Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2015-09-01

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:18 teran joined #fuel
00:21 teran_ joined #fuel
01:04 bookwar joined #fuel
02:10 tzn joined #fuel
02:12 tzn joined #fuel
02:34 _tzn joined #fuel
02:52 hakimo joined #fuel
03:28 gongysh joined #fuel
04:00 HeOS_ joined #fuel
04:25 tzn joined #fuel
05:03 xdeller joined #fuel
05:39 vsedelnik joined #fuel
06:27 stamak joined #fuel
06:40 mkwiek joined #fuel
06:55 vsedelnik joined #fuel
06:59 ddmitriev joined #fuel
07:05 vsedelnik joined #fuel
07:07 hyperbaba joined #fuel
07:12 JoeStack joined #fuel
07:18 pbrooko joined #fuel
07:32 aliemieshko_ joined #fuel
07:46 bazilza joined #fuel
07:53 stamak joined #fuel
08:08 pbrooko joined #fuel
08:13 e0ne joined #fuel
08:26 vsedelnik joined #fuel
08:33 HeOS joined #fuel
08:48 e0ne joined #fuel
08:54 tzn joined #fuel
09:06 gongysh joined #fuel
09:21 tzn joined #fuel
09:27 vsedelnik joined #fuel
09:34 mihag joined #fuel
10:07 vsedelnik joined #fuel
10:35 pal_bth joined #fuel
10:58 tzn joined #fuel
11:02 vsedelnik joined #fuel
11:13 dkusidlo joined #fuel
11:14 jaypipes joined #fuel
11:22 dkusidlo_ joined #fuel
11:24 JoeStack <xenolog> ping
11:34 pbrooko joined #fuel
11:39 ddmitriev1 joined #fuel
11:42 vsedelnik joined #fuel
11:48 vsedelnik joined #fuel
11:51 mihag joined #fuel
11:54 sergmelikyan joined #fuel
12:00 tzn joined #fuel
12:01 martineg_ any ideas what could cause a 500 error on instance creation, "Failed to allocate the network(s), not rescheduling." ? logs on the compute node shows VirtualInterfaceCreateException, and a "Timeout waiting for vif plugging callback .."
12:01 martineg_ (mos 6.0 ha-mode)
12:01 martineg_ (vif_plugging_timeout=300)
12:10 pbrooko joined #fuel
12:18 e0ne joined #fuel
12:25 dkusidlo joined #fuel
12:28 vsedelnik joined #fuel
13:00 tzn joined #fuel
13:04 samuelBartel joined #fuel
13:11 JoeStack <xenolog> update on https://bugs.launchpad.net/fuel/+bug/1488925
13:12 sergmelikyan joined #fuel
13:23 vsedelnik joined #fuel
13:38 vsedelnik joined #fuel
13:54 vsedelnik joined #fuel
13:54 sergmelikyan joined #fuel
13:58 pbrooko joined #fuel
14:03 pbrooko joined #fuel
14:06 julien_ZTE joined #fuel
14:09 ericjwolf joined #fuel
14:09 ericjwolf Greetings
14:09 ericjwolf is there any fuel documents on the details of the syntax of the deployment yaml files?
14:10 mwhahaha contents or sytnax
14:10 DevStok ERROR nova.api.openstack [req-d4b112c4-513f-43b4-9711-8cac93e25568 None] Caught error: Connection to neutron failed: Maximum attempts reached
14:10 DevStok how can i solve this issue?
14:10 ericjwolf I tried to modify the networking deployment section and after I deployed the node, no changes where actually done.
14:10 DevStok i cant reach the vm with floating IP
14:11 mwhahaha ericjwolf: so you manually modified the astute.yaml?
14:11 ericjwolf I guess really both.  My main concern is the syntax of the networking sections (network scheme, quantum settings, ..)
14:11 ericjwolf not the astute files.
14:11 ericjwolf the files when you do a fuel deployment --env x --default -- download
14:12 ericjwolf it creates a directory deployment_x in there are yaml files for each server.
14:14 angdraug joined #fuel
14:16 mwhahaha yea so now that i have that up, what are you trying to adjust?
14:17 ericjwolf a few place.  but on each compute, I want to add a new bridge that is going to be used by our app (directly connected to external networks).
14:19 mwhahaha so a new bridge would be under the network_scheme section
14:19 ericjwolf the bridge will be br-extnet and link off a bond0 device (the same way br-aux and pr-prv are).  so it is bond0 -> br-aux->br-extnet
14:19 mwhahaha wonder if we have that documented somewhere
14:19 mwhahaha i know that is xenolog's realm of expertise
14:20 ericjwolf I found like 4 places to modify it
14:20 mwhahaha oo it is in the docs
14:20 ericjwolf but after the deployment nothing got picked up.
14:20 mwhahaha did you upload it after you adjusted it?
14:20 ericjwolf so either I had a typo or when I uploaded it, I did it wrong.
14:21 mwhahaha https://docs.mirantis.com/fuel/fuel-master/reference-architecture.html#advanced-network-configuration-using-open-vswitch
14:22 ericjwolf That's what I did :  fuel --env 1 deployment --upload
14:23 ericjwolf one section I woundered about was the neutron roles section.
14:23 ericjwolf when I add a new bridge do I have to assign a role?
14:23 ericjwolf can I have multiple of the same type?
14:23 mwhahaha no i think it's a 1:1
14:23 dkusidlo joined #fuel
14:24 mwhahaha those are used within our puppet for determining what networks/ips a particular role has
14:24 mwhahaha we use it for figuring out ips for services
14:24 mwhahaha so if you are adding a bridge for yourself, unless you're creating a plugin that you want to query that info via puppet you don't need to touch that
14:25 ericjwolf that was my original thought.
14:25 ericjwolf so that did clear that up.
14:25 mwhahaha https://github.com/stackforge/fuel-library/blob/master/deployment/puppet/osnailyfacter/modular/openstack-controller/openstack-controller.pp#L29
14:25 ericjwolf thanks
14:25 mwhahaha that's an example of how we're using that info
14:26 ericjwolf one more question
14:27 ericjwolf under the section:  quantum_settings:
14:27 ericjwolf L2:     base_mac: fa:16:3e:00:00:00     phys_nets:       physnet2:         bridge: br-prv         vlan_range: 3000:4000     segmentation_type: vlan
14:27 ericjwolf arg
14:27 ericjwolf http://paste.openstack.org/show/437856/
14:28 ericjwolf I add a new network section here for my bridge.
14:28 ericjwolf in the ml2 file I can add multiple entries for VLAN identities for nonconcurrent ranges.
14:29 ericjwolf In here can I do the same?  Add the same network twice with different ranges:
14:30 ericjwolf http://paste.openstack.org/show/437857/
14:31 mwhahaha i'm not sure on that one
14:31 claflico joined #fuel
14:33 mwhahaha if it was me, i'd just try it :D
14:34 ericjwolf :)  I did and did not work.
14:34 ericjwolf but like i said I could have screwed up when I uploaded it.
14:35 ericjwolf I have a set of files I can push after the deployment but was hoping to get fuel to do this for me.
14:35 ericjwolf I am far from a puppet user.
14:35 ericjwolf :(
14:36 mwhahaha so you can see if it makes it to the end node by checking the deployed node's /etc/astute.yaml
14:36 mwhahaha there's a network_scheme section which should have your changes if it worked
14:38 vsedelnik joined #fuel
14:46 gparaskevas joined #fuel
14:46 ericjwolf that is actually on the node correct?
14:46 mwhahaha yea
14:47 gparaskevas Hello there after rebooting th cluster VIP is not accesible
14:47 gparaskevas Fuel 6.1
14:47 mwhahaha gparaskevas: public vip?
14:47 gparaskevas 3 controllers one compute
14:47 gparaskevas yes
14:47 mwhahaha is the public gateway pingable?
14:47 gparaskevas yes
14:47 gparaskevas this is a private network
14:47 gparaskevas though
14:48 gparaskevas corosync is up on all nodes
14:48 mwhahaha do me a favor and provide a crm status http://paste.openstack.org/
14:48 gparaskevas haproxy is up as well on all node
14:48 gparaskevas crm?
14:48 pbrooko joined #fuel
14:48 mwhahaha run  'crm status' on a controller
14:49 gparaskevas ERROR: status: crm_mon (rc=107): Connection to cluster failed: Transport endpoint is not connected
14:49 gparaskevas it was one line so i pasted here
14:49 mwhahaha that's fine, your corosync cluster isn't up
14:49 mwhahaha which may be your problem
14:49 gparaskevas yes
14:50 mwhahaha did you reboot them all at once?
14:50 gparaskevas yes
14:50 gparaskevas its just a test env on vms
14:50 gparaskevas so i wasnt very polite
14:51 mwhahaha it might take it a while to come up then
14:51 gparaskevas its two hours now
14:51 gparaskevas :P
14:51 mwhahaha then not so much
14:51 gparaskevas karma?
14:51 mwhahaha probably
14:52 dkusidlo joined #fuel
14:52 severion joined #fuel
14:53 mwhahaha you can check the pacemaker logs
14:53 mwhahaha see if there's anything obvious
14:53 blahRus joined #fuel
14:53 gparaskevas ok
14:55 rmoe joined #fuel
14:58 angdraug_ joined #fuel
15:00 gparaskevas http://paste.openstack.org/show/437936/
15:00 gparaskevas mwhahaha: take a look tthis is the same on all controllers
15:01 tzn joined #fuel
15:01 mwhahaha so it might not be running, you could try restarting it
15:02 gparaskevas i am doing a polite restart one by one
15:02 gparaskevas now the status changes
15:03 gparaskevas le me see once more
15:05 gparaskevas so cluster goes online one node at a time
15:07 gparaskevas so cluster s up crm status show all nodes online
15:07 gparaskevas but no horizon on VIP
15:07 mwhahaha give it a bit
15:08 gparaskevas ok
15:09 gparaskevas is mysql status stop/waiting what is should be?
15:10 gparaskevas i see it brings it back on automatically
15:10 mwhahaha it probably has to rebuild the galera cluster
15:10 mwhahaha so it may take a bit
15:10 gparaskevas ok
15:11 pbrooko joined #fuel
15:28 vsedelnik joined #fuel
15:39 neophy joined #fuel
16:00 skylerberg joined #fuel
16:02 tzn joined #fuel
16:20 RageLtMan joined #fuel
16:20 RageLtMan Is there a known issue with running fuel against a KVM node for the controller services? Fuel 6 and fuel 7 both screw up the public IPs such that services dont work - fuel thinks the master is at the first ip in the given range, and it always falls to the 2nd or 3rd.
16:25 mwhahaha i'm not aware of any issue
16:25 mwhahaha that being said, doesn't mean there isnt one
16:27 jaypipes joined #fuel
16:33 tzn joined #fuel
16:47 Akshik joined #fuel
16:48 tzn joined #fuel
16:50 sergmelikyan joined #fuel
16:50 Akshik Are there any hyper-v plug-in available for fuel?
16:50 vsedelnik joined #fuel
17:01 skylerberg joined #fuel
17:13 Akshik joined #fuel
17:14 sergmelikyan joined #fuel
17:17 nihilifer Akshik: currently no, there's no hyper-v plugin
17:30 sergmelikyan joined #fuel
17:31 neophy joined #fuel
17:33 vern quick question: does 7.0 use vxlan by default?
17:39 mwhahaha it will
17:40 thansen joined #fuel
17:40 sergmelikyan joined #fuel
17:43 vern thanks
17:44 mwhahaha well vxlan by default for the tunneled configuration instead of gre
17:48 sergmelikyan joined #fuel
17:59 neophy joined #fuel
17:59 skylerberg left #fuel
18:12 v1k0d3n joined #fuel
18:26 akshik joined #fuel
18:29 e0ne joined #fuel
18:40 tzn joined #fuel
18:42 jobewan joined #fuel
18:47 neophy_ joined #fuel
18:54 sergmelikyan joined #fuel
18:55 e0ne joined #fuel
18:59 HeOS joined #fuel
19:19 mkwiek joined #fuel
19:20 DevStok_ joined #fuel
19:21 DevStok_ hi
19:21 DevStok_ i cant use glance nova etc...
19:22 DevStok_ can I make a snapshot using ceph?
19:22 DevStok_ my vm disk is on compute :(
19:22 DevStok_ or using virsh commands
19:25 pbrooko joined #fuel
19:28 kaliya DevStok, whassup
19:28 julien_ZTE joined #fuel
19:29 kaliya yes you can snapshot with virsh snapshot-create
19:29 xarses joined #fuel
19:32 tzn joined #fuel
19:34 kat_pimenova joined #fuel
19:34 sc68cal joined #fuel
19:34 book` joined #fuel
19:35 HeOS joined #fuel
19:35 neophy_ joined #fuel
19:35 RageLtMan joined #fuel
19:35 hakimo joined #fuel
19:35 eliqiao joined #fuel
19:35 byrdnuge joined #fuel
19:35 Verilium joined #fuel
19:35 zimboboyd joined #fuel
19:42 tzn joined #fuel
19:46 julien_ZTE joined #fuel
19:51 stamak joined #fuel
19:52 HeOS joined #fuel
19:52 neophy_ joined #fuel
19:52 RageLtMan joined #fuel
19:52 hakimo joined #fuel
19:52 eliqiao joined #fuel
19:52 byrdnuge joined #fuel
19:52 Verilium joined #fuel
19:52 zimboboyd joined #fuel
19:55 e0ne joined #fuel
20:03 sergmelikyan joined #fuel
20:11 julien_ZTE joined #fuel
20:19 julien_ZTE joined #fuel
20:20 vsedelnik joined #fuel
20:37 JoeStack left #fuel
20:41 sbfox joined #fuel
20:41 vsedelni_ joined #fuel
20:45 tzn joined #fuel
20:51 vsedelnik joined #fuel
20:55 DevStok_ Hey Fabri
20:59 kaliya hey DevStok
20:59 kaliya hey DevStok_
21:02 DevStok_ what's up?
21:05 julien_ZTE joined #fuel
21:05 DevStok_ I've deployed a single node 5.1 version
21:05 DevStok_ I'm facing this issue
21:05 DevStok_ ERROR neutron.agent.linux.async_process [-] An error occurred while communicating with async process [['ovsdb-client', 'monitor', 'Interface', 'name,ofport', '--format=json']].
21:05 DevStok_ in neutron-plugin-openvswitch-agent.log
21:11 RageLtMan Hi all, are there any known issues with deploying the OS controller role to a kvm node with virtio NICs? After deployment it claims horizon is at the 2nd IP of the allocated pool, but it lands in the 3rd. This is fuel 7. With fuel 6.1, fuel thinks horizon is on the 1st IP of the allocated range, but its still on the 3rd
21:12 RageLtMan if allowed public IPs are 192.168.1.1-192.168.1.50, then the horizon node always ends up at 192.168.1.3 and services are broken since half of them think it is at .1 or .2
21:13 mwhahaha what's your gateway?
21:14 mwhahaha usually the gateway is .1 and you set the public ip range to 2-whatever
21:14 RageLtMan the subnet is /22 so the gateway is 192.168.0.1
21:15 mwhahaha what do you have configured for your public network?
21:15 RageLtMan 192.168.0.0/22 allocated range is 192.168.1.1-50 floating is 192.168.1.51-251
21:15 vsedelnik joined #fuel
21:16 mwhahaha sounds like a bug
21:16 RageLtMan This only started to happen when is tarted to use a KVM domain as the control node
21:16 e0ne joined #fuel
21:16 RageLtMan on both the release fuel and the current 7.0
21:16 RageLtMan or last night's 7 anyway
21:16 mwhahaha https://wiki.openstack.org/wiki/Fuel/How_to_contribute#Test_and_report_bugs
21:16 mwhahaha might be the /22, i'm not sure. best bet is to create a bug
21:17 RageLtMan the /22 worked fine before i switched to the kvm node
21:17 RageLtMan running ZFS underneath the storage, best way i could figure to do it out of the box
21:17 mwhahaha the kvm setting shouldn't have anything to do with the network ip allocation
21:17 RageLtMan i'm in 100% agreement with you there
21:18 mwhahaha best thing to do is create a bug and include a diagnostic snapshot
21:18 RageLtMan when it was a bare metal node, it worked. virtualbox worked too, kinda, but too slow. kvm deploys great, but then this happens
21:41 DevStok_ kaliya : I got the snapshot in status running
21:41 DevStok_ how can i download it?
21:54 kaliya DevStok_, ?
21:59 DevStok_ I wanna move the snapshot to another env
22:00 julien_ZTE joined #fuel
22:00 kaliya DevStok_, live migration (if conditions hold), or get the file to the other env and import it
22:01 DevStok_ using virsh?
22:01 DevStok_ can i get a qcow?
22:01 kaliya Yes and on the other side you can import it as a volume, and nova boot from this volume
22:01 kaliya DevStok_, yes you can't snapshot raw images
22:02 DevStok_ so i can't take that snapshot and then do a glance image-create
22:14 e0ne joined #fuel
22:54 julien_ZTE joined #fuel
22:55 rmoe joined #fuel
23:06 youellet_ joined #fuel
23:12 neophy_ joined #fuel
23:13 glidewell joined #fuel
23:14 glidewell TCP connection to AMQP failed: Could not establish TCP connection to 10.20.0.2:5672. Retry 5 sec later...
23:14 glidewell how can I fix this issue with Fuel ? please anyone
23:14 glidewell can help me out ?
23:15 kaliya glidewell, version of Fuel and oslo ?
23:15 kaliya are services up? usually you can ignore that
23:15 glidewell Fuel 6.1
23:15 kaliya rabbitmqctl cluster_status
23:16 glidewell where should I run that command ?
23:16 glidewell inside the rabbit mq container ?
23:18 glidewell also my nodes are stuck in "pending addition" for a while now
23:20 glidewell [root@fuel /]# rabbitmqctl cluster_status Cluster status of node rabbit@fuel ... [{nodes,[{disc,[rabbit@fuel]}]},  {running_nodes,[rabbit@fuel]},  {cluster_name,<<"rabbit@fuel.default.local">>},  {partitions,[]}] ...done. [root@fuel /]#
23:23 glidewell hi @kaliya, Can you please guide me a bit out of this trouble ?
23:39 glidewell got it working, thanks Kaliya
23:47 julien_ZTE joined #fuel

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary