I am trying to create virtual hosts using Libvirt compute resource through
Hammer. I am running Foreman 1.10 and hammer (0.6.1) with
hammer_cli_foreman (0.6.2).
There are multiple issues that I came across and not sure what the problem
is.
Success:
All bare metal machines provisions fine with provided IP address
Problem
Hosts continue to receive provisioning template upon reboot. Hammer
command used to provision machine is
hammer --output json host create --hostgroup=Virtual --name=collab
–interface=primary=true,managed=true,provision=true,compute_bridge=br0,compute_model=virtio,mac=00:16:3e:01:fd:42
–compute-resource=kvmA
–compute-attributes=start=1,cpus=6,memory=6291456000
–volume=format_type=qcow2,capacity=200G --puppet-classes=role::collab
–parameters=sensortype=collab
I had to specify MAC address for it to succeed. If I don't provide MAC it
returns an error with return value 70 and message
"Could not create the host:
Invalid MAC
"
Here is the snippet from foreman-tail output
==> /var/log/foreman/production.log <==
2016-04-19 07:22:59 [app] [I] Started POST "/api/hosts" for 192.168.130.227
at 2016-04-19 07:22:59 +0000
2016-04-19 07:22:59 [app] [I] Processing by Api::V2::HostsController#create
as JSON
2016-04-19 07:22:59 [app] [I] Parameters: {"host"=>{"name"=>"splunkSH",
"compute_resource_id"=>3, "hostgroup_id"=>1, "build"=>true,
"enabled"=>true, "managed"=>true, "compute_attributes"=>{"start"=>"1",
"cpus"=>"6", "memory"=>"8388608000",
"volumes_attributes"=>{"0"=>{"format_type"=>"qcow2", "capacity"=>"500G"}}},
"puppetclass_ids"=>[16], "overwrite"=>true,
"host_parameters_attributes"=>[{"name"=>"sensortype",
"value"=>"[FILTERED]"}], "interfaces_attributes"=>[{"primary"=>"true",
"managed"=>"true", "provision"=>"true", "ip"=>"192.168.130.34",
"compute_attributes"=>{"bridge"=>"br0", "model"=>"virtio"}}]}, "apiv"=>"v2"}
2016-04-19 07:22:59 [app] [I] Authorized user sensoradmin(Admin User)
==> /var/log/foreman-proxy/proxy.log <==
D, [2016-04-19T07:22:59.554258 #9597] DEBUG – : accept:
192.168.130.254:53261
D, [2016-04-19T07:22:59.557761 #9597] DEBUG – : Rack::Handler::WEBrick is
invoked.
D, [2016-04-19T07:22:59.558660 #9597] DEBUG – : verifying remote client
192.168.130.254 against trusted_hosts foreman.auto
I, [2016-04-19T07:22:59.559210 #9597] INFO – : 192.168.130.254 - -
[19/Apr/2016 07:22:59] "GET /serverName HTTP/1.1" 200 17 0.0009
D, [2016-04-19T07:22:59.600121 #9597] DEBUG – : close:
192.168.130.254:53261
==> /var/log/foreman/production.log <==
2016-04-19 07:22:59 [app] [W] Action failed
> Net::Validations::Error: Invalid MAC
> /usr/share/foreman/lib/net/validations.rb:40:in validate_mac' > /usr/share/foreman/lib/net/dhcp/record.rb:7:ininitialize'
> /usr/share/foreman/app/models/concerns/orchestration/dhcp.rb:20:in new' > /usr/share/foreman/app/models/concerns/orchestration/dhcp.rb:20:indhcp_record'
> /usr/share/foreman/app/models/concerns/orchestration/dhcp.rb:158:in
queue_remove_dhcp_conflicts' > /usr/share/foreman/app/models/concerns/orchestration/dhcp.rb:111:inqueue_dhcp'
Once it is provisioned, I see it informing Foreman via
==> /var/log/httpd/puppet_access_ssl.log <==
192.168.130.101 - - [19/Apr/2016:07:42:36 +0000] "POST
/production/catalog/splunksh.auto HTTP/1.1" 200 158772 "-" "-"
==> /var/log/foreman/production.log <==
2016-04-19 07:42:50 [app] [I] Started POST "/api/reports" for 127.0.0.1 at
2016-04-19 07:42:50 +0000
2016-04-19 07:42:50 [app] [I] Processing by
Api::V2::ReportsController#create as JSON
2016-04-19 07:42:50 [app] [I] Parameters: {"report"=>"[FILTERED]",
"apiv"=>"v2"}
2016-04-19 07:42:50 [app] [I] processing report for splunksh.auto
2016-04-19 07:42:54 [app] [I] Imported report for splunksh.auto in 3.99
seconds
2016-04-19 07:42:55 [app] [I] Rendered api/v2/reports/create.json.rabl
(980.8ms)
2016-04-19 07:42:55 [app] [I] Completed 201 Created in 5025ms (Views:
852.3ms | ActiveRecord: 2542.7ms)
==> /var/log/httpd/foreman-ssl_access_ssl.log <==
127.0.0.1 - - [19/Apr/2016:07:42:50 +0000] "POST /api/reports HTTP/1.1" 201
84571 "-" "-"
==> /var/log/httpd/puppet_access_ssl.log <==
192.168.130.101 - - [19/Apr/2016:07:42:49 +0000] "PUT
/production/report/splunksh.auto HTTP/1.1" 200 11 "-" "-"
==> /var/log/foreman/production.log <==
2016-04-19 07:42:56 [app] [I] Started GET
"/unattended/built?token=4dc3fe6b-54b6-49ce-a2d6-98e61fb55a64" for
192.168.130.101 at 2016-04-19 07:42:56 +0000
2016-04-19 07:42:56 [app] [I] Processing by UnattendedController#built as
/
2016-04-19 07:42:56 [app] [I] Parameters:
{"token"=>"4dc3fe6b-54b6-49ce-a2d6-98e61fb55a64"}
2016-04-19 07:42:56 [app] [I] Found splunksh.auto
2016-04-19 07:42:56 [app] [I] unattended: splunksh.auto is Built!
2016-04-19 07:42:56 [app] [I] Completed 409 Conflict in 517ms
(ActiveRecord: 19.8ms
==> /var/log/httpd/foreman_access.log <==
192.168.130.101 - - [19/Apr/2016:07:42:56 +0000] "GET
/unattended/built?token=4dc3fe6b-54b6-49ce-a2d6-98e61fb55a64 HTTP/1.0" 409
1 "-" "Wget/1.12 (linux-gnu)"
When machine reboots it receives same PXE Provisioning script and goes in
infinite loop of reboot/install/reboot.
Hammer_foreman_CLI documentation is somewhat out of date with respect to
current version. Is there a plan to update it soon?
Thanks,