Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2016-03-30

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

All times shown according to UTC.

Time Nick Message
00:12 BobBall_ joined #fuel
00:17 BobBall joined #fuel
00:38 dilyin joined #fuel
01:04 fatdragon joined #fuel
01:29 criss2016 joined #fuel
01:48 ilbot3 joined #fuel
01:48 Topic for #fuel is now Fuel 8.0 (Liberty) https://www.fuel-infra.org/ | Paste here http://paste.openstack.org/ | IRC logs http://irclog.perlgeek.de/fuel/
02:18 l4ng1t joined #fuel
02:39 krobzaur_ joined #fuel
03:42 askb joined #fuel
04:00 javeriak joined #fuel
04:01 l4ng1t joined #fuel
04:35 vsedelnik joined #fuel
04:47 l4ng1t1 joined #fuel
05:06 javeriak joined #fuel
05:26 vsedelnik joined #fuel
05:36 fatdragon joined #fuel
05:52 fredeerock I'm getting a fuel network verification error complaining about finding a DHCP server on my public network NIC: Node discovered DHCP server via with following parameters (...) This server will conflict with the installation. Should I just ignore this error?
05:59 magicboiz joined #fuel
06:04 javeriak joined #fuel
06:11 javeriak joined #fuel
06:17 akaszuba joined #fuel
06:25 javeriak joined #fuel
07:01 neilus joined #fuel
07:07 javeriak joined #fuel
07:22 cartik joined #fuel
07:23 magicboiz joined #fuel
07:31 permalac joined #fuel
07:38 elo joined #fuel
07:53 alex_didenko joined #fuel
07:58 vsedelnik joined #fuel
08:00 vsedelni_ joined #fuel
08:07 fatdragon joined #fuel
08:28 krogon joined #fuel
08:48 fatdragon joined #fuel
08:48 e0ne joined #fuel
08:50 zinovik joined #fuel
08:57 bazilza joined #fuel
09:23 mescanef joined #fuel
09:43 javeriak joined #fuel
09:51 ddmitriev joined #fuel
10:02 fatdragon joined #fuel
10:23 krypto joined #fuel
10:23 krypto in fuel 9 is there a plan to remove containers
10:24 javeriak joined #fuel
10:37 kozhukalov fredeerock: on the master node we usually need at least two network interfaces: one for admin network and another for public network. Fuel assumes slave nodes must be boot via PXE from the master node (slave nodes will get their IP and so called 'next server' parameter needed for PXE boot from DHCP server). That requires slave nodes to be at the same
10:37 kozhukalov untagged vlan with admin network interface of the master node. If there is another DHCP server available, that could break PXE flow. Here is the list of pre-installation requirements https://docs.mirantis.com/openstack/fuel/fuel-8.0/fuel-install-guide.html#pre-installation-checklists
10:38 kozhukalov fredeerock: and here https://docs.mirantis.com/openstack/fuel/fuel-8.0/fuel-install-guide.html#system-requirements see Network requirements section
10:39 kozhukalov krypto: yes, in 9.0 there won't be docker containers on the master node
10:44 hkominos joined #fuel
10:47 vsedelnik joined #fuel
11:12 hkominos Hi all. I have A connectivity problem with my fuel deployment. Can anyone help ?
11:14 yassine_ joined #fuel
11:18 vsedelnik joined #fuel
11:22 LanceHaig joined #fuel
11:25 LanceHaig joined #fuel
11:32 kozhukalov hkominos: what kind of problem you are facing?
11:49 hkominos joined #fuel
11:52 krypto kozhukalov any idea why Container support is dropped is it mentioned somewhere just curious
11:52 hkominos My Vms dont get a DHCP messages (or anything) I am trying to debug the problem and I am mirroring the traffic from the OVS switches to see what is going on. I can see the DHCP request message in br-int but not in Br_prv
11:53 hkominos (Virtualbox Deployment Fuel 7,0 1 Controller 1 Compute Ubuntu 14,04)
11:55 hkominos to be more specific,in the compute node I mirror the traffic from int-br-prv (and see the dhcp request). When i mirror from phy-br-prv i see nothing
12:05 kutija joined #fuel
12:12 javeriak joined #fuel
12:13 smachara joined #fuel
12:14 smachara Hi, I'm new with openstack and fuel an I have several question.
12:16 smachara Is someone here?
12:20 smachara How to add new nodes to the pool of nodes in openstack
12:21 kozhukalov holser_: xenolog sbog aglarendil could you please give some advice how to deal with DHCP problem that hkominos is facing? ^^
12:22 smachara I remove a node, now I want it back again?
12:23 kozhukalov smachara: what do you mean when say 'pool of nodes'? fuel is a deployment tool. before deployment you assign roles on nodes and deploy. once openstack cluster is deployed, you can add computes to the cluster and then again click "deploy changes" button
12:24 smachara the nodes that are  discovered by fuel
12:26 smachara the nodes that will be added to the enviromets.  total nodes is shown in the top right of the page in FUEL UI
12:29 kozhukalov smachara: if you remove a node from a cluster you can then add it back, but be aware it will be reprovisioned and redeployed from scratch
12:29 kozhukalov OS will be re-installed
12:29 kozhukalov and it works for computes and for controllers
12:30 smachara I haven't deploy openstack yet, I get the following error 'Deployment has failed. Timeout of deployment is exceeded.'
12:30 smachara What are the steps to add it back
12:31 kozhukalov smachara: you could try to figure out what went wrong taking a look at /var/log/docker-logs/astute/astute.log
12:31 kozhukalov anyway, you can just click "deploy changes " button to try again
12:32 kozhukalov smachara: just add it to cluster (assign a role to it) and click "deploy changes"
12:34 smachara In Fuel UI, I dont see how to re discover new nodes, for example I add a new VM to the cluster with new MAC and is not reconized automaticaly
12:38 smachara In the pool I have 2/2, 2 unalocated nodes / 2 total
12:40 smachara Maybe, I don't know how to add nodes to the cluster. I clone a vm that was reconized by FUEL, then I change the MAC and I reboot the machine. It is OK?
12:41 kozhukalov yottatsa: could you please help with DHCP related issue that hkominos has encountered? ^^
12:42 hkominos @smachara this may sound strange but just to clarify. Are you booting from the PXE ?
12:42 smachara Yes
12:43 kozhukalov smachara: when you remove a node you need to click "deploy changes" to perform real removal. fuel will remove this node from cobbler and then you can reboot this node. it will then be re-discovered
12:44 smachara I will try.
12:44 kozhukalov smachara: yes, mac change should be enough
12:50 javeriak joined #fuel
12:51 smachara To avoi my problem of timeout,  I'm going to deplo one node at the time, first I will deploy the controller (CONTROLLER-MONGO-BASE-OS)
12:53 kozhukalov smachara: FYI, HA cluster could not be deployed one by one, you have to have at least 3 nodes
12:53 kozhukalov smachara: you could try to understand what is the reason of timing out
12:54 kozhukalov I mean if it is slow internet connection, then it is better to create local mirror
12:57 smachara For the moment I want to install the enviroment to learn Openstack, I don't need a HA architecture.
12:58 kozhukalov smachara: then it should be ok to deploy nodes one by one
13:00 smachara However yesterday I tried to install only the controller node and I got a time out error. so, work with local mirror is not bad idea.
13:07 criss2016 Hi all... Anyone succeded to configure in Fuel 7.0 mutiple external networks using  the same l3-agent ?
13:08 smachara kozhukalov: Sorry, How can I set a local mirror? I imagine that I can in Settings/Repositories FUEL UI
13:08 smachara kozhukalov: Sorry, How can I set a local mirror? I imagine that I can do it in Settings/Repositories FUEL UI
13:10 hkominos You can from the FUel CLI
13:12 hkominos btw i had a similar issue with timeout and i could not create a mirror.(same timeout) better to find where it does timeout and why first.
13:16 smachara Ok, but how? the time out is at 99% I just find this warning in Astute.log '2016-03-30 11:56:31WARNING[582] Trying to reconnect to message broker. Retry 5 sec later...'
13:17 mescanef_ joined #fuel
13:18 [HeOS] joined #fuel
13:18 smachara but the warnings are after the time out error '2016-03-30 08:18:08ERR[640] Timeout of deployment is exceeded.'
13:21 jmhale joined #fuel
13:25 javeriak joined #fuel
13:28 hkominos Can anybody help with my DHCP issue ?
13:30 javeriak_ joined #fuel
13:31 kozhukalov smachara: if you have ready to use local mirror, then yes settings/repositories is the place where you configure it https://docs.mirantis.com/openstack/fuel/fuel-master/operations.html#configuring-repositories. But fuel also provides a tool to create local mirrors. it should be installed on the master node. fuel-mirror --help
13:32 javeriak joined #fuel
13:43 kozhukalov smachara: 99% could easily be not a real number. When progress bar is waiting for a log pattern (that means real progress) it still moves on slowly (that is how smooth progress bars work).
13:43 asaprykin joined #fuel
13:44 kozhukalov Trying to reconnect to message broker. - means something went wrong while deploying rabbitmq
13:44 kozhukalov probably, rabbitmq is up and running but listens to another port (or smth like this)
13:46 yottatsa joined #fuel
13:48 smachara I run it again to see what happen.
13:48 yottatsa kozhukalov said someone has problems with dhcp
13:50 javeriak joined #fuel
13:50 e0ne joined #fuel
13:53 yottatsa hkominos can you check that openvswitch-agent is running?
13:54 hkominos hi.All agents were running. However i have decided to redeploy now just to test another theory. I would appreciate however if you could just answer this one question. Is br-prv build on top of ETH1 (public traffic)
13:55 gongysh joined #fuel
13:55 hkominos because i found that is listed as down in ip a and i assumed that it should be up.
13:55 hkominos and i searched dsmesg|grep eth1 and it wrote eth1 not ready
13:56 hkominos (eth0 and eth2 were up properly)
13:56 yottatsa which interface do you choose as Private? eth2?
13:56 hkominos 2
13:56 srwilkers_ joined #fuel
13:57 hkominos eth0 was pxe, eth1 was public and eth2 everything else
13:58 yottatsa br-prv is connected to Private interface, in your case it should be eth2. eth1 will be used for br-floating (on controller)
13:58 hkominos so it should be down on the Compute ?
14:16 e0ne joined #fuel
14:21 jaypipes joined #fuel
14:25 fredeerock kozhukalov: Thanks for your help! I do have separate interfaces for admin and public on the master. Funny thing is network verification works with just 1 controller but gets the "Node discovered DHCP server" error when there's 2 controllers.
14:29 javeriak joined #fuel
14:58 fedexo joined #fuel
15:11 akaszuba joined #fuel
15:11 vsedelnik joined #fuel
15:15 fatdragon joined #fuel
15:19 dslevin left #fuel
15:19 blahRus joined #fuel
15:23 javeriak joined #fuel
15:25 askb_ joined #fuel
15:26 javeriak_ joined #fuel
15:27 askb_ joined #fuel
15:29 askb_ joined #fuel
15:30 criss2016 joined #fuel
15:31 askb_ joined #fuel
15:33 askb_ joined #fuel
15:34 Captain_Murdoch joined #fuel
15:34 javeriak joined #fuel
15:35 askb_ joined #fuel
15:37 askb_ joined #fuel
15:39 askb_ joined #fuel
15:39 venkat_ joined #fuel
15:44 permalac joined #fuel
15:56 l4ng1t joined #fuel
16:02 ericjwolf joined #fuel
16:22 notmyname left #fuel
16:25 vsedelnik joined #fuel
16:25 charz joined #fuel
16:28 vsedelnik joined #fuel
17:05 mescanef_ joined #fuel
17:34 mescanef joined #fuel
17:48 srwilkers joined #fuel
17:56 kaliya joined #fuel
18:00 e0ne joined #fuel
18:02 aimeeU joined #fuel
18:13 aimeeU left #fuel
18:14 ericjwolf Good afternoon
18:15 ericjwolf I am seeing a odd behavior on a new server I am trying to add to a cluster.  I am getting an rsync failure.
18:15 ericjwolf Deployment has failed. Method granular_deploy. Failed to execute hook 'sync' Failed to perform sync from rsync://10.20.0.2:/puppet/2014.2.2-6.1/ to /etc/puppet
18:15 ericjwolf When I go to the node and try to rsync I get connection refused.
18:16 ericjwolf root@node-57:~# rsync -v rsync://10.20.0.2:/puppet/2014.2.2-6.1/ /etc/puppet rsync: failed to connect to 10.20.0.2 (10.20.0.2): Connection refused (111) rsync error: error in socket IO (code 10) at clientserver.c(128) [Receiver=3.1.0]
18:16 ericjwolf I am not getting anyting in the logs that I can understand to determine the issue.
18:16 ericjwolf My assumption is there may be an issue with the docker-rsync
18:17 ericjwolf Should I restart the rsync docker?
18:23 mwhahaha yes
18:23 mwhahaha you should try restarting the rsync docker container
18:24 e0ne joined #fuel
18:25 cshahani joined #fuel
18:31 ericjwolf Thanks, I got past that by restarting the docker.  Was there any known issues or things to look out for in rsync?  Wondering why it would all of a sudden stop working.
18:31 e0ne joined #fuel
18:33 e0ne joined #fuel
18:35 mwhahaha I'm not aware of anything
18:35 mwhahaha would have to poke at the logs to see if there was anything that sticks out
18:35 e0ne joined #fuel
18:37 e0ne joined #fuel
19:00 xarses joined #fuel
19:06 vsedelnik joined #fuel
19:11 e0ne joined #fuel
19:31 jt708u joined #fuel
19:37 claflico joined #fuel
19:55 vsedelnik joined #fuel
19:55 angdraug joined #fuel
20:13 angdraug joined #fuel
20:32 angdraug joined #fuel
20:37 vsedelnik joined #fuel
20:45 cshahani joined #fuel
20:48 xarses joined #fuel
21:41 vsedelnik joined #fuel
22:04 preilly joined #fuel
22:16 jaypipes joined #fuel
22:42 vsedelnik joined #fuel
23:43 vsedelnik joined #fuel
23:45 jt708u Hi everyone, i'm looking at testing out fuel-plugin-designate.  First issue i noticed is that although it creates a node_role 'designate', and it shows up in the WebUI for when I add a node I can choose "designate" role.  However, on the CLI if i do 'fuel role|grep designate' nothing shows up.
23:47 jt708u is it supposed to?

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