Camelia, the Perl 6 bug

IRC log for #fuel, 2013-10-31

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

All times shown according to UTC.

Time Nick Message
00:03 rmoe_ joined #fuel
00:22 rmoe joined #fuel
01:27 xarses joined #fuel
01:47 fandikurnia01 joined #fuel
01:58 fandikurnia01 joined #fuel
02:58 vkozhukalov joined #fuel
03:16 IlyaE joined #fuel
03:27 ArminderS joined #fuel
03:41 tonyha joined #fuel
03:43 mihgen joined #fuel
06:27 ArminderS joined #fuel
06:28 teran joined #fuel
06:44 e0ne joined #fuel
06:56 ogzy joined #fuel
06:57 ogzy MiroslavAnashkin: i tried running the launch.sh script on the host machine without connecting it ssh, it is still the same, not connecting to 10.20.0.2 via ssh and the installation is not progressing
06:57 ogzy MiroslavAnashkin: i enabled IP forwarding
06:58 ogzy MiroslavAnashkin: ? (10.20.0.2) at <incomplete> on vboxnet0 is what arp -a says
06:58 ogzy MiroslavAnashkin: how do you run this script on ubuntu server?
07:14 ArminderS do you have a network 10.20.0.x?
07:15 ArminderS and your host is connected to that?
07:19 ArminderS the birtualbox helper scripts create that network
07:20 ArminderS or you can do it manually and try
07:21 ArminderS Configure the host-only interface vboxnet0 in VirtualBox by going to File -> Preferences -> Network and clicking the screwdriver icon.
07:21 ArminderS IP address: 10.20.0.1
07:21 ArminderS Network mask: 255.255.255.0
07:21 ArminderS DHCP Server: disabled
07:21 ArminderS and then attach the first adapter of your vm to this
07:30 mrasskazov1 joined #fuel
07:32 mihgen joined #fuel
07:34 e0ne joined #fuel
07:35 ogzy ArminderS: i found the problem, i manuallay created the vm and it works now, but the downloaded iso is for x86, i need x86-64 compatible of mirantis iso
07:36 ogzy ArminderS: when i boot is says, this kernel requires x86-64 cpu
07:37 ogzy ArminderS: my host os is amd64
07:38 ArminderS do you have nexted virtualization on?
07:38 ArminderS your host support virtualization extensions, right
07:38 ArminderS nested*
07:39 ogzy ArminderS: it should be, didn't check the bios settings
07:39 ogzy ArminderS: just installed virtualbox on it
07:40 ArminderS ok whats the base os?
07:41 ArminderS for your host
07:41 ArminderS ubuntu/centos/windows?
07:42 ArminderS or anything else
07:47 e0ne joined #fuel
07:49 ogzy ArminderS: ubuntu, i enabled virtualization from bios
07:51 ogzy ArminderS: is it possible to create custom mirantis iso?
07:51 e0ne joined #fuel
07:51 ogzy ArminderS: like ubuntu based one istead of centos
07:51 ArminderS can you past teh output for this -> egrep '(vmx|svm)' --color=always /proc/cpuinfo
07:53 ArminderS i mean could you see vmx or svm highlighted
07:53 ArminderS or no output et al?
07:53 ogzy ArminderS: it is started to install now
07:53 ArminderS no need to paste whole output, just for 1
07:53 ogzy ArminderS: after changing the bios setting
07:53 ArminderS great
07:54 ArminderS so virtualization wasn't enabled earlier
07:54 ogzy ArminderS: yes
07:54 ArminderS and its now, so working
07:54 ogzy ArminderS: the fuel iso is centos based, can i create an ubuntu based one?
07:54 ogzy ArminderS: where is the documentation for it?
07:54 ArminderS not sure, haven't yet ventured in that area
07:54 ogzy ArminderS: ok
07:54 ArminderS you need to ask this in #fuel-dev
07:55 ArminderS thats a proper channel for it, i think
07:56 ogzy ArminderS: it is installing som packages, what then, i will create some slave nodes?
07:57 ArminderS you may want to give a shot to the auto installer script
07:57 ArminderS the bagrant one
07:57 ogzy it is most probably the bios setting causing the vm not starting with the launch.sh
07:57 ArminderS vagrant*
07:57 ArminderS that will setup all automatically now
07:57 ogzy ok
07:57 ogzy ArminderS: ok this disk requirements are hig
07:57 ogzy ArminderS: i will make them 10GB, i just have 60GB
07:58 ArminderS use thin provisioning and you can create high sized drives
07:58 ArminderS you will be good unless you don't actually start using that much of space
07:58 e0ne joined #fuel
08:00 ogzy ArminderS: it created 3 slaves with each has 3 disks
08:01 ogzy ArminderS: not possible to test it on the same machine it seems
08:02 ArminderS go on, but keep any eye on free space on your host
08:05 ArminderS i have created a 6-node setup on vmware esxi and have 7-8 images including ISOs, Ceph with replication as 2 & 5 VMs including 1 windows 2012 server and all that is using 120GB
08:06 ArminderS so i guess 60gb might work for you
08:08 mattymo joined #fuel
08:18 vkozhukalov joined #fuel
08:34 aabashkin joined #fuel
08:38 ogzy ArminderS: i dont think so, it created every slave node with 3 disks, each one is at least 15GB
08:39 ogzy ArminderS: i will skip virtualbox installation and try iso booting
08:43 e0ne joined #fuel
08:44 vkozhukalov joined #fuel
09:07 tonyha joined #fuel
09:25 iscsi joined #fuel
09:25 iscsi moin
09:26 ogzy when i install from usb, the ip adress it is assigning is 10.x.x.x , should i chanege it according to my network?
09:29 iscsi http://docs.mirantis.com/fuel/fuel-​3.2/install-guide.html#changing-net​work-parameters-after-installation
09:31 ezziepat joined #fuel
09:34 iscsi ogzy: watch this http://docs.mirantis.com/fuel/fuel-3.2/i​nstall-guide.html#fuel-deployment-schema
09:35 * iscsi stuck on deployment stage
09:42 fandikurnia01 joined #fuel
10:25 iscsi https://dl.dropboxusercontent.com/u/217​0772/fuel-stuck-on-disks-configure.png
10:25 iscsi no errors in logs
10:27 ArminderS because you chose to configure disk for all 14 nodes
10:27 ArminderS seems to be a bug or something
10:27 ArminderS click on nodes and it will ask if you wish to cancel, cancel it
10:28 ArminderS and then try to configure few nodes at a time
10:28 ArminderS or 1 at a time
10:28 ArminderS then it will allow you to click Apply and the apply button will grey out
10:28 ArminderS allowing you to click on "Back to Node List"
10:29 iscsi 3 nodes is ok
10:29 iscsi 4th not
10:29 ArminderS then simply fix the 4th one separately
10:29 ArminderS just chose 4th node and click on configure disk
10:30 ArminderS i tried to do all 3 controllers in 1 go, but it refused to apply, cancelled and found first 2 nodes got the change, but the 3rd one didn't
10:31 ArminderS so configured the 3rd one separately
10:31 ArminderS the network parth though worked fine when i tried all 5 nodes in 1 go
10:34 ArminderS part*
10:36 e0ne_ joined #fuel
10:39 e0ne__ joined #fuel
10:48 iscsi https://dl.dropboxusercontent.co​m/u/2170772/stuck-on-deploy.png
10:49 iscsi this can happen from https://dl.dropboxusercontent.com/​u/2170772/ip-address-mismatch.png ?
11:25 iscsi bug in rhel6.4 ovs kmod (fuel vm host machine)
11:47 mihgen joined #fuel
11:52 aabashkin Guys, FUEL 3.2 creates initial image in Glance. Image has "Container Format"=OVF. is it really OVF or mistake?
12:01 xdeller it`s okay
12:01 teran joined #fuel
12:02 xdeller it can be qcow even if it displayed as ovf, may be kind of bug
12:15 teran joined #fuel
12:17 fandikurnia01 joined #fuel
12:20 e0ne joined #fuel
12:23 mrasskazov1 joined #fuel
13:19 e0ne joined #fuel
13:24 AndreyGrebenniko guys, how can I redeploy one controller with 3.2 if I improved manifests?
13:25 mihgen AndreyGrebenniko: you need to go on controller node and run "puppet agent -t"
13:25 AndreyGrebenniko mihgen, how it will apply facts then?
13:28 mihgen sorry we are on all hands meeting
13:28 mihgen so if you ever ran deploy
13:28 mihgen you will have /etc/astute.yaml
13:28 mihgen check if you have it there
13:29 mihgen otherwise the other option is to use CLI to get JSON's per role
13:29 mihgen if you are on 3.2. If you are on master, there is already YAML format
13:29 mihgen are you based on 3.2 or on master?
13:30 AndreyGrebenniko mihgen, ok, there is a trick with puppet. if I try to ping puppet master from the node it goes to 127.0.0.1 ;)
13:31 AndreyGrebenniko mihgen, not quite clear about "on 3.2 or on master" - I'm in the environment created by fuel3.2
13:32 ogzy i am at the fue interface and trying to define my network structure, i have only one network for internet access and one for management, it seems i need one more?
13:33 ogzy 10.102.235.0/24 fo rpublic
13:33 ogzy 10.102.237.0/24 for management
13:33 ogzy 10.102.236.0/24 for storage
13:33 ogzy it says Intersection with admin network(s) '10.102.237.0/24' found
13:33 ogzy eth0 has an ip with 237
13:33 ogzy though eth1 had 235
13:35 tonyha joined #fuel
13:50 mrasskazov2 joined #fuel
13:54 mrasskazov1 joined #fuel
13:54 e0ne joined #fuel
13:57 MiroslavAnashkin ogzy_: Admin network is used to discover available nodes. You may configure the scope of this network during the master node installation.
13:58 MiroslavAnashkin It should use untagged traffic only and no other IP addresses should be occupied inside this network.
13:59 e0ne joined #fuel
14:01 MiroslavAnashkin Other networks may - and in casse of single physical network must - be VLAN tagged.
14:06 MiroslavAnashkin ogzy_: Please consult docs about Fuel networking scenarios first http://docs.mirantis.com/fuel/fuel-3.2/install-gui​de.html#understanding-and-configuring-the-network
14:06 e0ne_ joined #fuel
14:10 ogzy MiroslavAnashkin: i read them and i am confused
14:11 mihgen AndreyGrebenniko: master branch I meant, not release version
14:12 mihgen looks like there is an issue, if it goes to 127.0.0.1. Check your /etc/hosts on master node
14:12 ogzy MiroslavAnashkin: according to your explanation my eth0 is at teh network 237
14:12 ogzy MiroslavAnashkin: which is admin network
14:12 ogzy MiroslavAnashkin: i have also 235 network
14:13 AndreyGrebenniko mihgen, release
14:13 mihgen it must not have your fuel master FQDN to be 127.0.0.1. If so - it's bug (I thought mattymo fixed it)
14:13 AndreyGrebenniko mihgen, clear))
14:18 MiroslavAnashkin ogzy_: What network type and network segmentation type you are going to implement? Nova or Newtron? Flat, VLAN or GRE?
14:20 mattymo mihgen, fqdn of master should be the IP of eth0 (or whatever is the interface for admin network)
14:21 AndreyGrebenniko mattymo, how should I change it on master so that it could give clients right address?
14:22 mattymo make sure /etc/hosts shows the right ip address for your master
14:22 mattymo then cobbler sync
14:22 AndreyGrebenniko mattymo, ah, forgot about it
14:22 mihgen mattymo: that
14:22 mihgen that's the issue we talked about on syncip
14:22 mihgen syncup*
14:22 AndreyGrebenniko mihgen, )))
14:23 mihgen AndreyGrebenniko: did you use fuel menu?
14:23 mattymo there were a couple of bugs which created /etc/hosts to break
14:23 AndreyGrebenniko mihgen, yes
14:23 AndreyGrebenniko mihgen, great, seems it works now
14:24 mihgen mattymo: we really need to fix these before we fly out ..
14:24 mihgen we want 3.2.1 code freeze on Monday
14:25 mattymo I'm waiting on nastya to QA ntp, which is where I bundled the fix
14:25 mattymo she's already left for the day, so I don't think it can be done tomorrow
14:27 ogzy MiroslavAnashkin: GRE
14:28 ogzy MiroslavAnashkin: and nova
14:34 MiroslavAnashkin ogzy_: Nova supports only Flat and VLAN segmentation
14:36 vnaboychenko joined #fuel
14:36 MiroslavAnashkin ogzy_: So, you may select Nova, select either Flat or VLAN and do the following:
14:37 MiroslavAnashkin ogzy_: Leave eth0 untagged for admin network. it should be physical network
14:38 MiroslavAnashkin ogzy_: Make all the remained networks VLAN tagged - actually it is default settings for each new environment.
14:38 ogzy MiroslavAnashkin: i had choosen Neutron with GRE segmentation
14:39 ogzy MiroslavAnashkin: sorry for confusion
14:40 dhblaz joined #fuel
14:44 MiroslavAnashkin ogzy_: Aah, OK. Then please keep in mind - not all VirtualBox network interfaces need to be connected to physical network.
14:44 ogzy MiroslavAnashkin: i made installation from iso, no virtual machine
14:45 MiroslavAnashkin ogzy_: You have virtual machine with Fuel master node under VirtualBox installed, right&
14:45 MiroslavAnashkin ?
14:45 ogzy MiroslavAnashkin: no
14:46 ogzy MiroslavAnashkin: i tried booting directly from iso, now i have the web interface running
14:46 ogzy MiroslavAnashkin: according to what you said if i chane the network configuration as eth0:1 as the network 10.102.236.* and make the management network 237 and public 235, things will fine i assume
14:47 ogzy MiroslavAnashkin: 236 and 237 are able to talk with each other
14:47 ogzy MiroslavAnashkin: eth0:1will be Fuel network in this case
14:51 MiroslavAnashkin ogzy_: What is eth0:1 ? Is is VLAN 1 on eth0?
14:52 MiroslavAnashkin ogzy_: What is eth0:1 ? Is it VLAN 1 on eth0?
14:56 ogzy MiroslavAnashkin: i have vlan on eth0 10.102.237.x and 10.102.235.x on eth1, eth1 is for internet access
14:56 ogzy MiroslavAnashkin: but i can define 236 also, 236 and 237 have access to each other
14:58 MiroslavAnashkin ogzy_: The best way to understand how Fuel assigns networks is to PXE boot 1-2 slave nodes from master node, create new env with Neutron+GRE, assign these nodes a role (one controller and other compute) and then click Gear icon on any added node and select Configure interfaces
14:58 ogzy MiroslavAnashkin: ok
15:03 IlyaE joined #fuel
15:22 e0ne joined #fuel
15:22 ogzy MiroslavAnashkin: i see nodes offline, why do you think that might be?
15:23 MiroslavAnashkin ogzy: Did  you boot these nodes wia PXE from master node?
15:23 ogzy MiroslavAnashkin: yes
15:24 ogzy MiroslavAnashkin: i can see the nodes on fuel interface but offline
15:27 MiroslavAnashkin ogzy: Does Fuel report there are unallocated nodes? In top right corner.
15:28 ArminderS joined #fuel
15:29 albionandrew joined #fuel
15:29 ogzy MiroslavAnashkin: yes
15:31 MiroslavAnashkin ogzy:  Then go to the Nodes tab of your created env and click Add Nodes button.
15:32 ogzy MiroslavAnashkin: i did, it is not possible to choose them, they look offline
15:33 MiroslavAnashkin ogzy: First, check one of roles mark.
15:33 ogzy MiroslavAnashkin: i did it also, not possible to check the unallocated ones
15:33 ogzy MiroslavAnashkin: it says offline
15:34 MiroslavAnashkin ogzy: Do you have DHCP server in 10.20.0.0/24 network segment?
15:35 ogzy MiroslavAnashkin: i have 10.102.235.0/24 and dhcp yes
15:35 ogzy MiroslavAnashkin: the machines are able to boot via pxe, and i can see the bootstrap login after they bıoı
15:36 ogzy MiroslavAnashkin: these nodes were available first, then i changed the IP address of the fuel master, somehow they are still there even the machines are closed
15:36 MiroslavAnashkin ogzy: No DHCP except the one on the master node should be accessible from slave nodes via network interface you use to PXE boot
15:37 MiroslavAnashkin ogzy: How did you changed IP address on the master node?
15:38 ogzy MiroslavAnashkin: /etc/sysconfig/network-scripts/ifcfg-eth0
15:38 ogzy MiroslavAnashkin: and also dnsmasq setting
15:40 MiroslavAnashkin ogzy: To change IP address on master node one should run bootstrap_admin_node.sh
15:42 ogzy MiroslavAnashkin: that may help
15:42 ArminderS joined #fuel
15:43 MiroslavAnashkin ogzy: It will not work if you change network settings directly on the interfaces.
15:45 e0ne joined #fuel
15:45 ogzy MiroslavAnashkin:
15:45 ogzy KeyError: 17
15:45 ogzy sed: -e expression #1, char 44: unterminated `s' command
15:46 ogzy MiroslavAnashkin: so what should i do?
15:50 ArminderS joined #fuel
15:51 MiroslavAnashkin ogzy: Please restore the network settings back first. Or simply redeploy the master node - it will ask you for custom network settings on first boot after OS installation.
15:53 ArminderS joined #fuel
16:02 ArminderS joined #fuel
16:06 e0ne joined #fuel
16:17 e0ne joined #fuel
16:22 rmoe joined #fuel
16:32 ogzy MiroslavAnashkin: bootstrap_admin_node.sh only configure eth0 and eth1
16:33 ogzy MiroslavAnashkin: i need another network also, one for pxe, one for public one for nodes, as i understood
16:44 xarses joined #fuel
16:47 albionandrew Morning xarses I updated the firmware, downloaded the latest SPP it did 4 updates but the important thing is it did the updates you suggested yesterday but I still get kernel panics - http://pastebin.com/egW6rWFn any idea what to try next?
16:52 angdraug joined #fuel
16:59 MiroslavAnashkin ogzy: Bootstrap_admin_node.sh configures only physical interfaces on master node.
17:17 vkozhukalov joined #fuel
17:18 xarses albionandrew: dhblaz: I received word of a possible work around that involves keeping the 2.6 kernel
17:19 xarses ovs-vsctl set interface eth2 other-config:enable-vlan-splinters=true
17:20 albionandrew That sounds like something I ight have to wait for dhblaz for? Can you talk me through it?
17:20 albionandrew 2.6 kernel has been deleted
17:22 albionandrew Although since the firmware update one of the blades, b5 has not kernel panicked
17:27 xarses albionandrew: you can find the packages from http://download.mirantis.com/fuelweb​-repo/3.2/centos/os/x86_64/Packages/
17:28 albionandrew So just put the [   ] kernel-2.6.32-358.118.1.openstack.el6.x86_64.rpm      and it's dependencies on the 2 blades we are having issues with?
17:29 albionandrew and update grub etc?
17:30 xarses the rpm installer should re-do grub for you
17:30 albionandrew OK Ido that now thanks
17:30 xarses if you need to put it into cobbler, they need to be copied to the fuel node and rebuild the yum repo
17:31 xarses /var/www/nailgun/centos/fuelweb/x86_64/Packages
17:31 albionandrew So I can't just download the rpms to those nodes and do it that way?
17:31 xarses you can
17:32 xarses if they don't lock up on you
17:32 albionandrew OK I will try downloading and applying them on one node, don't want to touch cobbler without talking to dhblaz, he'll be back in a while
17:35 book` joined #fuel
17:44 albionandrew xarses I can't get the new kernel on because the kernel panic, is there another way to do it? it even crashes in single user mode?
17:44 albionandrew the 2.6 kernel
17:45 xarses if you can hit the pxe menu, you could select the bootstrap image and use it to install the old kernel
17:46 albionandrew OK I'll try that thanks
17:49 e0ne joined #fuel
18:04 IlyaE joined #fuel
18:06 teran joined #fuel
19:18 ArminderS joined #fuel
19:18 jeremydei joined #fuel
19:18 teran joined #fuel
19:18 IlyaE joined #fuel
19:18 book` joined #fuel
19:18 vkozhukalov joined #fuel
19:18 angdraug joined #fuel
19:18 xarses joined #fuel
19:18 albionandrew joined #fuel
19:18 dhblaz joined #fuel
19:18 vnaboychenko joined #fuel
19:18 tonyha joined #fuel
19:18 iscsi joined #fuel
19:18 aabashkin joined #fuel
19:18 mattymo joined #fuel
19:18 mrasskazov joined #fuel
19:18 bogdando joined #fuel
19:18 korn__ joined #fuel
19:18 SteAle joined #fuel
19:18 akupko joined #fuel
19:18 ArminderS- joined #fuel
19:18 Arminder joined #fuel
19:18 jkyle joined #fuel
19:18 vk joined #fuel
19:18 MiroslavAnashkin joined #fuel
19:18 kcherchenko joined #fuel
19:18 sanek joined #fuel
19:18 Bomfunk joined #fuel
19:18 jkirnosova joined #fuel
19:18 ogelbukh joined #fuel
19:18 evgeniyl joined #fuel
19:18 xdeller joined #fuel
19:18 meow-nofer joined #fuel
19:18 AndreyGrebenniko joined #fuel
19:18 VladNaboychenko joined #fuel
19:20 xdeller probably
19:20 albionandrew So parses said to role back to 2.6 kernel still want me to try?
19:20 albionandrew xarses
19:20 albionandrew autocorrect sorry
19:20 xdeller what kernel you are using right now and what`s the problems except this panic?
19:20 xdeller xarses posted also nice panic from hp watchdog timer couple of days ago, not sure if it was yours
19:20 albionandrew I just deployed one of the problem nodes to 2.6, it was modified to use the 3.x in the previous KS
19:20 albionandrew watchdog was most liekly mine
19:20 ogzy joined #fuel
19:20 xdeller ok, so can you try with the option I mentioned? wdt can be turned off by modprobe.blacklist=hpwdt
19:21 xarses xdeller: the panic was from albionandrew  / dhblaz
19:21 xdeller fine
19:21 albionandrew we had  modprobe.blacklist=hpwdt
19:22 xdeller so we have dma issue on 2.6 rh kernel and wdt in the 3.10?
19:22 albionandrew let me try and find the paste bin
19:23 albionandrew Here was latest kernel args we tried - Command line: ro root=/dev/mapper/os-root rd_NO_LUKS  KEYBOARDTYPE=pc KEYTABLE=us LANG=en_US.UTF-8 rd_LVM_LV=os/root rd_NO_MD SYSFONT=latarcyrheb-sun16 crashkernel=auto  biosdevname=0 crashkernel=none rd_LVM_LV=os/swap rd_NO_DM console=tty0 console=ttyS0,9600n8 modprobe.blacklist=hpwdt nmi_watchdog=1
19:23 albionandrew Pastbin is here
19:23 albionandrew http://pastebin.com/egW6rWFn
19:23 xdeller why nmi wd was enabled explicitly/
19:23 xdeller ?
19:23 xdeller it is turned on by default i though
19:23 albionandrew that was with 3.x kernel and updated firmware
19:24 xdeller and what do you have here?
19:24 xdeller oh i get it
19:24 xarses xdeller: the the hpwdt module is supposed to not load its self if nmi_watchdog=1
19:24 xdeller ehm, you had blacklisted it anyway
19:24 xarses so I asked them to add it to the args
19:24 xarses it loaded anyway
19:25 xdeller so you have iommu dmar because of recent kernel
19:25 xdeller please disable it too
19:25 xdeller never worked for me :)
19:26 albionandrew I have the 2.6 kernel now the machine most of the testing was does on is sitting at the bootstrap please clarify what you need/want me to try.
19:26 albionandrew the other machine B5 remains with 3.x kern
19:27 xdeller please switch back to the 3.10 with disabled iommu and blacklisted hp wd timer
19:28 albionandrew ok will do that on b5
19:33 dhblaz joined #fuel
19:36 dhblaz xdeller, the hp watchdog driver was in the kernel panic but it was pretty clear that a NMI was triggered further down the stack.
19:37 teran1 joined #fuel
19:38 vkozhukalov_ joined #fuel
19:38 dhblaz xdeller: The problem is networking stops working (i.e. packets stop getting passed) after some amount of time.  I don't remember exactly in what way it stops working; I would have to look at it again.
19:39 teran2 joined #fuel
19:39 dhblaz xdeller: A /etc/init.d/network restart temporarily fixes the problem
19:40 vkozhukalov_ left #fuel
19:41 vkozhukalov joined #fuel
19:41 rmoe_ joined #fuel
19:45 rmoe joined #fuel
20:00 xarses joined #fuel
20:35 albionandrew xarses is there away once an environment has been created so make the master found a discovered node and then rediscover it fresh? I made a change to the flex 10 taking the NICS from 11 to 4 but even though I returned the node to the unallocated pool it still sees 11 NICS but if I login with the bootstrap it sees the correct/newly assigned 4.
20:40 xarses albionandrew: it should update dynamically as soon as the nailgun agent updates it's registration.
20:41 albionandrew OK Thanks.
20:41 xarses albionandrew: you can run /opt/nailgun/bin/agent manually from the bootstrap node to force the update and see if it has any errors
20:42 albionandrew Maybe I'll try that.
20:42 xdeller_ joined #fuel
20:42 xarses it shouldn't be necessary, but there could maybe be suck attributes, that if you remove it from the cluster and add it back might cause a refresh. If all that is fine, it might be a browser cache issue that a forced refresh might help with. Otherwise it's probably a bug
20:43 xdeller_ dhblaz - raising the question from history - can you check error counters before restart networking
20:44 xdeller_ i suspect one problem with ip-link
20:47 albionandrew xarses deleted the node and returned to unallocated pool, ran nail gun, still shows 11 NICS on gui and ifconfig -a on bootstrap shows the 4
20:48 xarses and a hard refresh of the browser didn't clear it either?
20:49 albionandrew xarses and I also tried an incognito window in chrome
20:51 dhblaz joined #fuel
20:52 xarses ok
20:52 xarses you'd want to remove the node from the cluster if it's in one, and then drop it from the db
21:30 dhblaz Found an interesting bug
21:30 dhblaz Click on Verify Networks
21:31 dhblaz Returns Green not sure of the All Clear message but you get the point
21:31 dhblaz Click Verify Networks again
21:32 dhblaz Click on the Nodes tab
21:32 dhblaz Come back and see "Red" notices about DHCP server
21:32 dhblaz Tried to repeat but got green again
21:33 dhblaz Some of the dhcp server messages I don't understand
21:33 dhblaz but the ones on the "Public" network I do.  That is my DMZ and I definately run a DHCP server there
21:34 dhblaz Not sure why it would not show up every time.
21:34 dhblaz I'm going to try clicking Verify several times to see what happens
21:35 dhblaz I was able to repeat the issue
21:35 dhblaz Without changing tabs
21:35 dhblaz Green: Verification succeeded. Your network is configured correctly.
21:35 dhblaz Red: Verification failed.
21:35 dhblaz Error occurred while running method 'verify_networks'. Inspect Orchestrator logs for the details.
21:36 dhblaz Red: DHCP Messages
21:38 dhblaz Green: Verification succeeded. Your network is configured correctly.
21:39 dhblaz Green: Verification succeeded. Your network is configured correctly.
21:39 dhblaz Green: Verification succeeded. Your network is configured correctly.
21:44 dhblaz I'm guessing that the agents don't wait long enough to receive the dhcp response
21:55 xdeller_ nice, thanks for the report
21:56 xdeller_ how`s about your networking? can you check the counters as I suggested?
21:56 xdeller_ and what you finally get with 3.10 w/ disabled features?

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