Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2014-06-19

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

All times shown according to UTC.

Time Nick Message
00:25 rmoe joined #fuel
01:18 xarses joined #fuel
03:37 icheng joined #fuel
03:40 icheng hi , i want get a feature list for Mirantis Openstack  , please have a hand
03:58 vidalinux joined #fuel
04:08 oak2278 joined #fuel
04:16 vkozhukalov joined #fuel
04:51 Arminder Kupo24z: I just used the centos iso to test install from iso
04:52 Arminder i had added iso to glance -> glance image-create --is-public true --name "centos-6.4-x86_64-minimal-iso" --container-format bare --disk-format iso --progress < CentOS-6.4-x86_64-minimal.iso
04:53 Arminder but when I try to create instance, it doesn't allow it to boot from iso
04:53 Arminder in grizzly i never had that issue
04:53 Arminder nor did havana
05:01 oak2278 joined #fuel
05:04 Arminder thats weird, I added the image in glance via horizon
05:05 Arminder and then chose Rebuild Instance option, selected the iso and on boot, it showed the OS install screen
05:15 Arminder but now its stuck in what kind of media from where the installation will take place
05:15 oak2278 joined #fuel
05:15 Arminder and there its unable to detect the local CD/DVD i.e. the iso from which it booted
05:27 Arminder ok so i used netinstall option and provcided a URL and now it comes to partitioning
05:27 Arminder the 80GB ephemeral I had used is showing up as /dev/sda and iso9660 type, not as disk
05:28 Arminder anyone did got a chance to make boot from iso work properly
06:12 neophy joined #fuel
06:19 vkozhukalov joined #fuel
06:22 oak2278 joined #fuel
06:44 al_ex6 joined #fuel
06:45 e0ne joined #fuel
07:07 mrasskazov joined #fuel
07:07 izinovik joined #fuel
07:07 MiroslavAnashkin joined #fuel
07:07 holser joined #fuel
07:20 al_ex6 joined #fuel
07:22 Kupo24z1 joined #fuel
07:22 Kupo24z1 getting extremly slow downloads in an instance (~4K/s) and im getting 1000Mbit on the compute host. Anything I can check?
07:24 Kupo24z1 seems to not matter what OS it is either
07:25 iunruh_ joined #fuel
07:25 sanek_ joined #fuel
07:27 mattymo joined #fuel
07:28 Kupo24z1 for instance downloading http://cachefly.cachefly.net/100mb.test results in 344 bytes/sec in windows, similar in centos & fedora. However on the host node im getting 100MB/s every time
07:28 AndreyDanin joined #fuel
07:28 jeremydei joined #fuel
07:28 ogelbukh joined #fuel
07:42 sbartel joined #fuel
07:42 icheng joined #fuel
07:56 saibarspeis joined #fuel
08:05 ddmitriev joined #fuel
08:05 e0ne joined #fuel
08:23 e0ne joined #fuel
08:26 Arminder- joined #fuel
09:11 e0ne_ joined #fuel
09:21 lromagnoli joined #fuel
09:39 e0ne joined #fuel
09:41 ykotko joined #fuel
09:45 lromagnoli hallo any of u was able to run kvm inside virtualbox? even if i enable VT-x/AMDV the output of grep -E 'svm|vmx' /proc/cpuinfo is null and when i tray to lunch a virtual machine on log i found node-12 nova-nova.compute.manager ERROR: Error: internal error no supported architecture for os type 'hvm'
10:27 sbog Ironmagnoli: you can try "VBoxManage modifyvm <vmname> --longmode on"
10:33 ykotko joined #fuel
11:05 e0ne joined #fuel
11:31 artem_panchenko joined #fuel
12:13 TVR_ trying to get multiple, separate floating IP ranges to work is kicking my arse ... has ANYONE been able to get this to work yet?
12:57 BillTheKat joined #fuel
13:11 alexz joined #fuel
14:00 alexz joined #fuel
14:02 Pho_Tai joined #fuel
14:02 TVR_ The OpenStack Networking API does not expose any facility for retrieving the list of available physical networks.
14:02 Pho_Tai left #fuel
14:03 TVR_ this seems like it would be a very handy feature to get the multiple CIDR floating ip networks to work.....
14:10 oak2278 joined #fuel
14:14 TVR_ ok.. maybe I can go about this another way....
14:17 e0ne joined #fuel
14:18 TVR_ if I do a net-show net04_ext I see the provider:physical_network | physnet1 ..... and I figure if I add another router and subnet ... I can get a second floating IP range... and I see my router binds to br-ex on the port "qg-1a8e3143-e6" ..... so there is a mapping somewhere from that port to physnet1....
14:19 TVR_ so... if I add another bridge and port, how do I bind it to a provider:physical_network
14:20 TVR_ or.. or am I smoking crack here and don't know what I am talking about?
14:21 vkozhukalov joined #fuel
14:22 evg TVR_: not exactly that you want, but m.b. it'll help (nice article): http://www.mirantis.com/blog/configuring-floating-ip-addresses-networking-openstack-public-private-clouds
14:23 TVR_ heh.. thank you for the reply... that is open in one of my tabs
14:27 TVR_ the issue there is it default connects the floating IP's to eth0... and it seems that with the setup I am using the /etc/nova/nova.conf is not used.
14:28 TVR_ So... if the consensus is switching off of neuton and using nova networking and then editing the conf file would solve my issue, I will do it.
14:28 oak2278 joined #fuel
14:35 jobewan joined #fuel
14:38 oak2278 joined #fuel
14:39 oak2278 joined #fuel
14:42 baboue joined #fuel
14:43 baboue hello
14:43 TVR_ hello
14:43 baboue got an issue with mirantis5
14:43 baboue insalled the img on a server
14:43 baboue but once the image is isntalled and rebooted none of the services are installed or available
14:44 baboue nginx for example is not there
14:44 baboue so basically, no UI to access, the "fuek --help" command returns no such command
14:44 baboue etc
14:44 baboue fuek = fuel
14:44 baboue there were no errros during installation
14:45 baboue any ideas/suggestions?
14:45 evg baboue: is /usr/bin/fuel exist?
14:46 baboue no, only  /usr/bin/fuelmenu
14:48 evg baboue: it's strange
14:49 baboue yes
14:51 evg i'd say it's impossible. please check md5sum of iso
14:51 baboue I used the .img
14:53 baboue md5sum MirantisOpenStack-5.0.img  673f5545bff815237806e8abbec61b28  MirantisOpenStack-5.0.img
14:54 vkozhukalov joined #fuel
14:54 baboue used dd to copy the img to a USB stick though
14:54 baboue can I check if that somehow could have been corrupted?
14:55 baboue where are the "fuel" bin contained in the expanded img?
14:57 evg baboue: sure, it's corrupted :)
14:57 evg baboue: "fuel" is in /usr/bin/
14:59 evg baboue: let me find img and check md5 myself
15:00 e0ne joined #fuel
15:07 casanch1 joined #fuel
15:25 sbartel hello
15:25 sbartel got a problem with fuel 4.1.1
15:25 sbartel i try to bootstrap 2servers
15:26 sbartel the servers are bootstraped (hotsname + address ip givent by the pxe)
15:26 sbartel this is visible also in fuel logs
15:26 sbartel but in fuel ihm no node discovered
15:26 sbartel neither in nailgun database
15:26 sbartel any clue?
15:28 MiroslavAnashkin sbartel: Please check the network interface on the target servers is still up. Looks similar with missing firmware bug in 5.0
15:28 icheng joined #fuel
15:32 sbog left #fuel
15:36 sbartel MiroslavAnashkin: just checked and they are up -/
15:43 sbog joined #fuel
15:43 blahRus joined #fuel
15:46 MiroslavAnashkin Please try to reboot both nodes to bootstrap one more time if you have not.
15:49 MiroslavAnashkin Then, if issue persists, please check /var/log/nailgun/api.log, for the messages marked with the IP adresses, given to the nodes
15:53 angdraug joined #fuel
15:54 MiroslavAnashkin You may use `tail -f /var/log/nailgun/api.log` - these messages should drop every minute
15:55 sbartel i have tried to reboot them with no success
15:55 sbartel the only message drop every minutes is 2014-06-19 15:53:57.399 DEBUG [7f7b8f5fe700] (logger) Request GET /api/notifications?_=1403190332021 from 10.193.118.30:46403 2014-06-19 15:53:57.401 DEBUG [7f7ba56cf700] (logger) Request GET /api/nodes/allocation/stats?_=1403190332020 from 10.193.118.30:46404 2014-06-19 15:53:57.421 DEBUG [7f7b8f5fe700] (logger) Response code '200 OK' for GET /api/notifications?_=1403190332021 from 10.193.118.3
15:56 sbartel no link with the ipaddress allocated which are part of the 10.20.0.0/24
15:57 e0ne joined #fuel
15:58 MiroslavAnashkin So, nothing comes from the nodes.
15:59 MiroslavAnashkin Then, please go to the bootstrap console of one of these nodes and check NIC settings with ip a. If OK - try to run the following:
16:03 e0ne joined #fuel
16:05 sbartel eth0 is set on bootproto dhcp  with ip 10.20.0.172 broadcast 10.20.0.255 and netmask 255.255.255.0, so nothing unusual there
16:06 sbartel and it is possible to ping or ssh to the target server from the fuel master node
16:06 Kupo24z1 MiroslavAnashkin: getting extremly slow downloads in an instance (~4K/s) and im getting 1000Mbit on the compute host.  for instance downloading http://cachefly.cachefly.net/100mb.test results in 344 bytes/sec in windows, similar in centos & fedora. However on the host node im getting 100MB/s Anything I can check?
16:08 xarses joined #fuel
16:08 christopheraedo Kupo24z1 which hypervisor are you using?
16:08 Kupo24z1 Ubuntu
16:10 christopheraedo If the vm is not using the kvm guest drivers, it's possible for the VM to be wired with a 10mb virtual nic.  Also potential for KVM (the hypervisor) to be defaulting to the slower virtual nic.  I have to run but I can find more info for you in 2 hrs or so (or else maybe someone here can give you pointers on checking which virtual nic KVM is assigning)
16:11 christopheraedo the details will be in the XML file on the compute node
16:11 MiroslavAnashkin sbartel: Then please run once the following `/opt/nailgun/bin/agent >> /var/log/nailgun-agent.log 2>&1`
16:12 MiroslavAnashkin On the bootstrapped node
16:13 MiroslavAnashkin Then check /var/log/nailgun-agent.log on this node
16:13 sbartel there seems to be on error
16:14 sbartel : /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:31:in `gem_original_require': no such file to load -- ohai/system (LoadError)         from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:31:in `require'         from /opt/nailgun/bin/agent:21
16:16 MiroslavAnashkin Do you use original bootstrap image or customized one?
16:17 sbartel it is a generated iso (generated with fuel-main branch stable/4.1)
16:18 sbartel and custom fuel-library and fuel-web
16:19 sbartel custom modules addes on fuel-lib for controler nodes and some textbox added on fuel web
16:20 MiroslavAnashkin Kupo24z1: Please check https://bugs.launchpad.net/fuel/+bug/1275650 - there is description on where to dig.
16:22 MiroslavAnashkin Looks like broken dependency in the bootstrap image (initramfs.img)
16:22 Kupo24z1 MiroslavAnashkin: im running 5.0 currently, should i try toggling 'GSO'?
16:23 MiroslavAnashkin Please try `ethtool -K eth1 gro off` first on all related physical NICs on the nodes
16:24 MiroslavAnashkin or both
16:24 Kupo24z1 do i need to restart anything?
16:24 MiroslavAnashkin No
16:24 MiroslavAnashkin ethtool applies settings immediately
16:25 Kupo24z1 MiroslavAnashkin: same issue
16:25 Kupo24z1 applied it to all compute nodes
16:26 MiroslavAnashkin sbartel: please try perform deep_clean before the new ISO build. Such errors possible when your local repository is not synchronized with the remote one
16:27 MiroslavAnashkin Kupo24z1: please apply to the controllers as well
16:28 Kupo24z1 MiroslavAnashkin: sweet that worked everything going at 100Mbps now
16:28 Kupo24z1 thanks
16:29 MiroslavAnashkin Then please follow the links to the change requests in the bug for details on how to configure NICs to make this change persistent between reboots
16:32 MiroslavAnashkin sbartel: Or - simply get the existing vanilla initramfs.img from Fuel media, if you have not added moduled to it, and change it on the deployed master node
16:34 MiroslavAnashkin sbartel: HowTo is described in http://docs.mirantis.com/fuel/fuel-5.0/release-notes.html#known-issues-in-mirantis-openstack-5-0 under the Bootstrap kernel issues on certain hardware
16:40 rmoe joined #fuel
16:41 sbartel joined #fuel
16:43 sbartel MiroslavAnashkin: sorry network issu
16:43 e0ne joined #fuel
16:43 sbartel => disconenction
16:43 sbartel there will be no problem to use initramfs.img 5.0 with 4.1.1?
16:44 MiroslavAnashkin No, both are compatible
16:44 MiroslavAnashkin But 5.0, shipped with ISO has missing firmware.
16:45 MiroslavAnashkin So, better to use older one, say this one, from release notes: http://9f2b43d3ab92f886c3f0-e8d43ffad23ec549234584e5c62a6e24.r60.cf1.rackcdn.com/bootstrap-5.0-kernel-2.6.zip
16:46 sbartel ok i'll try with this one
16:48 sbartel thank you
17:01 lromagnoli joined #fuel
17:21 lromagnoli joined #fuel
17:29 e0ne joined #fuel
17:29 mutex is it just me or did 5.0 ship with a lot of big bugs
17:53 e0ne joined #fuel
18:10 lromagnoli joined #fuel
18:10 rmoe joined #fuel
18:10 xarses joined #fuel
18:10 angdraug joined #fuel
18:10 blahRus joined #fuel
18:10 sbog joined #fuel
18:10 icheng joined #fuel
18:10 casanch1 joined #fuel
18:10 jobewan joined #fuel
18:10 ykotko joined #fuel
18:10 Arminder- joined #fuel
18:10 saibarspeis joined #fuel
18:10 ogelbukh joined #fuel
18:10 jeremydei joined #fuel
18:10 AndreyDanin joined #fuel
18:10 mattymo joined #fuel
18:10 sanek_ joined #fuel
18:10 iunruh joined #fuel
18:10 Kupo24z1 joined #fuel
18:10 holser joined #fuel
18:10 MiroslavAnashkin joined #fuel
18:10 izinovik joined #fuel
18:10 mrasskazov joined #fuel
18:10 bogdando joined #fuel
18:10 epsas joined #fuel
18:10 dilyin joined #fuel
18:10 obcecado_ joined #fuel
18:10 bas___ joined #fuel
18:10 dmitryme joined #fuel
18:10 Llew joined #fuel
18:10 ChanServ joined #fuel
18:10 Bomfunk_ joined #fuel
18:10 TVR_ joined #fuel
18:10 christopheraedo joined #fuel
18:10 evgeniyl joined #fuel
18:10 wayneeseguin joined #fuel
18:10 GeertJohan joined #fuel
18:10 aedocw joined #fuel
18:10 scroiset_ joined #fuel
18:10 jseutter joined #fuel
18:10 meow-nofer_ joined #fuel
18:10 boris-42 joined #fuel
18:10 Kupo24z joined #fuel
18:10 alex_didenko joined #fuel
18:10 mutex joined #fuel
18:10 dpyzhov joined #fuel
18:10 aglarendil joined #fuel
18:10 book` joined #fuel
18:10 jbellone joined #fuel
18:10 xenolog joined #fuel
18:10 vkramskikh joined #fuel
18:10 monester joined #fuel
18:10 evg joined #fuel
18:10 kat_pimenova joined #fuel
18:10 jkirnosova joined #fuel
18:10 apalkina joined #fuel
18:10 akislitsky joined #fuel
18:10 MorAle joined #fuel
18:10 meow-nofer joined #fuel
18:10 mihgen joined #fuel
18:10 dburmistrov joined #fuel
18:10 nurla joined #fuel
18:10 justif joined #fuel
18:16 lromagnoli joined #fuel
18:16 rmoe joined #fuel
18:16 xarses joined #fuel
18:16 angdraug joined #fuel
18:16 blahRus joined #fuel
18:16 sbog joined #fuel
18:16 icheng joined #fuel
18:16 casanch1 joined #fuel
18:16 jobewan joined #fuel
18:16 ykotko joined #fuel
18:16 Arminder- joined #fuel
18:16 saibarspeis joined #fuel
18:16 ogelbukh joined #fuel
18:16 jeremydei joined #fuel
18:16 AndreyDanin joined #fuel
18:16 mattymo joined #fuel
18:16 sanek_ joined #fuel
18:16 iunruh joined #fuel
18:16 Kupo24z1 joined #fuel
18:16 holser joined #fuel
18:16 MiroslavAnashkin joined #fuel
18:16 izinovik joined #fuel
18:16 mrasskazov joined #fuel
18:16 bogdando joined #fuel
18:16 epsas joined #fuel
18:16 dilyin joined #fuel
18:16 obcecado_ joined #fuel
18:16 bas___ joined #fuel
18:16 dmitryme joined #fuel
18:16 Llew joined #fuel
18:16 ChanServ joined #fuel
18:16 Bomfunk_ joined #fuel
18:16 TVR_ joined #fuel
18:16 christopheraedo joined #fuel
18:16 evgeniyl joined #fuel
18:16 wayneeseguin joined #fuel
18:16 GeertJohan joined #fuel
18:16 aedocw joined #fuel
18:16 scroiset_ joined #fuel
18:16 jseutter joined #fuel
18:16 meow-nofer_ joined #fuel
18:16 boris-42 joined #fuel
18:16 Kupo24z joined #fuel
18:16 alex_didenko joined #fuel
18:16 mutex joined #fuel
18:16 dpyzhov joined #fuel
18:16 aglarendil joined #fuel
18:16 book` joined #fuel
18:16 jbellone joined #fuel
18:16 xenolog joined #fuel
18:16 vkramskikh joined #fuel
18:16 monester joined #fuel
18:16 evg joined #fuel
18:16 kat_pimenova joined #fuel
18:16 jkirnosova joined #fuel
18:16 apalkina joined #fuel
18:16 akislitsky joined #fuel
18:16 MorAle joined #fuel
18:16 meow-nofer joined #fuel
18:16 mihgen joined #fuel
18:16 dburmistrov joined #fuel
18:16 nurla joined #fuel
18:16 justif joined #fuel
18:16 bookwar joined #fuel
18:20 xarses vnc console was broken in 4.1
18:21 xarses dhcp workes fin if you didn't go into fuelmenu =/
18:21 xarses I do agree that that one is annoying
19:05 vkozhukalov joined #fuel
19:28 e0ne joined #fuel
19:31 casanch1_ joined #fuel
19:36 e0ne joined #fuel
19:41 e0ne joined #fuel
19:49 e0ne joined #fuel
19:57 e0ne joined #fuel
20:08 casanch1 joined #fuel
20:18 mitcha joined #fuel
20:19 mitcha hello, I have a quick question about my fuel install... the controller is getting an error about the network interfaces not coming up when its to the point of "installing openstack" and I don't know what the login is for the controller node after the ubuntu 12.04 install
20:34 casanch1 joined #fuel
20:36 evg mitcha: it's not clean about your controller's interfaces, login to your controller with ssh from master node (by ssh keys without password)
20:36 mitcha evg: what username would i login with?
20:37 evg mitcha: from master node just ssh from root account
20:37 mitcha evg: root doesn't work
20:38 mitcha evg: I know it did with centos nodes, but I used ubuntu.
20:38 evg mitcha: ssh to master node root/r00tme
20:39 mitcha evg: doesn't work
20:39 evg mitcha: then just ssh to controller
20:39 evg you cannot ssh to master?
20:39 mitcha evg: they respond to ssh, but normal "r00tme" password doesn't work on the root account
20:40 mitcha evg: typically ubuntu doesn't use the root account so was wondering if fuel setup some other account
20:41 evg mitcha: stop. it's centos on master node
20:41 mitcha yeah... i can ssh into the master
20:41 mitcha evg: just can't get into the designated controller nor any of the other nodes I installed
20:42 evg mitcha: ok. then just run "ssh _ip_node_"
20:42 evg mitcha: from master
20:43 mitcha evg: I have... [root@fuel log]# ssh 10.17.90.12 Warning: Permanently added '10.17.90.12' (RSA) to the list of known hosts. root@10.17.90.12's password:
20:43 mitcha just asks for a password
20:44 mitcha evg: r00tme doesn't work
20:44 angdraug joined #fuel
20:46 evg mitcha: try to ssh to ip from your admin network
20:46 mitcha evg: 10.17.90.0/24 is the admin network
20:48 mitcha evg: Fuel master is: 10.17.90.5    installed 5 nodes and I can ping all 5 other IP's and get ssh to respond... just can't get into any of them... All but the controller show "Ready" from my deploy this morning
20:48 evg mitcha: hmm, cannot understand
20:49 evg mitcha: is it vm or hw intsallation?
20:50 mitcha evg: hw installed
20:51 evg mitcha: i'm thinking...
20:52 mitcha evg: no worries thanks for helping
20:52 mihgen mitcha: it's likely wrong node then
20:52 mihgen can you try "fuel node"
20:52 mihgen to get list of nodes, then there is gonna be ID of node in the left column
20:53 mihgen then try "ssh node-<ID>"
20:53 mitcha mihgen: ahh
20:54 mitcha mihgen: sure enough it was, they aren't responding to pings then
20:54 evg mitcha: ssh "node_ip"
20:55 mitcha evg: yup... "node_ip" was a guess from a nmap ping sweep from outside the network...
20:55 mihgen if they are not pingable, then you lost your node :) try to connect over KVM/iLO whatever and take a look what's up there....
20:55 mitcha evg: fuel node showed the proper ip
20:56 evg mihgen: no, hi wrote hi can ping them
20:57 mitcha mihgen: evg: ideas on what I should look for that would cause my "controller" + "mongodb" to fail the openstack install phase?
20:58 evg mitcha: could uoy please just ssh to your nodes before
20:59 mitcha evg: I'm now connected to the controller node
20:59 evg mitcha: sorry?
20:59 evg mitcha: you dont have access to your env?
21:00 mitcha evg: .... I'm connected to it now..
21:07 evg mitcha: oh, sorry i've misred
21:10 evg mitcha: so, ssh to controller works?
21:11 mitcha evg: yes
21:27 mitcha evg: looking through puppet logs, seeing alot of "Skipping because of failed dependencies"
21:27 mitcha evg: trying to get to the begining to see what failed
21:29 mitcha evg: Error: /Stage[netconfig]/L23network::L2/Package[ifenslave]/ensure: change from purged to present failed: Execution of '/usr/bin/apt-get -q -y -o DPkg::Options::=--force-confold --force-yes install ifenslave' returned 100: Reading package lists...
21:30 mitcha evg: would seem its failing because a package install failed somewhere...
21:32 mitcha evg: dpkg: error processing qemu-kvm (--configure):  subprocess installed post-installation script returned error exit status 1
21:33 Kupo24z1 joined #fuel
21:52 mitcha evg: looks like its a known bug: https://bugs.launchpad.net/fuel/+bug/1316812
21:53 mitcha There are updated ISO images... but... don't really want to reload everything...
22:32 lromagnoli joined #fuel
23:04 oak2278 joined #fuel

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