h1. Openstack Management TTNN

h2. Opération sur les VMs

h3. Reboot VM en error

<pre>

nova reset-state --active xxx.tetaneutral.net

nova reboot --hard --poll xxx.tetaneutral.net

</pre>

h3. Agrandissement d'un disque

Sur la machine 'openstack':

<pre>

root@openstack:~/tools ./os-grow-disk.sh 

usage: os-grow-disk.sh CINDER_VOLUME_ID SIZE_GB

root@openstack:~/tools# ./os-grow-disk.sh bc6f01cb-c6b6-4413-b2ef-1f7df0031049 60

Cinder:

bc6f01cb-c6b6-4413-b2ef-1f7df0031049 , test4-bootdisk | 40 -> 60

Enter YES to update

YES

Testing g3 ssh connection...

Resizing the kvm blockdevice...

Block device 'vda' is resized

Updating cinder database with the new value...

Update finish

</pre>

Puis sur la VM:

Il faut avoir un 'parted' récent ou patché avec le support du 'online resizepart':

'online resizepart' est natif depuis parted >= 3.1: 

* debian >= jessie

* ubuntu >= 14.10

'online resizepart' backporté dans:

* ubuntu >= 13.10 < 14.10 

Pour debian wheezy, utilisation les packages ubuntu de 14.04 devrait marcher 

(pas encore testé, à voir si le kernel veut bien reload la table des partitions après)

<pre> 

root@test4:~# parted /dev/vda print | grep vda

Disk /dev/vda: 64.4GB

root@test4# parted /dev/vda resizepart 1 

Warning: Partition /dev/vda1 is being used. Are you sure you want to continue?

Yes/No? Yes                                                               

End?  [42.9GB]? 64.4GB   # METTRE ICI LA TAILLE VU AU DESSUS                                                 

Information: You may need to update /etc/fstab.                           

root@test4:~# resize2fs /dev/vda1

resize2fs 1.42.9 (4-Feb-2014)

Filesystem at /dev/vda1 is mounted on /; on-line resizing required

old_desc_blocks = 3, new_desc_blocks = 4

The filesystem on /dev/vda1 is now 15722400 blocks long.

root@test4:~# df -h | grep vda

/dev/vda1        60G  797M   56G   2% /

</pre>

h3. Ajout d'une interface réseau à une VM:

<pre>

$ neutron net-list | grep vlan3199

| 460f49ce-f5ed-449e-9e0f-7626bcee4379 | vlan3199         | ba3f12af-074a-4a31-9a7c-a621993e13bb 192.168.99.0/24     |

$ neutron port-create 460f49ce-f5ed-449e-9e0f-7626bcee4379 --name nagios-eth2 --fixed-ips type=dict list=true ip_address=192.168.99.248

Created a new port:

+-----------------------+---------------------------------------------------------------------------------------+

| Field                 | Value                                                                                 |

+-----------------------+---------------------------------------------------------------------------------------+

| admin_state_up        | True                                                                                  |

| allowed_address_pairs |                                                                                       |

| binding:host_id       |                                                                                       |

| binding:profile       | {}                                                                                    |

| binding:vif_details   | {}                                                                                    |

| binding:vif_type      | unbound                                                                               |

| binding:vnic_type     | normal                                                                                |

| device_id             |                                                                                       |

| device_owner          |                                                                                       |

| fixed_ips             | {"subnet_id": "ba3f12af-074a-4a31-9a7c-a621993e13bb", "ip_address": "192.168.99.248"} |

| id                    | 918e0ec7-0909-423c-9a67-15b5ef7c7ae0                                                  |

| mac_address           | fa:16:3e:6b:32:07                                                                     |

| name                  | nagios-eth2                                                                           |

| network_id            | 460f49ce-f5ed-449e-9e0f-7626bcee4379                                                  |

| security_groups       | b6d5b5fb-a053-4f97-b964-9e1a12d243b1                                                  |

| status                | DOWN                                                                                  |

| tenant_id             | ed980105f9d047e2bee738b3f261f126                                                      |

+-----------------------+---------------------------------------------------------------------------------------+

$ nova interface-attach --port-id 918e0ec7-0909-423c-9a67-15b5ef7c7ae0   nagios.tetaneutral.net

</pre>

Dans la VM:

<pre>

$ dmesg | tail 

[ 1315.826088] virtio-pci 0000:00:06.0: using default PCI settings

[ 1315.826102] virtio-pci 0000:00:07.0: no hotplug settings from platform

[ 1315.826102] virtio-pci 0000:00:07.0: using default PCI settings

[ 1315.826116] pci 0000:00:08.0: no hotplug settings from platform

[ 1315.826117] pci 0000:00:08.0: using default PCI settings

[ 1315.826382] virtio-pci 0000:00:08.0: enabling device (0000 -> 0003)

[ 1315.829281] virtio-pci 0000:00:08.0: setting latency timer to 64

[ 1315.829535] virtio-pci 0000:00:08.0: irq 50 for MSI/MSI-X

[ 1315.829555] virtio-pci 0000:00:08.0: irq 51 for MSI/MSI-X

[ 1315.829563] virtio-pci 0000:00:08.0: irq 52 for MSI/MSI-X

$ ip l   show dev eth2

4: eth2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT qlen 1000

    link/ether fa:16:3e:6b:32:07 brd ff:ff:ff:ff:ff:ff

$ tail -4 /etc/network/interfaces

auto eth2

iface eth2 inet static

   address 192.168.99.248

   netmask 255.255.255.0

$ ifup eth2

$ ping 192.169.99.250

PING 192.169.99.250 (192.169.99.250) 56(84) bytes of data.

64 bytes from 192.169.99.250: icmp_req=1 ttl=119 time=170 ms

64 bytes from 192.169.99.250: icmp_req=2 ttl=119 time=204 ms

^C

</pre>

</pre>

h2. VM de management (nommé openstack)

h3. Live Migration d'un host à l'autre

Cela fait tout à notre place niveau kvm/libvirt/ceph

<pre>

virsh migrate --persistent --undefinesource --live --p2p openstack qemu+ssh://g2:2222/system

</pre>

h3. Recover depuis une perte de la machine hote:

Créé un fichier openstack.xml à partir d'un backup de '/etc/libvirt/qemu/openstack.xml' ou utiliser l'exemple ici: attachment:openstack.xml

_Note: il faut surement viré les uuid des interfaces openvswitch qui sont différent suivant le hosts_

Puis on recréé la VM

<pre>

$ virsh create openstack.xml

$ virsh start openstack.xml

</pre>