/313026 and https://review.openstack.org/313029 | 08:30 |
| shardy | EmilienM: thanks, will check it out | 08:31 |
| marios | shardy: k I +Ad it after the compelling case you made | 08:32 |
| marios | :D | 08:32 |
| shardy | hehe :D | 08:32 |
| EmilienM | marios: also feel fee to look my patches for haproxy roles ^ if you have time | 08:33 |
| * hewbrocca checks watch | 08:33 | |
| hewbrocca | EmilienM: are you in France? | 08:33 |
| EmilienM | hewbrocca: yes | 08:33 |
| marios | EmilienM: may have to bump til tomorrow at this rate man already been on reviews too long today | 08:33 |
| EmilienM | marios: but it's me ! | 08:34 |
| marios | EmilienM: i know it is breaking my heart | 08:34 |
| EmilienM | you have no idea how I feel now | 08:35 |
| marios | :( | 08:36 |
| *** chem`` has quit IRC | 08:45 | |
| *** ccamacho|afk is now known as ccamacho | 08:45 | |
| *** dtantsur is now known as dtantsur|bbl | 08:46 | |
| shardy | bandini: Hey, can you pls check out https://review.openstack.org/#/c/313029 ? | 08:46 |
| shardy | there is a question re the dummy pacemaker resource you added in c5049d8ffbf9579a6e726c0031411ddb57e1a625 and how it relates to enabling the loadbalancer | 08:47 |
| bandini | shardy: sure thing | 08:48 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Change ping wait flag https://review.openstack.org/314344 | 08:48 |
| openstackgerrit | Merged openstack/tripleo-common: Fix no-replace-server to accurately preview update https://review.openstack.org/296578 | 08:50 |
| *** bvandenh has joined #tripleo | 08:50 | |
| bandini | shardy: ok I replied in the review. I'll be around for any questions/doubts/thoughts/cluebats ;) | 08:54 |
| shardy | bandini: thanks, so basically we should move it outside the if $enable_load_balancer conditional then | 08:55 |
| shardy | I assume probably also the resource_defaults? Which would be good as then we can remove the entire conditional | 08:56 |
| bandini | shardy: yes you're right, we should move that out including the resource_defaults inclusion | 08:58 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Add missing ManagementIpSubnet https://review.openstack.org/301266 | 08:58 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: add heat-api-cfn to endpoint map https://review.openstack.org/315504 | 08:58 |
| shardy | EmilienM: ^^ | 08:58 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Remove ControllerClusterConfig from overcloud.yaml https://review.openstack.org/314260 | 08:58 |
| shardy | bandini: great, that cleans up the moving things into the loadbalancer composable service template then :) | 08:59 |
| bandini | gfidente: agree? I am positive about this, but maybe with external lb there are situations I haven't thought of? | 08:59 |
| bandini | shardy: aye :) | 08:59 |
| *** derekh has joined #tripleo | 09:00 | |
| *** jcoufal has joined #tripleo | 09:00 | |
| EmilienM | shardy: ok I'll address it | 09:02 |
| EmilienM | shardy, bandini: why did you +1 then? | 09:02 |
| shardy | EmilienM: thanks, +2 otherwise | 09:02 |
| EmilienM | ok | 09:02 |
| EmilienM | cool | 09:02 |
| bandini | EmilienM: because I noticed that part afterwards | 09:03 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Avoid repetition in endpoint map https://review.openstack.org/308182 | 09:03 |
| EmilienM | ok | 09:04 |
| EmilienM | no prob | 09:04 |
| *** chem has joined #tripleo | 09:06 | |
| *** bret_ has joined #tripleo | 09:08 | |
| EmilienM | shardy, bandini: are we ok to keep https://review.openstack.org/#/c/313029/10/puppet/manifests/overcloud_controller_pacemaker.pp and just drop my question comment? | 09:09 |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Add net-config-static.yaml https://review.openstack.org/301267 | 09:09 |
| EmilienM | or do you want me to move the dummy resource outside the if? | 09:09 |
| *** bret_ is now known as khrystoph | 09:11 | |
| shardy | EmilienM: I think the comment and the conditional need to be removed | 09:12 |
| bandini | EmilienM: I'd move it out, because if $enable_load_balancer is not true we will fail anyways when we create the first constraint requiring 'openstack-core' | 09:12 |
| *** chem has quit IRC | 09:12 | |
| *** chem has joined #tripleo | 09:12 | |
| *** akrivoka has joined #tripleo | 09:13 | |
| EmilienM | ok cool | 09:13 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: Add loadbalancer profile for ha & non-ha https://review.openstack.org/313026 | 09:15 |
| *** khrystoph has quit IRC | 09:15 | |
| *** khrystoph has joined #tripleo | 09:16 | |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: Deploy Loadbalancer as a composable role https://review.openstack.org/313029 | 09:16 |
| EmilienM | shardy, bandini: done, I'll ping you when CI passed so we can reconsider it | 09:16 |
| bandini | EmilienM: thanks! | 09:16 |
| EmilienM | I also improved puppet-tripleo patch | 09:16 |
| EmilienM | to drop useless params, and use hiera in the THT | 09:17 |
| *** khrystoph has quit IRC | 09:17 | |
| *** akrivoka has quit IRC | 09:17 | |
| *** khrystoph has joined #tripleo | 09:17 | |
| *** akrivoka has joined #tripleo | 09:17 | |
| *** sambetts|afk is now known as sambetts | 09:17 | |
| *** khrystoph has quit IRC | 09:18 | |
| *** khrystoph has joined #tripleo | 09:18 | |
| *** khrystoph has quit IRC | 09:19 | |
| *** khrystoph has joined #tripleo | 09:21 | |
| *** khrystoph has quit IRC | 09:25 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Composable roles within services - MongoDB https://review.openstack.org/315051 | 09:38 |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Adds memcached as a composable service https://review.openstack.org/312899 | 09:43 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Composable roles within services - NTP https://review.openstack.org/310421 | 09:49 |
| openstackgerrit | Carlos Camacho proposed openstack/puppet-tripleo: Composable roles within services - NTP https://review.openstack.org/310725 | 09:55 |
| openstackgerrit | Carlos Camacho proposed openstack/puppet-tripleo: Composable roles within services - NTP https://review.openstack.org/310725 | 09:58 |
| *** dixiaoli has joined #tripleo | 09:59 | |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Adds memcached as a composable service https://review.openstack.org/312899 | 10:00 |
| *** dixiaoli has quit IRC | 10:00 | |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Nodes Registration https://review.openstack.org/315136 | 10:02 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Replace NodeConfigIdentifiers with DeployIdentifier https://review.openstack.org/315616 | 10:05 |
| *** xinwu has quit IRC | 10:05 | |
| *** dtantsur|bbl is now known as dtantsur | 10:11 | |
| *** paramite is now known as paramite|afk | 10:16 | |
| *** zoli|wfh is now known as zoli|afk | 10:22 | |
| *** d0ugal has quit IRC | 10:24 | |
| openstackgerrit | Michele Baldessari proposed openstack/tripleo-specs: Add initial next generation HA architecture spec https://review.openstack.org/299628 | 10:27 |
| *** d0ugal has joined #tripleo | 10:28 | |
| *** d0ugal has quit IRC | 10:31 | |
| *** d0ugal has joined #tripleo | 10:31 | |
| *** apetrich has quit IRC | 10:33 | |
| *** ansiwen_ is now known as ansiwen | 10:41 | |
| yolanda | question, i'm trying to enroll some nodes to tripleo using pxe_ssh, but i got trouble with the mac address. My nodes are vms, and i'm just setting the mac address with the ether field of the nic . But i get error about "SSH driver was not ablet o find a VM with any of the specified MACs" | 10:46 |
| *** fzdarsky is now known as fzdarsky|afk | 10:47 | |
| *** ccamacho has quit IRC | 10:48 | |
| shadower | I'm trying tripleo-quickstart but it's failing to download: http://artifacts.ci.centos.org/artifacts/rdo/images/mitaka/delorean/stable/undercloud.qcow2.md5 | 10:49 |
| shadower | indeed the "images" directory's missign from http://artifacts.ci.centos.org/artifacts/rdo/ | 10:49 |
| ansiwen | yolanda: wow, I just wanted to ask, how I can automate installation of baremetal server (centos) | 10:49 |
| ansiwen | yolanda: sounds like pxe_ssh is interesting for me | 10:49 |
| shadower | any ideas what to do about that? Is that a centos ci issue or something in quickstart? | 10:49 |
| yolanda | ansiwen, i just learned from it today | 10:50 |
| shardy | shadower: does this patch fix it? https://review.openstack.org/#/c/318398/2/quickstart.sh | 10:50 |
| yolanda | but i have the problem of which mac address is needed, doesn't seem to be the one from the nic... | 10:50 |
| ansiwen | yolanda: is that a standalone tool? | 10:50 |
| yolanda | it's an ironic driver | 10:51 |
| yolanda | http://docs.openstack.org/developer/tripleo-docs/environments/environments.html | 10:51 |
| shadower | shardy: I checked open patches and didn't see this one :-(. I'll give it a try & comment | 10:51 |
| ansiwen | yolanda: oh, I see... that might be overkill for my single baremetal machine | 10:51 |
| yolanda | i'm trying a deploy with openstack vms | 10:51 |
| ansiwen | yolanda: I just want to make the installation with all packages easily repeatable | 10:52 |
| yolanda | ansiwen, i'm just trying to learn more about tripleo, and install with the resources i have available... | 10:53 |
| *** weshay has joined #tripleo | 10:55 | |
| *** zoli|afk is now known as zoli|wfh | 10:56 | |
| ansiwen | yolanda: yes, me too, but I though the hardware I want to run tripleO on doesn't have to be deployed by openstack. :-) | 10:56 |
| yolanda | no, it doesn't have to.. but it's the resources i have to test, that's my use case | 10:57 |
| shardy | ansiwen: IMO a kickstart file is the simplest solution to your requirement - doing a manual install creates a kickstart in /root that you can then customize and reuse | 10:57 |
| shardy | yolanda: perhaps dtantsur can help with your node registration issue | 10:57 |
| yolanda | shardy, asking on ironic channel, yep | 10:58 |
| *** ccamacho has joined #tripleo | 11:06 | |
| *** ccamacho has quit IRC | 11:13 | |
| ansiwen | shardy: oh, that's a great hint, so I can do it manually, and just replay it with that kickstart file? cool! | 11:13 |
| weshay | gbraad, for gates to pass for your review.. you should change your review to sit on top of mine.. git-review -d I152717fbc2d3beb6936468f6eb4724b89bcd0edd; then add a commit for yours and git-review | 11:17 |
| *** zoli|wfh is now known as zoli|lunch | 11:19 | |
| *** rodrigods has quit IRC | 11:20 | |
| *** rodrigods has joined #tripleo | 11:20 | |
| *** mburned_out is now known as mburned | 11:21 | |
| *** paramite|afk is now known as paramite | 11:26 | |
| *** paramite is now known as paramite|afk | 11:29 | |
| *** lucasagomes is now known as lucas-hungry | 11:30 | |
| *** rhallisey has joined #tripleo | 11:33 | |
| dtantsur | ansiwen, pxe_ssh is for virtual environments, it won't work on a bare metal | 11:34 |
| *** myoung|remote has joined #tripleo | 11:41 | |
| *** khrystoph has joined #tripleo | 11:43 | |
| *** ccamacho has joined #tripleo | 11:45 | |
| gbraad | weshay check\ | 11:45 |
| *** khrystoph has quit IRC | 11:49 | |
| openstackgerrit | Gerard Braad proposed openstack/tripleo-quickstart: Change artifacts location for the undercloud image https://review.openstack.org/318406 | 11:50 |
| gbraad | rebased | 11:51 |
| *** ccamacho has quit IRC | 11:52 | |
| *** athomas has joined #tripleo | 11:57 | |
| *** lucas-hungry is now known as lucasagomes | 11:57 | |
| *** trown|outtypewww is now known as trown | 12:02 | |
| *** penick has quit IRC | 12:06 | |
| *** jistr has quit IRC | 12:08 | |
| ansiwen | dtantsur: thanks for that important detail :-) | 12:10 |
| *** stendulker has quit IRC | 12:12 | |
| *** ccamacho has joined #tripleo | 12:12 | |
| gfidente | marios, so I'd merge https://review.openstack.org/#/c/312877/6 | 12:12 |
| *** jayg|g0n3 is now known as jayg | 12:12 | |
| gfidente | ack? | 12:12 |
| trown | gbraad: sorry just coming online, but we don't want to change to the CDN URL | 12:15 |
| *** ccamacho has quit IRC | 12:16 | |
| trown | it had its own issues, and the nuking of the artifacts server images was definitely not intentional, trying to work out what happened with folks in #centos-devel | 12:16 |
| weshay | panda, adarazs tripleo-ci stand up | 12:17 |
| marios | gfidente: ack thanks. yeah i think it is ready ... waiting for green ci on the tht side | 12:19 |
| marios | gfidente: esp since we agree on the openstack-core at least for now leave it there | 12:19 |
| gfidente | marios, yes the constraints depend on two services so wherever we put it's going to break if the other is missing | 12:20 |
| gfidente | but if openstack-core stays in _pacemaker.pp better to have the deps in the role | 12:20 |
| *** dprince has joined #tripleo | 12:20 | |
| *** apetrich has joined #tripleo | 12:23 | |
| *** hewbrocca is now known as hewbrocca-afk | 12:23 | |
| *** hewbrocca-afk is now known as hewbrocca | 12:24 | |
| openstackgerrit | Imre Farkas proposed openstack/python-tripleoclient: Add Ironic password generation during deployment https://review.openstack.org/316146 | 12:26 |
| *** morazi has joined #tripleo | 12:26 | |
| *** penick has joined #tripleo | 12:27 | |
| *** jistr has joined #tripleo | 12:28 | |
| *** jistr is now known as jistr|bbl | 12:28 | |
| *** anshul|bootcamp has quit IRC | 12:31 | |
| *** weshay is now known as weshay_mt | 12:33 | |
| *** weshay_mt is now known as weshay_mtg | 12:33 | |
| *** julim has joined #tripleo | 12:34 | |
| *** jpena is now known as jpena|lunch | 12:34 | |
| *** links has quit IRC | 12:36 | |
| *** ccamacho has joined #tripleo | 12:37 | |
| *** zoli|lunch is now known as zoli|wfh | 12:37 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add Cinder API/Scheduler/Volume roles https://review.openstack.org/310939 | 12:39 |
| *** _prad_ has joined #tripleo | 12:40 | |
| *** rhallisey has quit IRC | 12:41 | |
| *** anshul has joined #tripleo | 12:44 | |
| *** anshul is now known as Guest19218 | 12:44 | |
| *** rhallisey has joined #tripleo | 12:44 | |
| *** rlandy has quit IRC | 12:48 | |
| *** panda has quit IRC | 12:48 | |
| *** panda has joined #tripleo | 12:48 | |
| *** rlandy has joined #tripleo | 12:49 | |
| *** penick has quit IRC | 12:49 | |
| *** mbound has joined #tripleo | 12:51 | |
| *** Guest19218 has quit IRC | 12:51 | |
| *** myoung|remote has quit IRC | 12:51 | |
| *** abehl has joined #tripleo | 12:51 | |
| *** akshai has joined #tripleo | 12:56 | |
| *** fultonj has joined #tripleo | 12:56 | |
| *** zoli|wfh is now known as zoli|mtg | 12:58 | |
| *** akshai has quit IRC | 12:58 | |
| *** akshai has joined #tripleo | 13:03 | |
| larsks | ayoung: No longer comatose; you all set w/ networking? | 13:08 |
| larsks | trown: o/ | 13:08 |
| ayoung | larsks, well, not really | 13:08 |
| ayoung | larsks, but I have been watching OVN presentations, so I know how far I have to go | 13:08 |
| trown | larsks: o/ | 13:08 |
| ayoung | and I need to figure out how to reproduce the setup that the undercloud Node has on the additional nodes. | 13:09 |
| ayoung | Wondering if there is an easy way to "enroll" the second VM so neutron can do its magic | 13:09 |
| larsks | So, I missed the discussion yesterday...what exactly is it you're trying to do? | 13:09 |
| *** lblanchard has joined #tripleo | 13:11 | |
| larsks | Also, OVN? We're not using OVN right now... | 13:12 |
| jaosorior | ayoung: OVN? O_o | 13:13 |
| ayoung | jaosorior, OVS rather | 13:13 |
| ayoung | larsks, right...OK so here is what I need: | 13:13 |
| jaosorior | that makes more sense | 13:13 |
| ayoung | I have to install a bunch of different services that are not part of OpenStack, but that will integrate in with it. To simulate these being "pre-existing" I have created a second VM just like the undercloud VM | 13:14 |
| ayoung | but, this VM needs to be able to talk to the overcloud | 13:14 |
| ayoung | so I need to connect the VM to the control plane network | 13:14 |
| ayoung | it needs to talk to the undercloud, too | 13:15 |
| larsks | Right. | 13:15 |
| jaosorior | I guess putting that in the internal-api network would make sense too? | 13:15 |
| ayoung | so, I was looking at how the undercloud VM is set, and there is a stage post undercloud install where it creates the interface | 13:15 |
| ayoung | jaosorior, sounds right | 13:16 |
| *** noslzzp has joined #tripleo | 13:17 | |
| larsks | It seems the easiest solution is to set up your vm like an overcloud node (a single interface, attached to the "overcloud" bridge on the host). | 13:17 |
| ayoung | larsks, it looks like the br-ctl-plane bridge is set up by openstack undercloud install. That is what I think I need to replicate | 13:18 |
| *** sshnaidm has quit IRC | 13:18 | |
| larsks | Do you need to set up a bridge? | 13:18 |
| ayoung | and I think the commands to do that are all cached in the /etc/openvs .db file | 13:18 |
| ayoung | larsks, yes | 13:18 |
| larsks | Why's that? | 13:18 |
| ayoung | larsks, otherwise the new VM has no way to communicate with the undercloud. THere is no route | 13:19 |
| larsks | If you set up your vm like the overlcoud network -- a single interface, attached to the "overcloud" bridge on the host -- your vm is sitting directly on the ctlplane network. | 13:20 |
| larsks | Just like the compute hosts or controllers. | 13:20 |
| larsks | Give it an appropriate ip and it can communicate with both the undercloud and the overcloud hosts, no additional work required. | 13:20 |
| *** sshnaidm has joined #tripleo | 13:20 | |
| ayoung | larsks, OK, I'm rerunning quickstart right now. As soon as it is done, I'll take a look | 13:21 |
| *** noslzzp has quit IRC | 13:21 | |
| larsks | Just look at 'virsh dumpxml' for one of the overcloud nodes to see what the libvirt side looks like. | 13:21 |
| *** noslzzp has joined #tripleo | 13:22 | |
| larsks | trown: any chance you can take https://review.openstack.org/#/c/316264/ for a spin today? | 13:22 |
| *** rbrady has joined #tripleo | 13:22 | |
| trown | larsks: ya it is on my list... planned to do it first thing today, then got a bit sidetracked restoring all of our images getting deleted | 13:23 |
| *** jdob has quit IRC | 13:23 | |
| larsks | Yeah :) | 13:23 |
| *** jdob has joined #tripleo | 13:23 | |
| trown | wtf on that... but I think I have things back in good shape | 13:23 |
| ayoung | jaosorior, but as soon as I have this solved, I have a place to get you an IPA instance, and all of the rest of the TLS stuff should fall into line.... | 13:24 |
| ayoung | larsks, trown by the way. Quickstart is awesome. | 13:24 |
| jaosorior | yay :D | 13:25 |
| trown | :) | 13:25 |
| larsks | ayoung: glad it helps! | 13:25 |
| ayoung | I mean like, a game saver. We are talking Sam-Mayday-Malone | 13:25 |
| * trown googles Sam-Mayday-Malone | 13:25 | |
| trown | ah Ted Danson | 13:26 |
| *** trown is now known as trown|brb | 13:26 | |
| ayoung | And the Mayday nickname was because he was the relief pitcher.... | 13:26 |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Enable Manila integration https://review.openstack.org/188137 | 13:30 |
| *** liverpooler has quit IRC | 13:30 | |
| *** penick has joined #tripleo | 13:35 | |
| *** jtomasek_ has joined #tripleo | 13:35 | |
| *** jslim has joined #tripleo | 13:38 | |
| *** jdob has quit IRC | 13:40 | |
| *** jdob has joined #tripleo | 13:40 | |
| *** trown|brb is now known as trown | 13:41 | |
| *** thrash|g0ne is now known as thrash | 13:43 | |
| *** jpena|lunch is now known as jpena | 13:46 | |
| *** tzumainn has joined #tripleo | 13:48 | |
| *** akshai has quit IRC | 13:52 | |
| *** penick has quit IRC | 13:52 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add Cinder API/Scheduler/Volume roles https://review.openstack.org/310939 | 13:53 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Switch Cinder Api/Scheduler/Volume to composable roles https://review.openstack.org/310940 | 13:54 |
| *** masco has quit IRC | 13:57 | |
| jslim | Good morning everyone. Can anyone tell me if Liberty or Mitaka is the better choice for a production deployment when being conducted by someone who is relatively new to Openstack in general ? | 13:58 |
| jstir | jslim: ITYM 'OpenStack'. | 13:58 |
| *** veteran has joined #tripleo | 13:59 | |
| openstackgerrit | Merged openstack/tripleo-puppet-elements: Honor $DIB_DEFAULT_INSTALLTYPE in puppet-modules https://review.openstack.org/318293 | 13:59 |
| jaosorior | jslim: Well, for TripleO I think both Liberty and Mitaka give very similar deployment experiences. Will you try to deploy upstream? | 13:59 |
| dtantsur | jslim, (ignore our case-sensitive bot) I'd say both are good. Liberty got somewhat more testing, Mitaka has more features and was less affected by feature backports | 14:00 |
| dtantsur | jstir, don't confuse newcomers! | 14:00 |
| dtantsur | :) | 14:00 |
| *** jistr|bbl is now known as jistr | 14:00 | |
| jslim | Thanks for the answer. | 14:00 |
| *** akshai has joined #tripleo | 14:00 | |
| jrist | lol precious jstir | 14:00 |
| jrist | shadower: ^ | 14:01 |
| jistr | oh we have it here too | 14:01 |
| jistr | jstir: hi | 14:01 |
| jstir | jistr: Whatever. | 14:01 |
| jaosorior | holy crap, ^^ those nicknames are confusing | 14:01 |
| trown | lol | 14:02 |
| *** trown is now known as jsitr | 14:02 | |
| jaosorior | hahahaha | 14:02 |
| openstackgerrit | Brad P. Crochet proposed openstack/puppet-tripleo: Add Heat profiles https://review.openstack.org/313019 | 14:02 |
| *** jsitr is now known as trown | 14:02 | |
| *** zoli|mtg is now known as zoli|wfh | 14:03 | |
| gfidente | jaosorior, that's when you forget about jrist | 14:03 |
| gfidente | because that is real guy as well | 14:03 |
| jrist | lol | 14:03 |
| jslim | jaosorior: I am not totally sure what deploy upstream means I am afraid. Could you give me the highly abbreviated definnition ? :-) | 14:03 |
| jrist | yeah | 14:03 |
| jrist | that's why this started jaosorior | 14:03 |
| jrist | because I was here first, then jistr joined | 14:04 |
| jrist | jstir is a bot | 14:04 |
| gfidente | jstir, are you two bots? | 14:05 |
| pabelanger | derekh: slagle: do you have a moment to reply to my ML post about AFS. I'd like to see if we can come to an agreement on it today | 14:06 |
| jaosorior | jslim: Sure! So... Are you planning to use tripleo coming from RDO or coming from your provider? | 14:07 |
| jaosorior | or are you planning to clone tripleo's code and go from there? | 14:07 |
| openstackgerrit | Marios Andreou proposed openstack/puppet-tripleo: Adds the base and pacemaker profile for the memcached service https://review.openstack.org/312877 | 14:09 |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Adds memcached as a composable service https://review.openstack.org/312899 | 14:09 |
| dtantsur | shadower, a bug report: jstir does not understand "jstir,", only "jstir:" | 14:11 |
| trown | jstir: be smarter | 14:12 |
| jstir | trown: Whatever. | 14:12 |
| jslim | jaosorior: Oh. RDO all the way. I have been installing with quickstart and just using the deployment script generated but have only got as far as the undercloud deployment and have yet to even be able to login to the undercloud dashboard ergo my question about the releases as I was afraid I was working on one that might have been a little to knew.... | 14:12 |
| jaosorior | aaah I see | 14:13 |
| jaosorior | well, if you're using quickstart, you might need to tweak some things to get into the dashboard | 14:14 |
| jaosorior | ayoung had a similar issue ^^ | 14:14 |
| *** nico_auv has joined #tripleo | 14:16 | |
| jslim | jaosorior: Got it. I am assuming I only have to tweak items that are directly related to Apache and Django or are there other gremlins ? | 14:16 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Split patternfly-additions import https://review.openstack.org/318696 | 14:16 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Add Management net to ips-from-pool-all example https://review.openstack.org/318699 | 14:17 |
| *** myoung|remote has joined #tripleo | 14:18 | |
| jaosorior | well, mostly the issues ayoung had, IIRC, is that he needed to properly set up port forwarding to be able to access the dashboard. As it's not exposed automatically. | 14:18 |
| mandre | shardy, dprince: slagle had concerns about having ansible validations in instack-undercloud, since it would couple unnecessary coupling | 14:18 |
| mandre | see https://review.openstack.org/#/c/285860/ | 14:18 |
| mandre | any objection having the validations in tripleo-common instead? | 14:18 |
| dprince | mandre: is that the right patch link? | 14:20 |
| dprince | mandre: sorry, I'm not following how that relates to anything Ansible | 14:20 |
| akrivoka | dtantsur: thanks for the "Nodes management" email, I learned a lot about the various node states and correct ways to move between them | 14:21 |
| dtantsur | akrivoka, great to hear! | 14:21 |
| openstackgerrit | Brad P. Crochet proposed openstack/os-cloud-config: Add Heat CloudFormation service for endpoint creation https://review.openstack.org/318704 | 14:21 |
| mandre | dprince, shardy: sorry, wrong patch link :/ | 14:21 |
| mandre | have a look at https://review.openstack.org/#/c/316522/ | 14:22 |
| *** myoung|remote has quit IRC | 14:22 | |
| akrivoka | dtantsur: this is email is gold, I wonder if this is already covered by the Ironic docs - if not it could maybe be added almost verbatim :) | 14:22 |
| openstackgerrit | Brad P. Crochet proposed openstack/python-tripleoclient: Add Heat CloudFormation service for endpoint creation https://review.openstack.org/318705 | 14:22 |
| dtantsur | akrivoka, heh, I'll consider it | 14:23 |
| *** jpena is now known as jpena|brb | 14:23 | |
| mandre | dprince: though, I'd appreciate your review on https://review.openstack.org/#/c/285860/ too :D | 14:23 |
| *** akrivoka has quit IRC | 14:23 | |
| *** akrivoka has joined #tripleo | 14:23 | |
| openstackgerrit | Brad P. Crochet proposed openstack/tripleo-heat-templates: composable heat services https://review.openstack.org/313577 | 14:24 |
| dprince | mandre: I commented. Sorry, I thought we agreed that Ansible wouldn't require shelling into the undercloud in order to run it? | 14:25 |
| *** myoung has joined #tripleo | 14:26 | |
| openstackgerrit | Emilien Macchi proposed openstack/instack-undercloud: Deploy and Upgrade Ironic to run in mod_wsgi https://review.openstack.org/309140 | 14:27 |
| mandre | dprince: I don't understand your comment, is that about the dependency on ansible? | 14:30 |
| hdsdman_ | to be honest, my preferred deploy method is ansible and not heat :/ | 14:32 |
| jslim | jaosorior: Got it. I think I have the forwarding part with ssh down. I seem to be running into problems with missing configs though... mainly the actual apache config for the dashboard itself so I must be doing something wring. Can you tell me if it is acceptable to just re-run quickstart.sh from your seed host to re-deploy the undercloud VM or should I really re-install the OS on the undercloud host ? | 14:32 |
| *** jasond has joined #tripleo | 14:33 | |
| jaosorior | for quickstart related questions probably larsks and trown are better suited | 14:33 |
| jaosorior | ^^ | 14:34 |
| dprince | mandre: I guess it is about the hosts.sample and where and how it is created | 14:34 |
| dprince | mandre: like 'stack' user isn't what I use to create my cloud | 14:34 |
| dprince | mandre: so it won't work for me | 14:34 |
| mandre | dprince: forget about the hosts.sample, it's not used for the mistral based validations | 14:34 |
| jslim | jaosorior: Got it. Thanks. | 14:35 |
| *** zeroshft has joined #tripleo | 14:35 | |
| mandre | dprince: regarding the stack user, I'm trying to get rid of this dependency in my other patch | 14:35 |
| mandre | dprince: I was thinking about adding a 'validations' user to the undercloud, generate a keypair and upload it to nova in instack-undercloud's undercloud.py script | 14:36 |
| mandre | dprince: would that work for you? | 14:36 |
| *** aufi has quit IRC | 14:36 | |
| *** penick has joined #tripleo | 14:37 | |
| trozet | hi all, is there anyway to specify multiple compute pre config resources? OS::TripleO::ComputeExtraConfigPre: | 14:37 |
| trown | jslim: if it is just a matter of accessing the undercloud horizon, the instructions in https://github.com/openstack/tripleo-quickstart/blob/master/docs/accessing-overcloud.md could be modified slightly to work | 14:38 |
| mandre | dprince: we would run automated validations as the 'validations' user on the undercloud, which would be created in the 'openstack undercloud install' step | 14:38 |
| trown | jslim: I am not totally sure what the question is | 14:38 |
| *** penick has quit IRC | 14:39 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: Required change to adjust to openstack CI. https://review.openstack.org/318719 | 14:40 |
| openstackgerrit | Merged openstack/tripleo-ui: Tracking nodes operations in Nodes Table https://review.openstack.org/314233 | 14:40 |
| jasond | can anybody point me to the documentation for this template language? https://github.com/openstack/tripleo-image-elements/blob/master/elements/os-collect-config/os-apply-config/etc/os-collect-config.conf | 14:41 |
| *** penick has joined #tripleo | 14:41 | |
| dprince | mandre: for me all of the validations should be client driven I think | 14:41 |
| *** noslzzp has quit IRC | 14:41 | |
| dprince | mandre: so like, not instack-undercloud, not tripleo-common | 14:41 |
| dprince | mandre: it doesn't relate to any of those projects really | 14:41 |
| dprince | mandre: it is a separate, new project that allows you to create custom Ansible playbooks to validate things (as a client) | 14:42 |
| mandre | dprince: are you suggesting we create a tripleo-validations project then? shardy seemed opposed to it | 14:42 |
| *** yamahata has joined #tripleo | 14:43 | |
| *** noslzzp has joined #tripleo | 14:43 | |
| ayoung | jaosorior, I really want to come up with a solution that does not require setting up a proxy | 14:43 |
| *** jslim has quit IRC | 14:43 | |
| jaosorior | ayoung ?? | 14:44 |
| dprince | mandre: perhaps. | 14:44 |
| ayoung | jaosorior, your comment on reaching the undercloud Horizon | 14:44 |
| jaosorior | ah, that | 14:44 |
| ayoung | I was using sshuttle | 14:44 |
| mandre | dprince: tripleo-common seems like the best place to me if we chose to not go down the path creating a new project | 14:45 |
| dprince | mandre: tripleo-common is better than instack-undercloud, sure. But I'm still not sure it is the right place for these | 14:45 |
| *** jslim has joined #tripleo | 14:45 | |
| dprince | mandre: what is the expected end-user interface here? | 14:45 |
| dprince | mandre: if you expect the end-user to be able to run 'ansible-playbook ....' then I think all of this needs to be a client package | 14:46 |
| dprince | mandre: which tripleo-common is not | 14:46 |
| dprince | mandre: triplo common is helping to create workflows that are presented via an API | 14:46 |
| *** jpena|brb is now known as jpena | 14:47 | |
| mandre | dprince: so, I see 2 consumers for the validations | 14:47 |
| mandre | first one is tripleo-common and the mistral validation workflow | 14:47 |
| dprince | mandre: I didn't say we wouldn't (optionally) install the client package on the undercloud | 14:48 |
| mandre | second is end-user running the validation directly via ansible | 14:48 |
| dprince | mandre: we just don't require it | 14:48 |
| dprince | mandre: build it as a client package, and we can use it anywhere | 14:48 |
| dprince | mandre: because that is what your use cases tell me | 14:48 |
| mandre | dprince: yeah, I understand | 14:48 |
| dprince | mandre: and FWIW this lines up w/ how Mistral works anyways. It uses for example client libraries... like python-novaclient, python-ironicclient to access those repsective projects | 14:49 |
| dprince | mandre: so we do install those on the undercloud too, but that bit is transparent to the end user who just uses the API | 14:49 |
| mandre | dprince: I'd like to get shardy's opinion on creating a new client package, it seemed to me he really didn't want to go down that path | 14:50 |
| sshnaidm | trown, hi | 14:50 |
| sshnaidm | trown, can I know from the running script in quickstart which topology is being installed now? | 14:50 |
| dprince | mandre: sure, NP. We should definately get agreement | 14:50 |
| mandre | dprince: at least we agree instack-undercloud is not the right place | 14:51 |
| dprince | mandre: sorry if I miscommunicated anything before. This integration is a "fun" one :) | 14:51 |
| ayoung | larsks, OK, finally got it up (and dealt with other issues). The controller has the IP address 192.0.2.8 from neutron, and also 172.16.2.7 from vlan10 | 14:51 |
| openstackgerrit | Marios Andreou proposed openstack/puppet-tripleo: Adds the puppet-tripleo manifests for manila https://review.openstack.org/313527 | 14:51 |
| ayoung | er...something | 14:51 |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Composable manila tripleo-heat-templates side https://review.openstack.org/315658 | 14:51 |
| ayoung | there are 5 vlans | 14:51 |
| mandre | dprince: don't worry, there was no miscommunication on your side :) | 14:51 |
| *** penick has quit IRC | 14:51 | |
| larsks | ayoung: okay. So if you attach your host to the overcloud bridge and give it a 192.0.2.0/24 address, you should be all set, right? | 14:52 |
| ayoung | let me try that | 14:52 |
| ayoung | larsks, its not on the overcloud bridge per-se | 14:52 |
| ayoung | If I dump XML | 14:53 |
| trown | sshnaidm: which running script? | 14:53 |
| ayoung | I see brovc and brext | 14:53 |
| ayoung | so the domain knows about it... | 14:53 |
| trown | sshnaidm: you can definitely tell if HA is being used in the deploy script, by the pacemaker environment being passed in | 14:53 |
| larsks | ayoung: right. "brovc" == "the overcloud bridge" | 14:53 |
| ayoung | larsks, but inside the VM I don't see it | 14:53 |
| ayoung | I see | 14:53 |
| larsks | Just like the compute and controller hosts. | 14:54 |
| ayoung | http://paste.openstack.org/show/497733/ | 14:54 |
| jasond | can anybody tell me what this template language is? doesn't appear to be jinja https://github.com/openstack/tripleo-image-elements/blob/master/elements/os-collect-config/os-apply-config/etc/os-collect-config.conf | 14:54 |
| ayoung | that is what lead me down the path to madness before | 14:54 |
| *** weshay_mtg is now known as weshay | 14:54 | |
| sshnaidm | trown, I thnk about skipfile for tempest, it could be different depending on topology, I'd like to know if it's HA when I generate it | 14:54 |
| ayoung | so I need to somehow make the vm know about the bridge... | 14:54 |
| trown | sshnaidm: why would the skipfile be different for HA vs nonHA? | 14:55 |
| larsks | ayoung: I'm a little confused. I was suggesting that you *create* the vm just like the compute/controller hosts, so that it has an interface attached to brovc. | 14:55 |
| larsks | The vm doesn't need to "know about the bridge" otherwise. | 14:55 |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add Cinder API/Scheduler/Volume roles https://review.openstack.org/310939 | 14:55 |
| trown | sshnaidm: but you could check the length on "overcloud_nodes" var to see how many nodes there are | 14:55 |
| *** ibravo has joined #tripleo | 14:55 | |
| sshnaidm | trown, there is bug about identity v3 tests iirc, they fail on ha | 14:56 |
| *** dprince has quit IRC | 14:56 | |
| *** julim has quit IRC | 14:56 | |
| sshnaidm | trown, ok | 14:56 |
| ayoung | larsks, so I started with that, but then the Identity VM is not exposed to the outside world, and I have the same connectivity issue from there. I was hoping to mimic what the undercloud node was doing instead | 14:56 |
| *** dprince has joined #tripleo | 14:56 | |
| *** julim has joined #tripleo | 14:56 | |
| sshnaidm | trown, btw, do you want to run on quickstart the upstream version or installed from rdo is enough? | 14:57 |
| larsks | Okay, so you have two interfaces, then, but you still have one connected to brovc for interacting with the undercloud/overcloud. | 14:58 |
| larsks | Your second interface can be for whatever external connectivity you need. | 14:58 |
| larsks | It still seems very simple. | 14:58 |
| larsks | You don't need to create any bridges or anything inside the vm. | 14:58 |
| openstackgerrit | Merged openstack/tripleo-ui: Add missing key, remove redundant div in DeploymentPlan https://review.openstack.org/315137 | 14:58 |
| trown | sshnaidm: I would like the rdo version and upstream version to be the same like everything else in RDO ;p, but packaged version is what we want in quickstart | 14:59 |
| *** jaosorior has quit IRC | 14:59 | |
| *** jaosorior has joined #tripleo | 15:00 | |
| openstackgerrit | Merged openstack/tripleo-ui: Fix the small × close icons on plan modals https://review.openstack.org/317247 | 15:00 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/puppet-tripleo: Redirect to https for horizon https://review.openstack.org/318725 | 15:02 |
| *** jaosorior has quit IRC | 15:03 | |
| *** diazjf has joined #tripleo | 15:03 | |
| *** jaosorior has joined #tripleo | 15:03 | |
| openstackgerrit | Merged openstack/tripleo-ui: Nodes Registration https://review.openstack.org/315136 | 15:06 |
| *** diazjf1 has joined #tripleo | 15:07 | |
| *** shardy has quit IRC | 15:07 | |
| *** tremble has quit IRC | 15:09 | |
| *** jaosorior has quit IRC | 15:10 | |
| *** diazjf has quit IRC | 15:10 | |
| *** links has joined #tripleo | 15:11 | |
| *** zeroshft has quit IRC | 15:11 | |
| *** apetrich has quit IRC | 15:12 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge. https://review.openstack.org/310713 | 15:12 |
| jslim | ayoung: I was told by a little birdy that you had some issues getting your undercloud dashboard up after deploying with Quickstart. Can I ask if you decided to use IPTables or SSH to finally make it accessible and if there was anything you had to do outside of forwarding the traffic i.e. configuration etc.? I am doing the same thing and just keep getting 404 because I don't have any actual dashboard vhost configuration? | 15:13 |
| ayoung | jslim, So, I have to admit, I did not look at the undercloud dash, only the overcloud | 15:14 |
| jslim | ayoung: ah okay | 15:14 |
| ayoung | jslim, I used sshuttle, which is OK, but single user | 15:14 |
| ayoung | the same approach should work for undercloud, but different subnet? | 15:14 |
| trown | jslim I have never used the undercloud dashboard, and can not even confirm if it is packaged/installed in RDO | 15:15 |
| jslim | ayoung: are most folks not even bothering with a dashboard for undercloud ? | 15:15 |
| jslim | ayoung/trown: it's all there and I got the login page once but after re-running quickstart.sh it's broken at least it seems that way.... | 15:16 |
| jslim | glad I asked | 15:17 |
| trown | jslim: ya unless you are trying to test the undercloud GUI, it is really not all that useful in quickstart... everything is already prescribed via the config passed to quickstart | 15:17 |
| jslim | ah, i was under the impression it would somehow be useful for deploying overcloud VMs.... | 15:19 |
| *** jtomasek_ has quit IRC | 15:19 | |
| ayoung | jslim, I'm in another convo atm. Will be back focused here after | 15:19 |
| jslim | np, thanks guys | 15:20 |
| *** yamahata has quit IRC | 15:20 | |
| *** fultonj has quit IRC | 15:24 | |
| *** jasond has left #tripleo | 15:24 | |
| *** links has quit IRC | 15:26 | |
| *** akshai has quit IRC | 15:27 | |
| openstackgerrit | Bob Callaway proposed openstack/tripleo-heat-templates: Enable Neutron LBaaS Integration https://review.openstack.org/313933 | 15:29 |
| *** sthillma has joined #tripleo | 15:30 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/tripleo-ci: Delete more things to make room for centos-7 https://review.openstack.org/318743 | 15:31 |
| *** saneax is now known as saneax_AFK | 15:33 | |
| *** dmk0202 has quit IRC | 15:33 | |
| *** sthillma_ has joined #tripleo | 15:33 | |
| *** akshai has joined #tripleo | 15:34 | |
| *** sthillma has quit IRC | 15:36 | |
| *** sthillma_ is now known as sthillma | 15:36 | |
| jistr | marios, social: FYI i updated the whiteboard of https://blueprints.launchpad.net/tripleo/+spec/overcloud-upgrades-per-service with some semi-answers. Perhaps on our sync up call tomorrow we can chat about the options we have there etc. | 15:39 |
| *** dprince has quit IRC | 15:40 | |
| *** dprince has joined #tripleo | 15:41 | |
| marios | jistr: ack, thanks on a call now definite ack on catchup tomorrow | 15:43 |
| pabelanger | Wouldn't mind a review on diskimage-builder: https://review.openstack.org/#/c/318400/ | 15:46 |
| pabelanger | fixes hardcoded ext4 values | 15:46 |
| openstackgerrit | Paul Belanger proposed openstack-infra/tripleo-ci: Delete more things to make room for centos-7 https://review.openstack.org/318743 | 15:47 |
| openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart: WIP: Tempest support for quickstart https://review.openstack.org/318751 | 15:50 |
| *** florianf has quit IRC | 15:52 | |
| *** mikelk has quit IRC | 15:54 | |
| *** trozet has quit IRC | 15:56 | |
| *** dprince has quit IRC | 15:58 | |
| *** mcornea has quit IRC | 15:58 | |
| *** trozet has joined #tripleo | 15:59 | |
| *** dprince has joined #tripleo | 15:59 | |
| *** mcornea has joined #tripleo | 15:59 | |
| *** florianf has joined #tripleo | 16:01 | |
| *** penick has joined #tripleo | 16:02 | |
| *** julim has quit IRC | 16:02 | |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: Add timeout to get_host_info task https://review.openstack.org/317937 | 16:03 |
| *** marcusvrn_ has quit IRC | 16:03 | |
| *** ebarrera has quit IRC | 16:04 | |
| *** rcernin has quit IRC | 16:04 | |
| *** zoli|wfh is now known as zoli|gone | 16:09 | |
| *** veteran has quit IRC | 16:09 | |
| *** zoli|gone is now known as zoli_gone-proxy | 16:09 | |
| *** abehl has quit IRC | 16:09 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add Cinder API/Scheduler/Volume roles https://review.openstack.org/310939 | 16:11 |
| openstackgerrit | Dmitry Tantsur proposed openstack/python-tripleoclient: [WIP] Allow 'baremetal import' command to also start introspection https://review.openstack.org/299388 | 16:14 |
| openstackgerrit | Dmitry Ilyin proposed openstack/puppet-pacemaker: Merge with fuel-infra/puppet-pacemaker https://review.openstack.org/296440 | 16:16 |
| *** dprince has quit IRC | 16:16 | |
| trown | larsks: I had one question on https://review.openstack.org/#/c/316264 but I think it is good to go | 16:17 |
| larsks | trown: I wish we could run a test matrix, with both default and non default values for ha and non ha configs with and without network isolation. But I suspect that would limit us to, like, one test run per week. | 16:18 |
| openstackgerrit | Merged openstack/tripleo-quickstart: fix network configuration https://review.openstack.org/316264 | 16:18 |
| larsks | trown: but maybe using non-default values in ci would be a good place to start. Unless we change the default values... | 16:19 |
| trown | larsks: ya it is also tough, because minimal.yml is the default for all intents and purposes | 16:20 |
| trown | so by changing the default there, we are really changing the default for all new users | 16:20 |
| *** dtantsur is now known as dtantsur|afk | 16:21 | |
| trown | I think we need a job that tests ceph anyways, and I am hesitant to add that for new users (just one extra thing that could break), so maybe we could make the ceph job also change a bunch of defaults that we are concerned might get secretly hardcoded | 16:21 |
| trown | that way we test defaults (minimal.yml), and non-defaults (something-else.yml) | 16:22 |
| jslim | trown: do you think it would be wise do use quickstart.sh with Mitaka to install the deps necessary before following the howto at https://remote-lab.net/rdo-manager-ha-openstack-deployment or should I just follow the upstream documentation for TripleO like the howto refers to ? | 16:22 |
| *** yamahata has joined #tripleo | 16:22 | |
| larsks | trown: Would it make sense to set up some sort of periodic job that would test minimal + an override config? It wouldn't run on every change, but it would maybe help spot problems. | 16:23 |
| openstackgerrit | Dmitry Ilyin proposed openstack/puppet-pacemaker: Merge with fuel-infra/puppet-pacemaker https://review.openstack.org/296440 | 16:23 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: Required change to adjust to openstack CI. https://review.openstack.org/318719 | 16:26 |
| *** jcoufal has quit IRC | 16:27 | |
| *** xinwu has joined #tripleo | 16:27 | |
| trown | larsks: sorry heavily parallelized today... ya I would like to get some periodic jobs setup as well | 16:29 |
| *** absubram has quit IRC | 16:31 | |
| *** xinwu has quit IRC | 16:32 | |
| *** ebarrera has joined #tripleo | 16:33 | |
| *** paramite|afk is now known as paramite | 16:34 | |
| *** dprince has joined #tripleo | 16:36 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: Required change to adjust to openstack CI. https://review.openstack.org/318719 | 16:38 |
| *** akshai has quit IRC | 16:40 | |
| *** ifarkas has quit IRC | 16:41 | |
| *** tremble has joined #tripleo | 16:45 | |
| *** tremble has quit IRC | 16:46 | |
| *** nico_auv has quit IRC | 16:46 | |
| *** panda has quit IRC | 16:48 | |
| *** panda has joined #tripleo | 16:49 | |
| *** sambetts is now known as sambetts|afk | 16:52 | |
| *** penick has quit IRC | 16:54 | |
| *** penick has joined #tripleo | 16:54 | |
| *** electrofelix has quit IRC | 16:55 | |
| *** penick has quit IRC | 16:55 | |
| *** rasca has quit IRC | 16:55 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: Required change to adjust to openstack CI. https://review.openstack.org/318719 | 16:55 |
| *** derekh has quit IRC | 16:55 | |
| openstackgerrit | Merged openstack/tripleo-ui: Split patternfly-additions import https://review.openstack.org/318696 | 16:57 |
| *** sthillma has quit IRC | 16:58 | |
| *** yamahata has quit IRC | 16:58 | |
| *** yamahata has joined #tripleo | 16:58 | |
| *** trown is now known as trown|lunch | 16:59 | |
| Erming_ | larsks: I deployed the overcloud, got one error on postdeployment: | resource_status_reason | Error: resources.ControllerNodesPostDeployment.resources.ControllerOvercloudServicesDeployment_Step6.resources[2]: Deployment to server failed: deploy_status_code: Deployment exited with non-zero status code: 6 | | 17:01 |
| *** penick has joined #tripleo | 17:01 | |
| Erming_ | larsks: which log you think I should check? | 17:01 |
| *** mcornea has quit IRC | 17:05 | |
| larsks | Erming_: probably start by investigating the system log on the controller; that may be due to a puppet failure. | 17:05 |
| *** florianf has quit IRC | 17:05 | |
| leifmadsen | for the tripleo-quickstart common values, I think this line might be wrong (and the following 2): https://github.com/openstack/tripleo-quickstart/blob/master/playbooks/roles/common/defaults/main.yml#L43 | 17:05 |
| leifmadsen | I believe default should be undercloud_memory, not default_memory? | 17:06 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: Required change to adjust to openstack CI. https://review.openstack.org/318791 | 17:06 |
| openstackgerrit | Honza Pokorny proposed openstack/tripleo-ui: Disable links when deploy is in progress https://review.openstack.org/312473 | 17:06 |
| *** lucasagomes is now known as lucas-afk | 17:06 | |
| ayoung | jslim, didn't forget about you, just been tied up since then. | 17:08 |
| jistr | Erming_: yea what larsks said. Also i think "resources[2]" in the failure message means that the error happened on overcloud-controller-2, and the that might be best way to find the puppet error could be `journalctl -u os-collect-config` | 17:09 |
| jslim | ayoung, no worries, thanks.... | 17:09 |
| larsks | yea what jistr said... :) | 17:09 |
| *** florianf has joined #tripleo | 17:10 | |
| *** julim has joined #tripleo | 17:10 | |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-pacemaker: Add other-requirements.txt for bindep https://review.openstack.org/318802 | 17:11 |
| openstackgerrit | Leif Madsen proposed openstack/tripleo-quickstart: Update undercloud default variable location https://review.openstack.org/318806 | 17:12 |
| *** akshai has joined #tripleo | 17:13 | |
| *** fragatina has joined #tripleo | 17:15 | |
| jslim | ayoung, going to lunch...be back later.... | 17:16 |
| *** fragatina has quit IRC | 17:16 | |
| *** fragatina has joined #tripleo | 17:17 | |
| *** rasca has joined #tripleo | 17:19 | |
| *** jpena is now known as jpena|off | 17:20 | |
| *** jslim has quit IRC | 17:20 | |
| *** ebarrera has quit IRC | 17:20 | |
| *** devvesa has quit IRC | 17:27 | |
| *** pkovar has quit IRC | 17:28 | |
| *** dmsimard is now known as dmsimard|afk | 17:39 | |
| Erming_ | jistr: Thank you and larsks. Sounds awesome for the resource [2] explaination. I checked the journal but don't see anything special. Just this: No local metadata found (['/var/lib/os-collect-config/local-data']) I guess it's not | 17:42 |
| Erming_ | an real issue. then I checked the var/log/message on controller 2. Only found something like: warning: Processing failed op monitor for openstack-gnocchi-metricd:0 on overcloud-controller-0: not running (7) | 17:44 |
| *** paramite is now known as paramite|afk | 17:44 | |
| larsks | Erming_: let's step back and take a look at what heat is telling us. Can you pastebin the output of 'heat resource-list -n5 overcloud'? | 17:44 |
| Erming_ | larsks: from undercloud? | 17:45 |
| *** apetrich has joined #tripleo | 17:45 | |
| larsks | Correct. | 17:45 |
| Erming_ | larsks: jistr: here you are: https://paste.fedoraproject.org/368597/46368002/ | 17:47 |
| *** akshai has quit IRC | 17:47 | |
| *** ohamada has quit IRC | 17:48 | |
| Erming_ | larsks: see the bottom line: | 2 | cf4e4f29-c0ef-415d-a286-8231a99e6db8 | OS::Heat::StructuredDeployment | CREATE_FAILED | 2016-05-17T23:28:28 | overcloud-ControllerNodesPostDeployment-ui2suxzxptit-ControllerOvercloudServicesDeployment_Step6-h7ozeig3utdr | | 17:49 |
| *** dmacpher has joined #tripleo | 17:51 | |
| larsks | Erming_: can you pastebin: heat resource-show overcloud-ControllerNodesPostDeployment-ui2suxzxptit-ControllerOvercloudServicesDeployment_Step6-h7ozeig3utdr 2 | 17:53 |
| ayoung | larsks, do we even run Horizon on the undercloud? | 17:54 |
| larsks | ayoung: I don't think we do. | 17:54 |
| *** akshai has joined #tripleo | 17:54 | |
| *** sthillma has joined #tripleo | 17:56 | |
| Erming_ | larsks: https://paste.fedoraproject.org/368598/63680570/ | 17:56 |
| larsks | Well, that doesn't help much. | 17:57 |
| larsks | How about pastebin the output of "journalctl" on controller 2? An easy way to do that is "journalctl | curl -s -T- chunk.io" | 17:59 |
| EmilienM | the haproxy roles are ready, can we review and eventually land https://review.openstack.org/#/c/313029/ and https://review.openstack.org/#/c/313026/ please? (already +2 and CI passing) | 18:05 |
| *** trown|lunch is now known as trown | 18:07 | |
| *** TSCHAK has joined #tripleo | 18:09 | |
| *** mbound has quit IRC | 18:14 | |
| Erming_ | larsks: sorry I didn't notice your reply. this is what you needed: http://chunk.io/f/c851ca4ac6964c0c9588639471f29c24 | 18:15 |
| Erming_ | larsks: looks ceph issue? as I read: error connecting to the cluster | 18:17 |
| openstackgerrit | Dan Sneddon proposed openstack/tripleo-heat-templates: Updates and fixes to NIC config template routes https://review.openstack.org/318389 | 18:17 |
| *** dprince has quit IRC | 18:18 | |
| larsks | Erming_: oddly enough, that log seems to start too late; it doesn't include the logs from the puppet installation process. | 18:21 |
| larsks | So it doesn't have the information to determine why the deploy failed. | 18:21 |
| larsks | I suspect there must have been a log rotation or something. | 18:21 |
| openstackgerrit | Ryan Brady proposed openstack/tripleo-common: Adds Plan List Action https://review.openstack.org/300221 | 18:22 |
| larsks | Erming_: See if you have a /var/log/messages* file that contains the line "writing /etc/puppet/hiera.yaml", and paste that one. | 18:22 |
| openstackgerrit | Dan Sneddon proposed openstack/tripleo-heat-templates: Updates and fixes to NIC config template routes https://review.openstack.org/318389 | 18:24 |
| *** noslzzp has quit IRC | 18:26 | |
| *** zaneb has quit IRC | 18:26 | |
| *** pabelanger has quit IRC | 18:26 | |
| *** jayg has quit IRC | 18:26 | |
| *** crinkle has quit IRC | 18:26 | |
| *** lifeless has quit IRC | 18:26 | |
| *** zoli_gone-proxy has quit IRC | 18:26 | |
| panda | trown: WDYT = what do you think ? | 18:26 |
| *** jayg has joined #tripleo | 18:26 | |
| *** lifeless has joined #tripleo | 18:26 | |
| *** crinkle has joined #tripleo | 18:27 | |
| *** zoli_gone-proxy has joined #tripleo | 18:27 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge. https://review.openstack.org/310713 | 18:27 |
| trown | panda: ack | 18:27 |
| Erming_ | larsks: Probably I report this issue too late as took a rest yesterday at home. I have learned your chunk.io way: http://chunk.io/f/a702b3336f354772a1924385f9e5300b | 18:27 |
| *** pabelanger has joined #tripleo | 18:28 | |
| *** zaneb has joined #tripleo | 18:29 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge. https://review.openstack.org/310713 | 18:29 |
| Erming_ | larsks: noticed "May 18 01:00:03 localhost galera(galera)[15643]: ERROR: MySQL is not running " by a very quick glance | 18:29 |
| *** rhefner has quit IRC | 18:30 | |
| *** igorbelikov has quit IRC | 18:30 | |
| openstackgerrit | Brad P. Crochet proposed openstack/puppet-tripleo: Add Sahara profiles https://review.openstack.org/318839 | 18:32 |
| *** igorbelikov has joined #tripleo | 18:33 | |
| jistr | Erming_: also i can see a couple of errors like Error: /Stage[main]/Nova::Cert/Nova::Generic_service[cert]/Service[nova-cert]: Could not evaluate: Cannot allocate memory - fork(2) | 18:34 |
| *** rhefner has joined #tripleo | 18:34 | |
| larsks | jistr: oh, good catch. I missed that. | 18:34 |
| openstackgerrit | Brad P. Crochet proposed openstack/tripleo-heat-templates: Composable Sahara services https://review.openstack.org/318840 | 18:34 |
| jistr | this could be the core issue -- needing more RAM on the overcloud nodes perhaps | 18:34 |
| larsks | There are definitely post-deploy errors, e.g., gnocchi can't connect to ceph, but those are probably symptoms rather than the cause. | 18:34 |
| *** jprovazn has quit IRC | 18:35 | |
| Erming_ | jistr: larsks: Does it mean out-of-memory issue | 18:35 |
| openstackgerrit | Brad P. Crochet proposed openstack/tripleo-heat-templates: Composable Sahara services https://review.openstack.org/318840 | 18:35 |
| larsks | The errors that jistr pointed out mean out-of-memory, yes. | 18:35 |
| Erming_ | I have 128GB in the host | 18:35 |
| Erming_ | the total allocate to undercloud/overcloud is about 40GB | 18:35 |
| larsks | Doesn't matter how much host memory you have; how much is allocated to the controller vms? You' | 18:35 |
| Erming_ | let me show you the ha.yml | 18:36 |
| Erming_ | https://paste.fedoraproject.org/368612/83003146/ | 18:36 |
| *** akrivoka has quit IRC | 18:37 | |
| larsks | So you're only allocating 6GB to each controller. Make that bigger and see what happens. | 18:37 |
| Erming_ | larsks: It's the default one. Isn't it your ha.yml from github? | 18:37 |
| larsks | It is the default, yes, but that's designed to try to fit in a 32GB machine so is not necessarily optimal. | 18:38 |
| Erming_ | I will try right away. Just wondering why works for you but not for me always. | 18:38 |
| Erming_ | how much you think is good for mine each? | 18:38 |
| larsks | I don't know that it works or not; I'm generally trying out the minimal install myself, and I know the ha CI jobs are often problematic. | 18:39 |
| larsks | trown: you around? Any thoughts on ^^^^? | 18:39 |
| Erming_ | larsks: trown: from which step I should re-try? Change the ha.yml and then re-deploy undercloud? | 18:40 |
| larsks | Erming_: If you're trying to minimize things, maybe set it to 8GB and try again. If you've got a lot of headroom and spare the extra memory, just make it bigger (12GB?). You can always try turning it down later if that turns out to solve the problem. | 18:40 |
| trown | hmm... but the HA jobs are not failing because of a lack of memory | 18:40 |
| larsks | trown: Yeah, that's why I was asking you, because you've paid more attention to that :) | 18:40 |
| * jistr finished digging up the link | 18:41 | |
| jistr | https://github.com/openstack-infra/tripleo-ci/blob/master/test-environments/worker-config.yaml | 18:41 |
| jistr | this is what we use in CI to downsize the overcloud a bit | 18:41 |
| trown | most of the time they fail due to timeouts, mostly because of the nova/ironic race | 18:41 |
| jistr | the defaults are geared more towards production | 18:41 |
| trown | jistr: ya that is baked into quickstart | 18:41 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: Merge with fuel-infra/puppet-pacemaker https://review.openstack.org/296440 | 18:41 |
| larsks | Erming_: I would just crank control_memory up to at least 8GB and try again, unless trown has another suggestion. | 18:41 |
| jistr | trown: ah didn't know that. So this won't be the cause. hmm... | 18:41 |
| * larsks has to step away for a bit. | 18:42 | |
| Erming_ | larsks: OK. let me try with 12GB servers. Thanks! | 18:42 |
| *** jistr has quit IRC | 18:43 | |
| trown | ya, I don't have a better suggestion... although it does lead me to a question... Erming_ did you do the default quickstart run, and then follow docs for the rest? | 18:43 |
| Erming_ | trown: sure. as we discussed. I just passed -c ha.yml | 18:43 |
| trown | Erming_: right and then you didnt use the scripts provided, and followed docs? | 18:44 |
| Erming_ | which scripts? | 18:44 |
| trown | ah ok, then you don't have https://github.com/openstack-infra/tripleo-ci/blob/master/test-environments/worker-config.yaml which is what is wrong | 18:44 |
| trown | Erming_: quickstart drops scripts to finish the deploy in /home/stack on the undercloud | 18:45 |
| Erming_ | trown: I never known about this yaml | 18:45 |
| trown | Erming_: ya it is a CI thing to make it possible to deploy HA with limited hardware | 18:45 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: Merge with fuel-infra/puppet-pacemaker https://review.openstack.org/296440 | 18:45 |
| Erming_ | trown: what I should do now? | 18:45 |
| trown | it is included in the overcloud-deploy script that quickstart creates, but if you just follow docs you would not have it (or even see mention of it) | 18:46 |
| *** rcernin has joined #tripleo | 18:46 | |
| Erming_ | trown: I see. I should have run the individual scripts after undercloud right? | 18:46 |
| trown | Erming_: you could just `heat stack-delete overcloud` and then `/home/stack/overcloud-deploy.sh` and it will likely "just work" | 18:46 |
| trown | actually in betwwen, you need to `watch heat stack-list` until it is empty | 18:47 |
| leifmadsen | you normally need to ironic node-delete too right? | 18:47 |
| leifmadsen | oh, maybe not in this case if things didn't complete | 18:47 |
| Erming_ | trown: really. how about the rest scripts, like overcloud-deploy-post.sh and overcloud-validate.sh | 18:47 |
| trown | leifmadsen: no that would be a bug... there was a bug like that in Kilo timeframe, but I have not seen anything like that in a while | 18:47 |
| trown | Erming_: you can run those if you want, but after overcloud-deploy you would have a cloud up | 18:48 |
| leifmadsen | trown: ah ok, I ran into that with OSP-d 7 (Kilo I think), and haven't adjusted my workflow for mitaka :) | 18:48 |
| Erming_ | trown: leifmadsen: now I have a cloud up with HA. Just reports the post deployment error | 18:48 |
| trown | Erming_: post deployment is a bit tricky term in tripleo-heat-templates | 18:49 |
| Erming_ | so I am wondering if deploy-post is necessary. and should I enlarge the mem at first | 18:49 |
| leifmadsen | HA test is next, I'm just doing minimal right now for some heat-engine performance tests | 18:49 |
| trown | Erming_: it is refering to ironic node provisioning as "deployment" so post deployment is really configuration and service starting | 18:49 |
| trown | Erming_: if the overcloud deploy fails, you do not have a cloud :( | 18:50 |
| Erming_ | trown: but why the memory issue reported and MySQL not started | 18:50 |
| trown | Erming_: because the HA config in tripleo-quickstart is very small memory for CI purposes, and therefore relies on that heat-environment to reduce memory consumption. | 18:51 |
| *** florianf has quit IRC | 18:52 | |
| Erming_ | trown: OK. then. let me try your solution, if it doesn't work I enlarge mem and start from scratch. thanks! | 18:52 |
| trown | I think there is a bug there, in that a valid "Choose your own adventure" path will result in failure, but I am not sure how to solve it | 18:52 |
| Erming_ | trown: one more concern. after a overcloud deployed in the virt environment, can I use the 3 controller environment to deploy/add physical compute nodes in the baremetal way. | 18:54 |
| Erming_ | trown: hope you can follow what I mean | 18:54 |
| trown | Erming_: it is technically possible... you just need the physical nodes to be on the provisioning network of the virt undercloud | 18:55 |
| *** florianf has joined #tripleo | 18:55 | |
| trown | that can be tricky though | 18:55 |
| Erming_ | trown: I can imagine. Can I know a bit about how to add compute nodes to an existing env? | 18:56 |
| Erming_ | physical ones | 18:56 |
| *** mbound has joined #tripleo | 18:56 | |
| trown | Erming_: you would just create an instackenv.json with there power management details and register them to Ironic just like virt nodes | 18:57 |
| trown | s/there/their/ | 18:57 |
| Erming_ | trown: makes sense. Last time I tried add virt nodes but failed. as I told you. | 18:57 |
| Erming_ | failed to do introspection. maybe not necessary. I can try again. don't worry. | 18:58 |
| Erming_ | if the path is correct | 18:58 |
| trown | like I said the tricky part is making sure those baremetal nodes can PXE boot from the underclouds provisioning network | 18:58 |
| Erming_ | OK! | 18:58 |
| trown | Erming_: well it failed with virt, because there were no actual virtual machines being registered to Ironic :P | 18:58 |
| Erming_ | trown: Ah. you mean for physical nodes, we don't need this step? | 18:59 |
| trown | Erming_: for physical nodes there is nothting to create...they were created in a factory at some point. so we just need to register them to ironic | 19:00 |
| trown | for virts though we need to create them before registering to ironic | 19:00 |
| Erming_ | trown: understood! | 19:00 |
| *** florianf has quit IRC | 19:02 | |
| *** liverpooler has joined #tripleo | 19:09 | |
| openstackgerrit | Jason E. Rist proposed openstack/tripleo-ui: Refactor notification for better display https://review.openstack.org/316874 | 19:30 |
| EmilienM | wow, gate-tripleo-ci-f22-nonha in 1h10 | 19:32 |
| *** mbound has quit IRC | 19:36 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-docs: Composable roles within services Tutorial https://review.openstack.org/311512 | 19:37 |
| *** dsariel has quit IRC | 19:38 | |
| *** mbound has joined #tripleo | 19:38 | |
| leifmadsen | much CI! | 19:41 |
| openstackgerrit | Brad P. Crochet proposed openstack/puppet-tripleo: Add Heat profiles https://review.openstack.org/313019 | 19:44 |
| *** fultonj has joined #tripleo | 19:46 | |
| *** athomas has quit IRC | 19:47 | |
| *** chem` has joined #tripleo | 19:48 | |
| *** chem has quit IRC | 19:50 | |
| *** rhallisey has quit IRC | 20:03 | |
| openstackgerrit | Jason E. Rist proposed openstack/tripleo-ui: Refactor notification for better display https://review.openstack.org/316874 | 20:05 |
| *** rhallisey has joined #tripleo | 20:05 | |
| slagle | has anyone started looking into the recurring "No Valid Hosts" errors in CI? | 20:06 |
| slagle | it seems to be the RamFilter that is disqualifying hosts | 20:06 |
| trown | that can be a symptom of the scheduler race... since there is not enough RAM on the host because it already has an instance scheduled | 20:08 |
| trown | or are you talking about some newish issue? | 20:09 |
| openstackgerrit | Merged openstack/puppet-tripleo: Add loadbalancer profile for ha & non-ha https://review.openstack.org/313026 | 20:09 |
| EmilienM | slagle: is it implemented this way ? https://review.openstack.org/#/c/288188/ | 20:09 |
| slagle | trown: not sure honestly | 20:10 |
| *** xenthree3 has joined #tripleo | 20:11 | |
| *** xenthree3 has left #tripleo | 20:12 | |
| slagle | actually, maybe it's related to my patches | 20:13 |
| ayoung | larsks, trown got it!. Blog post incipient with details | 20:14 |
| larsks | ayoung: \o/ | 20:15 |
| trown | slagle: I was also seeing that when testing out the ironic packaging... when in reality Ironic was just failing to deploy, but it somehow bubbled up as a nova no valid hosts | 20:15 |
| openstackgerrit | Tzu-Mainn Chen proposed openstack/tripleo-common: Add TripleO Mistral documentation https://review.openstack.org/313109 | 20:16 |
| trown | nice one ayoung | 20:16 |
| slagle | trown: yea i think it might just be a symptom of nova trying to reschedule, and everything is in use | 20:17 |
| slagle | trown: yea, it was ironic failing to deploy | 20:19 |
| slagle | i see the problem with my patch, it's assigning br-ctlplane 192.0.2.1/32 instead of 192.0.2.1/24 | 20:19 |
| trown | ah yep | 20:20 |
| *** akshai has quit IRC | 20:22 | |
| *** morazi has quit IRC | 20:23 | |
| *** ayoung has quit IRC | 20:24 | |
| openstackgerrit | James Slagle proposed openstack/instack-undercloud: Consolidate undercloud* elements https://review.openstack.org/315096 | 20:24 |
| openstackgerrit | James Slagle proposed openstack/instack-undercloud: Add timestamps to puppet apply output https://review.openstack.org/315128 | 20:24 |
| openstackgerrit | James Slagle proposed openstack/instack-undercloud: Minimize element usage https://review.openstack.org/315099 | 20:24 |
| openstackgerrit | James Slagle proposed openstack/instack-undercloud: Move 02-undercloud-stack-heat-metadata to python module https://review.openstack.org/314219 | 20:24 |
| openstackgerrit | James Slagle proposed openstack/instack-undercloud: Remove truncate nova config workaround https://review.openstack.org/315184 | 20:24 |
| openstackgerrit | James Slagle proposed openstack/instack-undercloud: Don't use os-net-config element https://review.openstack.org/315095 | 20:24 |
| pradk | slagle, i did a quick test with liberty overcloud deploy and installing aodh manifest alongside + turning off ceilo-alarms .. services come up fine, seems to work | 20:26 |
| pradk | slagle, if you can show me an example of how to put together a heat software deployment to integrate a puppet manifest I can try to come up with something | 20:27 |
| *** jayg is now known as jayg|g0n3 | 20:28 | |
| slagle | pradk: there are some examples in tripleo-heat-templates already, but this might be easier to understand: https://github.com/openstack/heat-templates/blob/master/hot/software-config/example-templates/example-puppet-template.yaml | 20:29 |
| slagle | pradk: so the main thing to notice is the "config" resource there that has group: puppet, and is using get_file to reference the manifest | 20:29 |
| pradk | ok cool so that file would be my puppet manifest aodh.pp or whatever? | 20:30 |
| slagle | yep | 20:30 |
| pradk | slagle, where in the upgrade sequence would this heat template sit? in major_upgrade_pacemaker.yaml ? | 20:31 |
| pradk | slagle, or keep it separate and pass it in with -e ? | 20:32 |
| *** bvandenh has quit IRC | 20:33 | |
| slagle | pradk: good question, i was thinking it might be a seperate step that no-ops the rest of the puppet from running, kind of like major_upgrade_pacemaker.yaml does | 20:34 |
| slagle | that way it could be run separate from the major upgrade | 20:34 |
| EmilienM | slagle: well, I saw the problem elsewhere too. (No Valid Hosts), I'll monitor it next time | 20:34 |
| rwsu | rbrady: hi, out of curiosity, is tripleo.create_container supposed to to accept a json file as input containing the name of the container? when I specify the container name as input it doesn't work | 20:35 |
| slagle | pradk: i'll have to think on that, and chat with jistr | 20:35 |
| slagle | pradk: maybe it's not a good idea to have it seperate, b/c then people could just update to aodh, and then they'd have mixed liberty/mitaka | 20:36 |
| pradk | slagle, cool ok, i'll play around with the template examples for now.. yea we can chat with jistr tomorrow when he's around | 20:36 |
| *** dmk0202 has joined #tripleo | 20:39 | |
| *** rdopiera has quit IRC | 20:40 | |
| *** julim has quit IRC | 20:42 | |
| *** rdopiera has joined #tripleo | 20:42 | |
| *** julim has joined #tripleo | 20:42 | |
| *** fragatina has quit IRC | 20:43 | |
| *** panda has quit IRC | 20:48 | |
| *** panda has joined #tripleo | 20:49 | |
| *** rdopiera has quit IRC | 20:50 | |
| *** rdopiera has joined #tripleo | 20:51 | |
| *** morazi has joined #tripleo | 20:51 | |
| *** lblanchard has quit IRC | 20:58 | |
| *** chem` has quit IRC | 21:02 | |
| *** rhallisey has quit IRC | 21:03 | |
| *** fragatina has joined #tripleo | 21:03 | |
| trozet | slagle, dsneddon, EmilienM: would tripleO accept a patch that allows disabling of Sahara? I know composable services is coming in master, but this would be for stable/mitaka. | 21:04 |
| *** trown is now known as trown|outtypewww | 21:09 | |
| *** weshay has quit IRC | 21:10 | |
| *** yamahata has quit IRC | 21:11 | |
| *** morazi_ has joined #tripleo | 21:12 | |
| *** morazi has quit IRC | 21:16 | |
| *** julim has quit IRC | 21:20 | |
| *** dmk0202 has quit IRC | 21:22 | |
| *** chem` has joined #tripleo | 21:34 | |
| openstackgerrit | Emilien Macchi proposed openstack/instack-undercloud: Cleanup hieradata to reduce Puppet warnings https://review.openstack.org/318956 | 21:39 |
| EmilienM | trozet: yes, probably | 21:41 |
| trozet | EmilienM: should i directly commit it to stable/mitaka? is that allowed in tht, or does it need to go to master then backported? | 21:42 |
| EmilienM | master then backport | 21:42 |
| trozet | EmilienM: thanks | 21:42 |
| EmilienM | cool | 21:42 |
| openstackgerrit | Jason E. Rist proposed openstack/tripleo-ui: Refactor notification for better display https://review.openstack.org/316874 | 21:43 |
| *** shadower has quit IRC | 21:44 | |
| *** shadower has joined #tripleo | 21:45 | |
| *** morazi_ has quit IRC | 21:48 | |
| slagle | EmilienM: trozet : i actually dont think we'd allow such a backport unless it's solving a bug | 21:56 |
| EmilienM | oops ok | 21:56 |
| trozet | slagle: but what about customers who don't want Sahara, which is the most memory consuming service we see in our deployment? | 21:57 |
| trozet | slagle: there is no mechanism to turn it off... | 21:57 |
| *** apetrich has quit IRC | 21:57 | |
| slagle | i know, i just dont think "cant disable sahara" meets with the stable backport policy we are trying to adhere to | 21:58 |
| slagle | but you can make the case for it :) | 21:58 |
| slagle | i'm not trying to shut it down or anything | 21:58 |
| slagle | just my initial reaction, based on the backport policy | 21:59 |
| trozet | slagle: ok, bug fixes only, i get it | 21:59 |
| trozet | slagle: question about undercloud, why does swift-proxy-service run there? | 21:59 |
| EmilienM | we use Swift for Glance backend | 22:00 |
| EmilienM | we store overcloud images in Swift, hosted on undercloud | 22:00 |
| trozet | EmilienM: what's the reason for using Swift? | 22:00 |
| EmilienM | Swift has a nice API that allow us to use it to store multiple artifacts, such images but we're also investigating the storage of more bits like puppet modules, etc | 22:01 |
| EmilienM | end of day here (midnight) - ttyl | 22:03 |
| trozet | EmilienM: thanks. later on | 22:03 |
| *** ramishra has quit IRC | 22:05 | |
| *** fultonj has quit IRC | 22:05 | |
| *** ramishra has joined #tripleo | 22:06 | |
| *** ayoung has joined #tripleo | 22:08 | |
| *** paramite|afk has quit IRC | 22:11 | |
| *** paramite|afk has joined #tripleo | 22:11 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Deploy Loadbalancer as a composable role https://review.openstack.org/313029 | 22:21 |
| *** rcernin has quit IRC | 22:23 | |
| *** dkelson has joined #tripleo | 22:25 | |
| *** paramite|afk is now known as paramite | 22:45 | |
| *** paramite has quit IRC | 22:45 | |
| *** fragatin_ has joined #tripleo | 22:49 | |
| *** fragatina has quit IRC | 22:52 | |
| *** diazjf1 has quit IRC | 22:54 | |
| rbrady | rwsu: I assume you are hitting the message timeout error? | 22:55 |
| ayoung | larsks, trown|outtypewww https://adam.younglogic.com/2016/05/reproducing-ovs/ | 22:55 |
| *** mbound has quit IRC | 22:59 | |
| rbrady | rwsu: https://bugs.launchpad.net/mistral/+bug/1581649 | 23:00 |
| openstack | Launchpad bug 1581649 in Mistral "Action Execution Response Timeout" [High,New] | 23:00 |
| rbrady | rwsu: if you follow those steps to disable httpd+wsgi for serving mistral and use eventlet for now it will work. I'm currently working on the bug for a fix | 23:01 |
| *** yolanda has quit IRC | 23:03 | |
| *** yolanda has joined #tripleo | 23:05 | |
| *** ibravo has quit IRC | 23:05 | |
| *** gfidente has quit IRC | 23:06 | |
| *** MaxPC has joined #tripleo | 23:06 | |
| *** MaxPC has quit IRC | 23:10 | |
| *** MaxPC has joined #tripleo | 23:12 | |
| rwsu | rbrady: ok, I'll give that a try, but my question above is on a different topic | 23:19 |
| rwsu | rbrady: is tripleo.create_container supposed to to accept a json file as input containing the name of the container? when I specify the container name as input it doesn't work | 23:19 |
| rbrady | rwsu: I just tested it and it worked. http://paste.openstack.org/show/497817/ | 23:20 |
| rbrady | rwsu: you can also put the json in a file and pass that to the mistral cli | 23:21 |
| *** ayoung has quit IRC | 23:22 | |
| rbrady | rwsu: if this doesn't help can you paste what you're doing and explain your undercloud age/config | 23:23 |
| rwsu | rbrady: ah ok, I see what I did wrong, thank you | 23:28 |
| rbrady | rwsu: ack | 23:37 |
| *** MaxPC has quit IRC | 23:40 | |
| *** ayoung has joined #tripleo | 23:49 | |
| *** MaxPC has joined #tripleo | 23:50 | |
| *** fragatin_ has quit IRC | 23:50 | |
| *** fragatina has joined #tripleo | 23:51 | |
| *** dkelson has quit IRC | 23:54 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!