Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2013-11-21

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

All times shown according to UTC.

Time Nick Message
00:37 teran joined #fuel
01:27 anoopl joined #fuel
01:47 fandikurnia01 joined #fuel
02:19 vkozhukalov joined #fuel
02:29 teran joined #fuel
02:54 sanek joined #fuel
02:54 anoopl joined #fuel
03:05 xarses joined #fuel
03:18 fandikurnia01 joined #fuel
03:28 ArminderS joined #fuel
03:59 teran joined #fuel
04:33 sanek joined #fuel
05:20 IlyaE joined #fuel
05:29 teran joined #fuel
05:35 mihgen joined #fuel
06:00 Arminder- joined #fuel
06:11 IlyaE joined #fuel
06:21 ToTsiroll joined #fuel
06:38 teran joined #fuel
06:47 IlyaE joined #fuel
06:56 r0mikiam joined #fuel
07:46 ToTsiroll how to fix the dashboard redirection problem in httpd logs?
07:47 ToTsiroll "[warn] [client 192.168.151.11] incomplete redirection target of '/dashboard/' for URI '/' modified to 'http://node-26.testbed.cloudtop.ph/dashboard/'"
07:50 xdeller joined #fuel
07:51 tatyana joined #fuel
07:54 vkozhukalov joined #fuel
07:55 e0ne joined #fuel
07:58 ToTsirolls joined #fuel
08:01 Valyavskiy joined #fuel
08:07 vvalyavskiy joined #fuel
08:09 vvalyavskiy joined #fuel
08:18 Valyavskiy joined #fuel
08:19 Valyavskiy is it possible create partitions with fuel node --node-id 3 --disk -d / -u then combine this partitions in raids, and then feed raid to lvm? if no, can i choose specific anaconda profile for specific role(controller,compute)?
08:22 b-zone joined #fuel
08:29 teran joined #fuel
08:30 mihgen joined #fuel
08:34 xdeller joined #fuel
08:37 jkirnosova joined #fuel
08:47 ruhe joined #fuel
08:59 vk_ joined #fuel
09:04 xdeller left #fuel
09:13 dnikishov joined #fuel
09:20 anotchenko joined #fuel
09:21 IlyaE joined #fuel
09:37 teran joined #fuel
09:43 vk_ joined #fuel
09:58 ruhe joined #fuel
10:01 e0ne joined #fuel
10:06 xdeller joined #fuel
10:07 ToTsirolls anyone knows how to remove the warning in httpd logs about "Incomplete redirection target of '/dashboard/' for URI '/' modified to 'http://node-01.testbed.website.com/dashboard/' ??
10:10 xdeller that`s haproxy pinger, you may raise default debug level in apache if you want to
10:29 anotchenko joined #fuel
10:29 r0mikiam joined #fuel
10:31 tatyana joined #fuel
10:31 Fecn joined #fuel
10:34 Fecn Hi Guys - I seem to have an orphaned node which is not part of any deployment, but still thinks that it is 'Pending Deletion'. (In reality it has been returned back to the bootstrap image).
10:34 Fecn How can I tell fuel that it's back in the bootstrap state?
10:37 b-zone joined #fuel
10:41 ruhe joined #fuel
10:43 AndreyDanin Fecn, please provide an output of this URL http://master_node_ip:8000/api/nodes/
10:44 IlyaE joined #fuel
10:46 e0ne_ joined #fuel
10:46 Fecn AndreyDanin: That is so much easier said than done... (mainly because my silly employer blocks IRC access, so I have to use an isolated system for IRC)... can't cut/paste from one to the other
10:47 IlyaE joined #fuel
10:47 Fecn Try http://paste.openstack.org/show/53737
10:48 Fecn It should show that the node ending in 8E:A0 says pending_delete : true
10:50 AndreyDanin Fecn, just put it to pastebin.org
10:50 AndreyDanin oops sorry
10:50 Fecn I used the paste from the channel topic
10:56 mihgen joined #fuel
10:59 anotchenko joined #fuel
11:00 AndreyDanin Fecn, how did you do that? Did you delete cluster and node got stuck in this state? Do you use Fuel 3.2?
11:01 Fecn Nope - we found out that the little javascript button for 'show logs' doesn't always update... sometimes it's still on 'delete additions' whilst the deployment is ongoign
11:01 Fecn if you click it at that stage, the node is removed from the cluster and orphaned
11:01 Fecn it still thought it was installing centos... but I told cobbler that it should be back at bootstrap in the hope that Fuel would recognise it as available again
11:03 Fecn The issue was that there were two of us viewing the Fuel UI from two different browser... My mate clicked on 'Deploy'... and after a few mins, I clicked on the 'show logs' link (or what I'd thought was show logs) but it turned out to still be 'delete additions'
11:06 Fecn ... and yes.... Fuel 3.2
11:08 AndreyDanin Did you create a bug about it?
11:09 AndreyDanin Actually we don't support for simultaneous work with one cluster, I think
11:12 AndreyDanin try this command: curl -X DELETE http://fuel_master_ip:8000/api/nodes/106
11:17 Fecn This is one issue where we haven't yet raised a bug... (we've been doing about 4 tickets a day with mirantis this past week)
11:17 teran_ joined #fuel
11:19 Fecn It tells me 'The requested method DELETE is not allowed for the url /api/nodes/106"
11:21 Fecn bah.. my bad.. forgot hte :8000
11:21 Fecn That's better
11:21 Fecn Thanks AndreyDanin
11:21 Fecn problem solved
11:22 AndreyDanin good
11:25 AndreyDanin When you clicked to the 'show logs' icon it looked like a list of paper or like a circle with arrow?
11:25 Fecn circle with arrow... but I didn't notice that until just after I'd clicked
11:25 Fecn the progress bar was 50% green at the time
11:25 Fecn but the icon hadn't changed to the paper
11:26 AndreyDanin Was the progress bar on the node presented?
11:27 AndreyDanin Or you talk about the global progress bar
11:27 Fecn It was on that specific node
11:27 Fecn It was actually the case that 90% of the nodes were up to 100% complete... and the one I clicked on only looked about 50% done... hence the reason I wanted to look at the logs for it
11:28 Fecn the progress bar for the node WAS presented
11:29 AndreyDanin Please file the bug. We will fix it. Thank you.
11:29 Fecn Thanks - I'll get my mate to file it when he arrives at work :)
11:34 AndreyDanin Fecn, Did it occur for this node only? Or other nodes had incorrect icons too?
11:34 Fecn All nodes had the incorrect Icon (I think) but I only clicked on one of them
11:41 anotchenko joined #fuel
11:42 mihgen joined #fuel
11:55 Fecn Can anyone tell me what '(/Stage[main]/Osnailyfacter::Cluster_ha/Nova_floating_range[10.96.192.60-10.96.192.124]) Could not evaluate: Unable to reconnect to 10.96.192.2 after 5 attempts' is all about... and whether it's fatal or not... (seemed to be non-fatal on one deployment we did, but fatal in a different rack of nodes)
11:56 Fecn On this current failed deploy it is followed up by 'Could not access hsadow instance 'vip__management_old' - no such file or directory
12:04 teran joined #fuel
12:07 dburmistrov joined #fuel
12:10 mattymo Fecn, floating range fails because openstack-nova-api isn't started
12:10 mattymo meaning there was a problem earlier
12:10 Fecn It was all green lines in the log until that one about not being able to evaluate the range
12:13 Fecn Logs show that it succeeded in starting nova-api 2 seconds before that message about floating ip range was logged
12:22 isviridov joined #fuel
12:26 ruhe joined #fuel
13:10 teran_ joined #fuel
13:11 teran__ joined #fuel
13:17 ruhe joined #fuel
13:21 xdeller joined #fuel
13:29 e0ne joined #fuel
13:31 e0ne_ joined #fuel
13:33 e0ne joined #fuel
13:34 mihgen joined #fuel
13:48 e0ne joined #fuel
13:57 rmoe joined #fuel
14:00 e0ne_ joined #fuel
14:23 anotchenko joined #fuel
14:25 rmoe joined #fuel
14:34 aglarendil_ joined #fuel
14:42 AndreyDanin Fecn, try to filter the logs by ERROR status
14:50 xdeller joined #fuel
15:01 ArminderS joined #fuel
15:16 ruhe joined #fuel
15:23 teran joined #fuel
15:47 anotchenko joined #fuel
16:31 ruhe joined #fuel
16:59 TSCHAKMac joined #fuel
17:00 TSCHAKMac in the Installation Pre-req's guide, I see for the Fuel Master Node that 1 NIC is needed (obviously for admin PXE), but for Red Hat OpenStack, two are needed, what is the second NIC interface for?
17:11 xarses joined #fuel
17:13 rmoe joined #fuel
17:14 ArminderS joined #fuel
17:17 vkozhukalov joined #fuel
17:24 angdraug joined #fuel
17:31 teran joined #fuel
17:32 ArminderS joined #fuel
17:36 teran_ joined #fuel
17:37 ArminderS joined #fuel
17:39 ruhe joined #fuel
17:41 aleksandr_null joined #fuel
17:42 ArminderS joined #fuel
17:42 mihgen joined #fuel
19:41 dhblaz joined #fuel
19:42 xarses joined #fuel
19:54 mihgen joined #fuel
20:00 ArminderS joined #fuel
20:27 mihgen__ joined #fuel
20:48 xarses joined #fuel
22:14 TSCHAKMac joined #fuel
22:27 Fecn joined #fuel
22:28 Fecn Hi Folks - I've done a seemingly successful deployment of a Nova-Network/Grizzly/Centos (all green lines, Success message etc)... but I seem to have an issue with the Swift(Ceph) containers... The dashboard says "Error: Unable to retrieve container list." - Is this a known issue?
22:29 Fecn Could it be because port 6780 is not lised in the haproxy config file?
22:42 tatyana joined #fuel
22:53 xarses I'm guessing this is a HA deployment?
22:53 xarses Fecn: ^
22:55 xarses Also is this neutron, or nova networks?
22:57 Fecn Hi xarses
22:57 Fecn Sorry...
22:57 Fecn yes HA and nova
22:57 xarses ya, you noted, its nova networks
22:58 Fecn and it seems that the problem was down to having the wrong auth key in the keyring.radosgw.gateway file
22:58 xarses you need to apply https://github.com/xarses/fuel/compare/radosgw-fixes
22:58 Fecn Nice - thanks
22:58 Fecn we'll do that now
22:58 xarses which should fix ha proxy, and the keys
22:59 Fecn Perfect - Thank you :-D
22:59 xarses you can patch -p2 them over /etc/puppet/manifests
23:00 xarses if you use neutron ha there is a separate bug which prevents all mons from starting
23:00 xarses Also, 3.2.1 should be out early next week
23:01 xarses which has these all in it
23:01 Fecn Even better :)
23:32 dhblaz joined #fuel
23:37 anoopl joined #fuel

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