Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2016-05-19

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

All times shown according to UTC.

Time Nick Message
00:09 xarses rps2: please clarify, what do you mean "fuel will not display the port"?
00:11 xarses also, where are these bridges coming from? they don't follow fuel's naming standard
00:12 rps2 xarses: I'm getting this second-hand from the guy trying the deployment. From what I can gather, the virtual machines are running in libvirt under KVM and not virtualbox for one thing.
00:13 xarses ok, so for libvirt based envs, the best thing to do is to use our fuel-devops scripts, and the setupmaster testcase from fuel-qa
00:14 xarses these are the same tools we use to develop and test on libvirt/kvm
00:14 rps2 The KVM host has bridges set up for the four networks that the VMs will use and are named "br1" through "br4". This was done because of conflicting reports that you should have bridges and not NATted interfaces.
00:15 xarses if you are using virtualbox, you should use the virtual box helper scripts. same story
00:16 rps2 Uhm, ok. I'll let him know. He's been futzing around with this for over a week. I've never even seen this stuff before.
00:16 rps2 :-(
00:18 xarses http://paste.openstack.org/show/497631/
00:19 xarses this is how my net's look like when set up by fuel-devops
00:20 xarses rps2: this should be up to date https://docs.fuel-infra.org/fuel-dev/devops.html#installation
00:21 xarses sorry, you want this list https://docs.fuel-infra.org/fuel-dev/devops.html#environment-creation-via-devops-fuel-qa-or-fuel-main
00:24 xarses and I run the setup master test class with https://gist.github.com/xarses/80e07a88c1b63c40cd1c783e03c03e30
00:25 rps2 Ok. I queried the guy. Essentially the intent is to have three physical machines. One is the jumpstart server running Ubuntu 14. Fuel is run as a libvirt VM on that machine which has booted the Fuel iso image.
00:26 rps2 A second physical machine running Ubuntu 14 will run two libvirt VMs on it: one VM as a controller and one VM as a compute node. The third physical machine would be another Ubuntu system that would run additional compute nodes as VMs under libvirt.
00:34 byrdog55 joined #fuel
00:35 fatdragon joined #fuel
00:37 xarses there are docs somewhere about how to use devops to do multiple node groups, this should detail the steps needed to have the bridges to cross communicate, it assumes that they are separate L2 networks on the same KVM host, but that should be enough clues on how to plumb the bridges
00:38 rps2 Because of this use of several physical machines, we've had to use multiple physical NICs on each machine to implement the networks required. Because of some restrictions regarding the numbers of NICs available, one has had to be pressed into service to handle two networks so we used VLAN tagging to do it. The untagged traffic on that NIC is for PXEboot, the tagged traffic was for the "public" network.
00:41 rps2 The gotcha is that doesn't seem to work in the GUI. The chap doing this is having all kinds of issues and he's told me that the NIC with tagging doesn't appear as an option you can select in the GUI (neither the native, untagged part called "br1" nor the tagged part called "br1.51").
00:41 rps2 I haven't seen what he's talking about since every time I go into his office, he's "started over" and it's not in a state I can look at. Grrrrr!
01:23 gardlt joined #fuel
01:29 raunak 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/
01:52 fatdragon joined #fuel
02:14 v1k0d3n joined #fuel
02:21 byrdog55 joined #fuel
02:55 xarses joined #fuel
03:03 raunak joined #fuel
03:39 jobewan joined #fuel
03:43 book` joined #fuel
03:47 mciccone joined #fuel
04:26 cartik joined #fuel
04:59 jobewan joined #fuel
05:35 javeriak joined #fuel
05:40 nineseconds joined #fuel
05:42 jobewan joined #fuel
05:50 xarses_ joined #fuel
05:51 xarses_ joined #fuel
06:11 fatdragon joined #fuel
06:22 AlexPost joined #fuel
06:25 kszukielojc joined #fuel
06:51 LanceHaig joined #fuel
06:58 Miouge joined #fuel
07:15 akaszuba joined #fuel
07:20 javeriak joined #fuel
07:45 Miouge joined #fuel
07:48 mciccone joined #fuel
07:57 HeOS joined #fuel
08:07 Miouge joined #fuel
08:10 Miouge joined #fuel
08:27 hezhiqiang joined #fuel
08:30 fatdragon joined #fuel
08:33 noshankus joined #fuel
08:36 Julien-zte joined #fuel
08:40 yassine joined #fuel
09:05 venkat joined #fuel
09:13 LanceHaig joined #fuel
09:19 Miouge joined #fuel
09:29 AlexPost joined #fuel
09:30 parlos joined #fuel
09:42 chopmann joined #fuel
10:12 javeriak joined #fuel
10:31 Miouge joined #fuel
10:31 fatdragon joined #fuel
10:37 Miouge joined #fuel
10:55 rmoe joined #fuel
10:58 javeriak joined #fuel
11:00 DavidRama joined #fuel
11:01 DavidRama Hi, what is the command line to get the fuel inital config screen ?
11:02 DavidRama or how do I change the ACL set for accessing the fuel node using ssh ?
11:05 DavidRama found the answer sorry
11:19 byrdog55 joined #fuel
11:19 DavidRama ok I changed the security in the fuelmenu to 0.0.0.0/0 and still can't ssh to the host any idea ?
11:20 cartik is it possible to update neutron alone to mitaka from kilo?
11:30 mmalchuk joined #fuel
11:32 fatdragon joined #fuel
11:50 mciccone joined #fuel
12:02 DavidRama Hi trying to add a node in a running config using fuel9.0 and got this errors: 2016-05-19 11:02:48
12:02 DavidRama WARNING
12:02 DavidRama 11:02:48.131312 #2265]  WARN -- : netio.rb:387:in `_init_line_read' PLMC7: Exiting after signal: SignalException: SIGTERM
12:02 DavidRama 2016-05-19 11:01:29
12:02 DavidRama ERR
12:02 DavidRama 11:01:28.429382 #2265] ERROR -- : agents.rb:71:in `rescue in loadagent' Loading agent puppetd failed: cannot load such file -- puppet
12:02 DavidRama 2016-05-19 11:01:29
12:02 DavidRama ERR
12:02 DavidRama 11:01:28.429320 #2265] ERROR -- : pluginmanager.rb:171:in `rescue in loadclass' Failed to load MCollective::Agent::Puppetd: cannot load such file -- puppet
12:20 javeriak joined #fuel
12:21 asvechnikov DavidRama, Hi, what's exactly you have tried to do?
12:21 javeriak_ joined #fuel
12:22 DavidRama Hi adding a compute node + osd
12:25 asvechnikov DavidRama, to already deployed cluster?
12:25 DavidRama yes
12:25 DavidRama provision is ok, deploy fails
12:30 asvechnikov DavidRama, Could you take a look at /var/log/puppet.log on failed node? What is error there?
12:33 DavidRama ok
12:34 DavidRama I have no errors, last lines:
12:34 DavidRama 2016-05-19 11:31:08
12:34 DavidRama NOTICE
12:34 DavidRama Finished catalog run in 2.35 seconds
12:34 DavidRama 2016-05-19 11:31:08
12:34 DavidRama NOTICE
12:34 DavidRama Finished catalog run in 2.35 seconds
12:34 DavidRama 2016-05-19 11:31:08
12:34 DavidRama NOTICE
12:34 DavidRama (/Stage[main]/Ntp::Service/Service[ntp]) Triggered 'refresh' from 1 events
12:34 DavidRama 2016-05-19 11:31:08
12:34 DavidRama NOTICE
12:34 DavidRama (/Stage[main]/Ntp::Service/Service[ntp]) Triggered 'refresh' from 1 events
12:34 DavidRama 2016-05-19 11:31:06
12:34 DavidRama NOTICE
12:35 DavidRama oh sorry
12:35 DavidRama got some
12:35 DavidRama 2016-05-19 11:24:01
12:35 DavidRama ERR
12:35 DavidRama /usr/bin/puppet:8:in `<main>'
12:35 DavidRama 2016-05-19 11:24:01
12:35 DavidRama ERR
12:35 DavidRama joined #fuel
12:36 DavidRama let met put a link with that :D
12:37 DavidRama Errors in puppet logs: http://postimg.org/image/ebtvf9dj5/
12:43 Miouge joined #fuel
12:44 DavidRama asvechnikov you saw my link ?
12:45 asvechnikov DavidRama, please check existence of the task `rsync_core_puppet` in this log
12:46 asvechnikov DavidRama, is it there? Also you may check existence of the folder `/etc/puppet/modules`
12:46 DavidRama let me check
12:48 DavidRama in puppet.log I have:
12:48 DavidRama 2016-05-19 11:24:03 +0000 Scope(Class[main]) (notice): MODULAR: rsync_core_puppet
12:48 DavidRama 2016-05-19 11:24:03 +0000 /Stage[main]/Main/Exec[rsync_core_puppet_shell]/returns (notice): skipping non-regular file "osnailyfacter/modular/master"
12:48 DavidRama 2016-05-19 11:24:03 +0000 /Stage[main]/Main/Exec[rsync_core_puppet_shell]/returns (notice): executed successfully
12:49 DavidRama root@node-11:~# ls /etc/puppet/modules/ceph/
12:49 DavidRama examples  Gemfile  lib  manifests  Rakefile  README.md  templates
12:50 asvechnikov DavidRama, what about /etc/puppet/modules/osnailyfacter/modular/ceph ?
12:51 Miouge joined #fuel
12:56 DavidRama root@node-11:~# ls /etc/puppet/modules/osnailyfacter/modular/ceph
12:56 DavidRama ceph_compute_post.rb  ceph-osd.pp    ceph_ready_check.rb  mon.pp               radosgw_post.rb  radosgw_pre.rb  updatedb.pp
12:56 DavidRama ceph_compute.pp       ceph_pools.pp  enable_rados.pp      radosgw_keystone.pp  radosgw.pp       tasks.yaml
13:04 asvechnikov DavidRama, what is iso build id you use? You can find it on master node /etc/fuel_build_id
13:08 DavidRama joined #fuel
13:10 MikeG451 joined #fuel
13:30 Julien-zte joined #fuel
13:34 fatdragon joined #fuel
13:34 Julien-z_ joined #fuel
13:38 javeriak joined #fuel
13:55 chopmann joined #fuel
14:03 mciccone joined #fuel
14:05 mciccone_ joined #fuel
14:05 mciccone joined #fuel
14:35 fatdragon joined #fuel
14:37 raunak joined #fuel
14:40 parlos joined #fuel
14:45 mciccone joined #fuel
14:53 Reepicheep joined #fuel
14:56 mciccone joined #fuel
15:04 gardlt joined #fuel
15:22 nurla joined #fuel
15:36 fatdragon joined #fuel
15:43 mciccone joined #fuel
15:54 gardlt joined #fuel
15:58 fatdragon joined #fuel
16:00 gomarivera joined #fuel
16:30 brig-man Where do you store the credentials for the fuel cli?  This is for 8.0.
16:32 brig-man I tried to switch my bootstrap image and gave me this:
16:33 brig-man fuel-bootstrap activate centos
16:33 brig-man WARNING: switching to depracated centos-bootstrap
16:33 brig-man Starting new HTTP connection (1): 10.20.0.2
16:33 brig-man Starting new HTTP connection (1): 10.20.0.2
16:33 brig-man Starting new HTTP connection (1): 10.20.0.2
16:33 brig-man The request you have made requires authentication. (HTTP 401) (Request-ID: req-8c80a6a0-f54d-4725-83f1-ea1fd8ee3204)
16:33 brig-man Now nothing boots.
16:33 srwilkers joined #fuel
16:33 brig-man The node cannot find an image to boot
16:35 raunak joined #fuel
16:39 sbfox joined #fuel
16:49 Jabadia_ joined #fuel
16:53 DavidRama Hi, can someone explaines what provisioning OS only host is for ? is this a MAAS like function ? ie provisionning basic bare-metal ?
16:54 brig-man I switched back and it now find the ubuntu image.  Problem with those is that they just continue to reboot after the first/pxe interface comes up.
16:56 mwhahaha brig-man: you'll probably need to clear out those systems from cobbler
16:57 mwhahaha DavidRama: it just drops the OS on only using the same image that is used to provision the other nodes. But we don't do any network configuration or anything beyond that. It's just an OS deployment essentially
16:58 brig-man I had these nodes up and running before but I used the centos image which works.
16:59 DavidRama mwhahaha meaning I can log to it and donfigure it and later on deprovision it via fuel ?
17:00 mwhahaha DavidRama: Yea i think so, as long as the mcollective agent is still running on it and it can communicate with the master
17:00 DavidRama This is done via the PXE LAN ?
17:00 mwhahaha yea
17:01 DavidRama ok I'll try it out
17:07 brig-man mwhahaha: Removing nodes from cobbler?
17:07 mwhahaha yea cobbler runs on the master to provide the bootstrap image
17:08 mwhahaha you may try removing them from the cobbler instance on the master to get them to reregister
17:12 brig-man Where is this or command?
17:19 mwhahaha the cobbler command can be used to query/manage cobbler
17:20 mwhahaha http://cobbler.github.io/manuals/quickstart/
17:20 krobzaur joined #fuel
17:42 javeriak joined #fuel
17:52 kur1j joined #fuel
17:53 kur1j What would cause a system to get to the PXE boot screen but never progress past that and start booting?
17:54 kur1j I'm using 9.0 .
17:54 kur1j Would 10gb nic cards be an issue?
17:55 brig-man kurl: No network response from the Fuel node.  Typically a network setup.
17:56 brig-man kurl: could be the 10gb cards.  I was seeing something about 10gb cards not working correctly on IPXE.
17:56 kur1j Hmmm
17:56 kur1j That's not good
17:57 kur1j What would be my options?
17:57 brig-man kurl: Is the pxe interface a 10gb interfaces?
17:57 kur1j Yes
17:57 kur1j It's on an esxi machine
17:58 kur1j I'm trying to get OS stack setup on 32 node blade system
17:58 kur1j Each node has 2 10gb NICs and that's it
17:59 brig-man I would go back to the vendor and check for a firmware update to the nodes.
17:59 kur1j Hmm
18:00 DavidRama left #fuel
18:02 kur1j So it apparently worked over the 10gb card if the fuel node vm network cards were made to be e1000
18:08 Jabadia joined #fuel
18:26 seeg_ joined #fuel
18:26 Captain_Murdoch joined #fuel
18:26 Captain_Murdoch joined #fuel
18:27 dmellado joined #fuel
18:29 Obi-Wan joined #fuel
18:30 srwilkers joined #fuel
18:33 kgalanov joined #fuel
18:33 teran joined #fuel
18:33 igorbelikov joined #fuel
18:33 holser joined #fuel
18:33 mwhahaha joined #fuel
18:35 apopovych joined #fuel
19:04 MikeG451 joined #fuel
19:05 srwilkers joined #fuel
19:11 brig-man I'm having problems getting my nodes to complete the loading of a discovery image.  It gets to link ready and waits then reboots.
19:12 brig-man This is with the ubuntu image.  These did load previously when I was getting a menu and selected centos.
19:21 kur1j joined #fuel
19:22 kur1j Ok after updating the esxi machine I see it got "!PXE structure was not found in UNDI driver code segment"
19:24 DavidRama joined #fuel
19:26 DavidRama joined #fuel
19:27 DavidRama joined #fuel
19:38 Jabadia joined #fuel
19:41 xenthree3 joined #fuel
19:41 xenthree3 left #fuel
19:53 cartik joined #fuel
19:53 gomarivera joined #fuel
19:57 srwilkers joined #fuel
20:13 HeOS joined #fuel
20:21 raunak joined #fuel
20:25 gomarivera joined #fuel
20:29 Zer0Byte__ joined #fuel
20:29 Zer0Byte__ hey
20:29 Zer0Byte__ someone have idea how to delete files from this path
20:29 Zer0Byte__ on the fuel node
20:30 Zer0Byte__ .../cobbler/boot/pxelinux.cfg/
20:30 Jabadia joined #fuel
20:30 Zer0Byte__ for one strange reason
20:30 Zer0Byte__ i delete the enviroment
20:30 Zer0Byte__ and one controller node still try to boot from the hard drive
20:34 kur1j Why would a network card show "down" when it is up and has a green link light?
20:35 LGee What tool says it's down?
20:37 kur1j Fuel
20:37 kur1j I logged into the machine and did ifconfig and it doesn't show the device
20:37 kur1j But it is listed in the hardware
20:38 kur1j Ah hell I figured it out
20:38 LGee Do an ifconfig -a, plain ifconfig only shows ifaces that are up
20:38 kur1j So I guess fuel doesn't config it if it can't get correct networking information
20:41 kur1j I could have sworn that fuel would show "up" if the link light was lit
20:41 kur1j But I guess not
20:46 mwhahaha Zer0Byte__: that can be managed via the cobbler command (remove the old node)
20:49 Zer0Byte__ got it
21:02 gardlt joined #fuel
21:19 gardlt joined #fuel
21:20 gomarivera joined #fuel
21:20 brig-man I just got two nodes that dropped into grub rescue after they finished installing ubuntu.  Ideas?
21:34 elo joined #fuel
21:38 gardlt joined #fuel
21:57 eric_lopez joined #fuel
22:12 HeOS joined #fuel
22:39 srwilkers joined #fuel
22:55 srwilkers joined #fuel
23:06 Zer0Byte__ someone can tell me why
23:06 Zer0Byte__ if i delete a system with fuel
23:06 Zer0Byte__ i have to delete with cobbler also?
23:06 Zer0Byte__ ant not automatically remove from there?
23:23 Zer0Byte__ rghhh
23:23 Zer0Byte__ i delete from cobbler
23:23 Zer0Byte__ and the computer still booting from the hard drive
23:26 ericlopez joined #fuel
23:53 Zer0Byte__ solved the problem
23:53 Zer0Byte__ but the question is
23:54 Zer0Byte__ why i have to delete the comptuer manually from cobbler

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