Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2015-07-22

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

All times shown according to UTC.

Time Nick Message
00:04 sergmelikyan joined #fuel
00:06 osryan_ @mihgen: thanks, i thought thats what it was
00:11 monester_ joined #fuel
00:12 omolchanov1 joined #fuel
00:19 nurla joined #fuel
00:46 sergmelikyan joined #fuel
00:55 sergmelikyan joined #fuel
01:00 serg_melikyan joined #fuel
01:00 user01 joined #fuel
01:18 sergmelikyan joined #fuel
01:25 ashtokolov joined #fuel
01:45 sergmelikyan joined #fuel
01:54 sergmelikyan joined #fuel
01:55 sergmelikyan joined #fuel
02:02 sergmelikyan joined #fuel
02:52 hakimo joined #fuel
04:31 tlbr joined #fuel
04:35 stamak joined #fuel
04:55 vidalinux joined #fuel
05:03 kutija joined #fuel
05:14 user01 joined #fuel
05:14 user01 joined #fuel
05:30 stamak joined #fuel
06:06 Miouge joined #fuel
06:11 monester joined #fuel
06:24 barthalion asilenkov: please join #fuel-infra; we need to know the status of packaging librarian-puppet and honestly I have no idea who's responsible for it
07:06 ub joined #fuel
07:09 sergmelikyan joined #fuel
07:22 e0ne joined #fuel
07:36 plicploc joined #fuel
07:45 mkwiek joined #fuel
07:56 tatyana joined #fuel
08:02 user01 joined #fuel
08:13 stamak joined #fuel
08:14 hyperbaba joined #fuel
08:35 tzn joined #fuel
08:36 tzn joined #fuel
08:49 stamak joined #fuel
08:57 ashtokolov joined #fuel
09:01 subscope joined #fuel
09:07 e0ne joined #fuel
09:18 plicploc Hi, i am trying to use swift as backend for cinder-backup but it is not workling, does someone have advices about that issue ?
09:19 plicploc I am working on a HA multimode environment in ubuntu 12.04 , deployed with fuel 6.0
09:19 HeOS joined #fuel
09:20 plicploc I installed cinder-backup package from the fuel master node and it seems correctly configured
09:26 Miouge If I enable SSL according to the documentation (https://docs.mirantis.com/openstack/fuel/fuel-6.1/operations.html?highlight=ssl#howto-configure-horizon-on-https) do I need to do something to also enable SSL on the WebSocks for noVNC?
09:28 plicploc hmm AFAIK you only need to configure ssl option in nova.conf for novnc
09:35 tatyana joined #fuel
09:36 mattymo_ joined #fuel
09:36 NERvOus joined #fuel
09:36 Miouge Ok, but shouldn’t it also be config in horizon to say “hey use secure websockets” ?
09:37 ashtokolov joined #fuel
09:40 plicploc idk, i don't use horizon
09:40 plicploc :/
09:41 tatyana joined #fuel
09:46 gongysh joined #fuel
10:17 e0ne joined #fuel
10:30 NERvOus joined #fuel
10:32 ashtokolov joined #fuel
10:58 ashtokol_ joined #fuel
11:01 ashtokolov joined #fuel
11:15 ashtokolov joined #fuel
11:38 dklepikov joined #fuel
11:55 jaypipes joined #fuel
12:17 jdandrea joined #fuel
12:24 NERvOus joined #fuel
12:41 ashtokolov joined #fuel
13:00 stamak joined #fuel
13:03 ClaudeD joined #fuel
13:06 ClaudeD Hi. I'm testing Fuel 6.1 on Ubuntu with Neutron (vlans). I have 2 tenants. VMs of tenant #1 can't ping floating addresses of tenant #2's VM. Any clues on how to solve this?
13:06 ClaudeD Of course ping is allowed in the security rules :)
13:07 ClaudeD And ping works fine when issued from a network outside of OpenStack...
13:25 ashtokolov joined #fuel
13:43 subscope joined #fuel
13:50 claflico joined #fuel
14:07 jaypipes joined #fuel
14:08 devStok joined #fuel
14:11 devStok can I delete the guestfs files in /var/tmp on the compute?
14:18 neophy joined #fuel
14:36 tatyana joined #fuel
14:36 ashtokolov joined #fuel
14:41 blahRus joined #fuel
14:44 ashtokolov joined #fuel
14:58 rmoe joined #fuel
15:00 devStok can I delete the guestfs files in /var/tmp on the compute?
15:02 mwhahaha https://bugs.launchpad.net/mos/+bug/1387429
15:02 mwhahaha looks like yes
15:03 mwhahaha what version are you on?
15:03 devStok icehouse
15:03 devStok in the last months  i found comptes full of this files
15:03 devStok 90% of disk
15:05 devStok No downloadable files for bash version 4.3-ubuntu1; perhaps it is a local or obsolete package?
15:06 devStok when i try to update
15:06 devStok update-guestfs-appliance
15:08 ashtokolov joined #fuel
15:11 ashtokolov joined #fuel
15:13 ashtokolov joined #fuel
15:15 ashtokolov joined #fuel
15:18 xarses_ joined #fuel
15:19 xarses_ joined #fuel
15:44 dmitryme joined #fuel
15:50 aignatov joined #fuel
16:03 SergK joined #fuel
16:08 ashtokolov joined #fuel
16:20 warpc__ joined #fuel
16:20 xarses joined #fuel
16:24 rawat joined #fuel
16:35 rawat hi all
16:35 ggolin halleluja, i got an env to deploy correctly
16:36 rawat i am trying to deploy openstack using fuel but after deployment openstack is not working...... even befor installing every slave node had internet but now there is no internet...
16:37 ggolin rawat: does network verification pass?
16:37 rawat ggolin : ye
16:38 ggolin also, define "not working"
16:38 rawat ggolin : yes and deployment also done..
16:38 sergmelikyan joined #fuel
16:39 rawat ggolin : actualy first i got error for ntp , then i had changed ntp server to local but after deployment i am unable to get acces ...
16:40 rawat ggolin : i had verified with controller node there is no nova compute showing....
16:41 ggolin get access where?
16:42 rawat ggolin : from horizon
16:42 ggolin you cant get TO horizon?
16:43 rawat ggolin : why   ?
16:44 rawat ggolin : i have one more setup of fuel with vertualbox and that is working fine...
16:45 ggolin you're not explaining the problem very well, im afraid
16:46 mwhahaha he's using virtual box and the deployed nodes don't have internet
16:46 mwhahaha you may need to add iptables/pf rules to provide network access to those networks for the deployed nodes
16:46 rawat ggolin : in this setup i am using vsphare..
16:48 osryan_ Hmm...I've scoured our network and can verify all the VLANs and subnets are functioning properly but I still can't get any network but the admin(pxe) network to pass traffic. It seems like an issue within openvswitch
16:48 osryan_ so far ive only tried deploying a controller
16:49 ggolin network seems to be the hardest part to get right, it seems
16:49 ggolin at least from my perspective
16:50 ggolin took me a couple of weeks to figure out the required ucs configuration
16:50 osryan_ it certainly can be. i suspect its a config issue on my part and not actually openvswitch, as everything else has been pretty smooth
16:50 ggolin being new to ucs didnt help :)
16:51 osryan_ :)
16:51 ggolin osryan_: everything deploys correctly?
16:51 osryan_ the fuel node and bootstrap os install works
16:51 osryan_ but deploying controller does not
16:51 ggolin network verification passes?
16:51 osryan_ only admin network functions
16:52 rawat yes network verification passes
16:53 ggolin did you change pxe iface?
16:54 rawat pxe is in seprate network
16:55 rawat thet network don't have internet ...
16:55 ggolin i was askin osryan_
16:55 rawat ok
16:55 ggolin sorry
16:57 angdraug joined #fuel
16:57 Rodrigo_BR joined #fuel
16:58 Rodrigo_BR Can I remove the object storage option in horizon dashboard ?
16:59 neophy joined #fuel
17:04 xarses Rodrigo_BR: yes, but not from fuel settings. I cant remember if it automatically picks up the object store listed in keystone endpoints, or if its directly in the /etc/openstack-dashboard config
17:05 xarses osryan_: have you used network checker in the networks settings page?
17:11 osryan_ rawat: yes i changed it
17:12 osryan_ err, i didnt change it. i verified it works
17:24 ggolin so about a couple of hours is an okay timeframe to deploy and env? two controllers, three nodes?
17:26 mwhahaha 3 controllers is recommended for HA
17:26 ggolin yesyes
17:26 mwhahaha but yea it may take a few hours
17:27 mwhahaha depends on speed of hardware/network
17:27 ggolin is it the puppet stuff that runs the longest then?
17:27 mwhahaha yea
17:27 * ggolin taps fingers on desk
17:28 ggolin i guess ill hack some hubot scripts meanwhile
17:37 neophy joined #fuel
17:44 ggolin ninenty minutes using external repos - i guess thats cool
17:48 sergmelikyan joined #fuel
17:55 ggolin gah! i cant add images >.<
17:55 ggolin queued status
18:01 sergmelikyan joined #fuel
18:04 ashtokolov joined #fuel
18:10 ub joined #fuel
18:15 ashtokolov joined #fuel
18:16 ub joined #fuel
18:17 ashtokolov joined #fuel
18:22 osryan_ @xarses: sorry, i missed your response. i tried the checker but it can only check if there are two nodes deployed, which there isnt. i am failing on my first controller node
18:22 xarses osryan_: it will run if all of the nodes are in discovery
18:22 xarses with an error to repos access?
18:27 aglarendil joined #fuel
18:28 ashtokolov joined #fuel
18:28 sergmelikyan joined #fuel
18:30 xarses when controller runs it will switch from fw-admin (admin/pxe) network as default route to public, when that occurs it will start fetching packages from default route (public) if it cant reach the repos it will fail
18:33 ashtokol_ joined #fuel
18:34 ub joined #fuel
18:42 sergmelikyan joined #fuel
18:43 osryan_ @xarses: yep, the repo access is where it is failing on me. so for network test, i can add the nodes and configure interfaces and run the network test prior to deploy?
18:44 ashtokolov joined #fuel
18:44 ClaudeD left #fuel
18:45 e0ne joined #fuel
18:47 osryan_ @xarses: ok, i ran that and everything succeeded. i didnt realize i had that option. i will try and deploy again
18:52 ub joined #fuel
18:58 aglarendil joined #fuel
18:59 nurla joined #fuel
19:01 tzn joined #fuel
19:02 xarses osryan_: ya, as you likely just discovered, you can check them prior to deployment (once the nodes are added to the env) It also allows you to verify the interface layout priory to deployment which i guess was likely your problem
19:05 osryan_ so during that network check, does it also verify that the public network can access the repo sites on the internet beyond the gateway?
19:18 jdandrea left #fuel
19:31 ashtokolov joined #fuel
19:32 ashtokolov joined #fuel
19:36 stamak joined #fuel
19:37 sergmelikyan joined #fuel
19:41 sergmelikyan joined #fuel
19:47 osryan_ it worked! I can't believe that's all it took
19:48 osryan_ does that mean the network test is a requirement of the setup? i didnt actually change anything network-wise, just added in the test
19:50 Verilium Is there a way to login by SSH to the hosts that have been bootstrapped by fuel?  I'd be looking for the location of the private key.
19:50 HeOS joined #fuel
19:55 Verilium Ah, nevermind.  Found it, nailgun container in bootstrap.rsa.
19:57 mwhahaha that should also be in the root user's keys on the fuelmaster i think
19:58 Verilium ...and the only reason I'm actually looking for this is that 2 hosts got detected that I wasn't aware of.  I guess they've been trying to boot by PXE on this network continuously for who knows how long. :P
20:01 mwhahaha someone's laptop ;)
20:01 osryan_ @xarses: thank you for the help, looks like im up and running
20:02 xarses osryan_: woot!
20:02 osryan_ mwhahaha: thanks for getting me past the fuel install a few days ago too. huge help
20:02 xarses osryan_: yes, it checks the public requirement
20:03 xarses Verilium: you can ssh to the discovered nodes from root@fuel-masternode
20:03 xarses the sshkeys are in the .ssh/config fore you, and in the /root/.ssh
20:12 ub joined #fuel
20:12 aglarendil joined #fuel
20:47 gongysh joined #fuel
20:56 sergmelikyan joined #fuel
21:06 skylerberg joined #fuel
21:06 sergmelikyan joined #fuel
21:06 skylerberg I am having trouble setting up Fuel with Virtual Box on Mac. I have NAT port forwarding set up to forward TCP 8000. This doesn't work unless I run dhclient eth0. The guest does not connect to the internet. That is what is currently blocking me.
21:06 skylerberg I think this has something to do with the way the fuel iso sets up the network.
21:06 skylerberg Has anyone got this working or have any ideas about this problem?
21:09 mwhahaha you could try an older version of the scripts that used pfctl to allow for internet
21:09 mwhahaha https://github.com/stackforge/fuel-main/commit/8a1433341fd1f36b949c226d29d5656248f18928
21:13 tzn joined #fuel
21:30 e0ne joined #fuel
21:35 kaliya joined #fuel
21:39 skylerberg mwhahaha: I tried the script, but I still don't have any internet connection.
21:39 mwhahaha i know i'm using a version of that
21:41 skylerberg So did you have the same problem where the internet was not working out of the box when you used virtual box?
21:41 skylerberg And did NAT work out of the box for you, or did you have to run dhclient (or did you not use NAT)?
21:49 skylerberg Okay, I think I have a workaround. I just added another NAT adapter that won't going to be used for anything except internet connectivity.
21:50 skylerberg Nvm, now I am getting a bad gateway error whenever I try to do anything with fuel.
21:53 mwhahaha check the docker containers
21:53 mwhahaha dockerctl check
21:53 sergmelikyan joined #fuel
21:59 ashtokolov joined #fuel
22:10 sergmelikyan joined #fuel
22:12 skylerberg mwhahaha: I am getting ProtocolError, 404 Not found when I check docker.
22:16 skylerberg Oh, I should mention that I killed httpd a couple of minutes ago and restarted nginx. This got most of it working. I would not be surprised if there are some problems because of that.
22:17 skylerberg It got things working enough that I can debug how my plugin shows up on the interface.
22:24 ashtokolov joined #fuel
22:27 ashtokolov joined #fuel
22:30 skylerberg joined #fuel
22:35 ub joined #fuel
22:51 sergmelikyan joined #fuel
23:23 skylerberg Is there a way to dynamically add UI elements in fuel? For example, if I specify that I have three backends for cinder, can I have a configuration section show up for each backend?
23:25 aglarendil joined #fuel
23:28 sergmelikyan joined #fuel
23:40 rmoe joined #fuel
23:41 tzn joined #fuel
23:45 xarses joined #fuel
23:51 skylerberg In the network menu, there is the IP Range section that has a button to add more sections. I can't figure out how to put that in my environment_config.yaml for my plugin though.
23:58 sergmelikyan joined #fuel

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