[Crowbar] Admin node goes in a No Data (off) state

Gangur, Hrushikesh hrushikesh.gangur at hp.com
Thu Jul 26 15:17:48 CDT 2012


The admin node shows state as "No Data (Off).  This issue reproduce twice and am not sure how to go ahead. Is there a way to manually bring crowbar components online. Here are the setup details and steps followed:



Environment Details:

8/16 blades are used from HP-C7000

These 8 blades have eth0 connected to an isolated network. Eth1 to public (lab) network

Crowbar installation to use default network configuration (i.e. flat DHCP on eth0).



Steps followed:

Download the latest iso image available at: http://crowbar.zehicle.com/openstack-fred3.iso

Using ILO attach the iso image as virtual CD/DVD drive

System boots using the CD drive

Ignore the following error:

[Description: Machine generated alternative text: orne f irrnware floppy. missing ardware needs non-free firmware files to e loaded from removable media, such as a . irmware <Tab> moves; <Space> se1ects' <Enter> a'=ti.t buttons]


Had the timezone set correctly

Had all the systems down on 192.168.x network down (to avoid any previous conflict)

Disabled all ethx except eth0

Modified ipmi template file to disable ipmi crowbar

Had the windows client machine configured with static ip in 192.168.x network, but out from the DHCP scope of the admin node.

Installed crowbar.

Installation was successful. Kept the system for 4 hrs to see if something weird behavior comes up. No issues observed.

Brought one of the nodes in the same network up and had to select PXE boot. W/o this selection, it was going in to the previous OS boot option.

The node was detected in the crowbar dashboard as discovered.

Created a new group OpenStack for the controller node.

The node rebooted, and started installing ubuntu OS. The prompt "Detect and Mount CD Rom" as per Fig 1 showed up. Had to manually skip it.

After OS installation was done, the node showed up as  Ready. See the screenshot:


[Description: Machine generated alternative text:]

Note: Node2 is under works. Node 1 is ready for taking any of the OpenStack roles, specifically assigning controller role.


While the other nodes are joining crowbar deployment, I started assigning mysql and keystone role to node1:

[Description: Machine generated alternative text:]



Ughh! After I brought node 4 up (rosesx0068) the admin node state went into (no data) off. This was the node which I accidentally booted in the previous OS (failed to select the pxe boot in the boot order). I hard booted the node, and got it PXE Booted. The node was discovered by the admin node (though it was in Off state). Guess this might be the cause.



[Description: Machine generated alternative text: Node Dashboard $Y?Group Add Gmup node4 (Edit) Full Name d3c-4a-92-78-Ob-76.quasar.local Description Not set -J State Hardware Installed Hardware ProLiant BL49Oc G7 Uptime Unavailable CPU Intel(R) Xeon(R) CPU X5650 ® 2.67GHz Switch NamelPort 64-J1-50-fO-51-dS / Unknown Memory 70.92 GB MAC Address 3c:4a:92:78:Ob:76 Disk Drives 9, RAID: Not Set Allocated Allocated Service Tag U5E0494T35 IP Address admin: Unknown: 192.168,124.84 hmr: bmc: 192.168.124,166 mot m anagedj: eth7,eth6,eth5,eth4,eth3,eth2,ethl,ethü Links IP Mgmtlnterface, Ganglia, Nagios, Chef Barclamps Deployer Default Ipmi Default Roles bmc-nat-client, crowbar-h3c-4a-92-78-Ob-76_quasar_local, deployer-client, ipmi-discover Delete Reset Reinstall Hardware Update O compute (_) osgrp * sw-&4-31-SO-fS-S1-dS ! E! 7 >node1 O admin Ls ____ __________]



Turning off the node 4 for now. And, will reboot admin node. Before rebooting checked the interfaces (ifconfig), looks OK to me (only eth0 had assigned IP, other ethx were up and running w/o any IP Assigned). Nodes can PXE boot, DHCP server is working fine.



The reboot did not work.



Irrespective of the admin node state, I kept adding/applying O~S components, to node 1. Interestingly, they were all successful except the nova.

[Description: Machine generated alternative text: POWERED DV Open Stac Create and apply proposals in order from top to bottom. Allow each proposal to complete to ready before applying the next one. openstack Ct3uD c:trwARr Name Status Description ! b Mysql * Configures a mysql server ___________________ b Keystone Centralized authentication and authorization service for openstack b Swift Part of openstack, and provides a distributed blob storage b Glance Glance service (image registry and delivery service) forthe cloud b Nova Dashboard User interface for openstack projects (aka code name horizon) b Nova Installs and configures the openstack nova component. it relies upon the network and glance barclamps for normal operation.]



During this time, node 3 also went into State: No data (off) and came online. Now, it is waiting on NOVA component to get installed on Node 1 (the status has not changed yet from 'pending')



[Description: Machine generated alternative text: Node Dashboard jtaGroup 1g Add Group nodel (Edit) Full Name d68-b5-99-74-30-7a.quasar.local Description Not set State Ready Hardware ProLiant BL49Oc G7 Uptime 12 hours 15 minutes 11 seconds CPU Intel(R) Xeon(R) Cpu X5650 @' 2.67GHz Switch NamelPort 64-31-50-f8-51-d5 /unknown Memory 70.78 GB MAC Address 68:b5:99:74:30:7a Disk Drives 9, RAID: RAID 10 Allocated Allocated Service Tag USE0494T35 IP Address admin: ethü: 192.168.124.81 mot m anagedi: eth 7, eth 6, eth 5, eth 4, eth 3, eth 2, ethi bmc: bmc: 192.168.124.163 Links IP Mgmt Interface, Nagios, Ganglia, Nova Dashboard, Chef Barclamps Deployer Default Dns Default Ganglia Default Glance Proposal, Ipmi Default Keystone Proposal, Logging Default Mysql Proposal, Nagios Default Network Default Nova Dashboard Proposal, Ntp Default Provisioner Default Roles bmc-nat-client crowbar-d68-b5-99-74-30-7a_quasar_local, deployer-client, dns-client, ganglia-client glance-server, keystone-server, logging- client mysql-server, nagios-client, network, nova_dashboard-server, ntp-client provisioner-base Delete Reset Reinstall Hardware Update * compute (_) osgrp _________ * sw-64-31-S#fS-S1-d5 'i node2 nodel ! () admin node3 node4 ________ ________________ I I]

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.us.dell.com/pipermail/crowbar/attachments/20120726/42b27491/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 19567 bytes
Desc: image001.jpg
Url : http://lists.us.dell.com/pipermail/crowbar/attachments/20120726/42b27491/attachment-0006.jpg 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.jpg
Type: image/jpeg
Size: 21496 bytes
Desc: image002.jpg
Url : http://lists.us.dell.com/pipermail/crowbar/attachments/20120726/42b27491/attachment-0007.jpg 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image003.jpg
Type: image/jpeg
Size: 19277 bytes
Desc: image003.jpg
Url : http://lists.us.dell.com/pipermail/crowbar/attachments/20120726/42b27491/attachment-0008.jpg 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image004.jpg
Type: image/jpeg
Size: 19134 bytes
Desc: image004.jpg
Url : http://lists.us.dell.com/pipermail/crowbar/attachments/20120726/42b27491/attachment-0009.jpg 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image005.jpg
Type: image/jpeg
Size: 23480 bytes
Desc: image005.jpg
Url : http://lists.us.dell.com/pipermail/crowbar/attachments/20120726/42b27491/attachment-0010.jpg 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image006.jpg
Type: image/jpeg
Size: 35591 bytes
Desc: image006.jpg
Url : http://lists.us.dell.com/pipermail/crowbar/attachments/20120726/42b27491/attachment-0011.jpg 


More information about the Crowbar mailing list