This step adds the mac hosts running Anka Agent package to the Anka Build cloud Controller.
sudo ankacluster join <http://controllerIP:port>
When joining the nodes using
ankacluster join, use sudo. Anka is multi user environment like unix. When you join with sudo, controller will start VMs with sudo. You can view them with
sudo anka list and operate with these VMs using sudo with other anka commands.
You can also set other parameters like name(the name that shows up in controller dashboard for this host), max-vm-count (number of concurrent VMs to run) etc.
All of the following flags are optional.
|–max-vm-count||-m||number||Maximum number of VMs this node is allowed to run (default 2)|
|–host||-H||string||Specify host name or IP for this machine (Default: resolve IP automatically)|
|–name||-n||NAME||Node name alias|
|–reserve-space||-||Size||Disk space to reserve when pulling. Number followed by magnitude (1024B, 10KB, 140MB, 45GB…). When |
|–heartbeat||-||duration (5s is 5 seconds)||Override the duration between updates|
|–no-central-logging||-||no value||Don't send logs to central logging system|
|–groups||-G||string||Specify group name (or names sepearated by ‘,') to add this node to|
|–capacity-mode||-M||string||Capacity mode (resource|
|–vcpu-override||-V||int||Override the max vcpu that this node can handle (when using capacity-mode resource). defaults to (cpu count * 2)|
|–tls||-t||no value||Use tls for communicating with the queue (optional)|
|–ram-override||-R||int||Override the max RAM in GB that this node can handle (when using capacity-mode resource). defaults to (total ram - 2GB)|
|–cacert||-c||path||to CA bundle file (PEM/X509)|
|–root-cert||-r||path||Path to CA bundle file (PEM/X509)|
|–cert||-C||path||client certificate file (PEM/X509)|
|–cert-key||-K||path||client certificate key file (PEM/X509)|
|–keystore||-k||path||Path to certificate and keystore (PEM, PKCS12)|
|–keystore-pass||-p||string||password for certificate and keystore|
|–skip-tls-verification||-||no value||Skip TLS verification|
|–skip-tests||-||no value||Skip the connection test before starting the agent|
|–force-no-sudo||-f||no value||Force the agent to start without sudo|
|–global||-g||no value||DEPRECATED! Install agent into system domain|
Executing the command with the
--help flag will show the usage:
ankacluster join --help Joins the current machine to Anka Cloud cluster Usage: ankacluster join [controller_address] [flags] Flags: -c, --cacert PATH PATH to CA bundle file (PEM/X509) (optional) -M, --capacity-mode string Capacity mode (resource|number). 'resource' will take vm jobs according to resources, number will take vm jobs according to max-vm-count (default "number") -C, --cert PATH PATH client certificate file (PEM/X509) (optional) -K, --cert-key PATH PATH client certificate key file (PEM/X509) (optional) -f, --force-no-sudo Force the agent to start without sudo -g, --global DEPRECATED! Install agent into system domain (optional) -G, --groups string Specify group name (or names sepearated by ',') to add this node to (optional) --heartbeat duration Override the duration between updates -h, --help help for join -H, --host string Specify host name or IP for this machine (optional) -k, --keystore PATH PATH to certificate and keystore (PEM, PKCS12) (optional) -p, --keystore-pass PASSWORD PASSWORD for certificate and keystore (optional) -m, --max-vm-count NUMBER Maximum NUMBER of VMs this node is allowed to run (optional) (default 2) -n, --name NAME Node NAME alias (optional) --no-central-logging Dont send logs to central logging (optional) -R, --ram-override int Override the max RAM in GB that this node can handle. defaults to (total ram - 2GB) --reserve-space string Disk space to reserve when pulling. Number followed by magnitude (1024B, 10KB, 140MB, 45GB...) -r, --root-cert PATH PATH to CA bundle file (PEM/X509) (optional) --skip-tests Skip testing the connection before starting the agent (optional) --skip-tls-verification Skip TLS verification (optional) -t, --tls Use tls for communicating with the queue (optional) -V, --vcpu-override int Override the max vcpu that this node can handle. defaults to (cpu count * 2)
Starting from version 1.2.0,
ankacluster command will now ask to be run with sudo. So
ankacluster join http://<host> will return available flags and a error message. It’s possible to override this behavior with the
-f or the
New in version 1.2.0 - Control the concurrent VM count execution on a node using a static number of dynamic capacity available. For dynamic capacity, use the
-M, --capacity-mode string in conjuction with
-m, --max-vm-count NUMBER and
-R, --ram-override int flags. Default is based on static Number and is 2.
In version 1.2.0,
-g flag is deprecated but backward compatibility is supported.
In version 1.3.0, controller will auto shutdown any running unmanaged VMs under sudo anka list. For any manual maintenance of VMs under sudo, disjoin the node first. This was removed in version 1.4.0 onward.
In version 1.3.0, the anka agent running on the nodes sends logs to the central controller. This is useful for debugging purposes. This can be disabled with –no-central-logging during ankacluster join command.
This step will remove the node from the Anka Build cloud. Execute this when you want to do maintainence on the node etc. You don't need to disjoin nodes to upgrade Anka on the node.
sudo ankacluster disjoin
In version 1.2.0,
-gflag is deprecated for the
anka stop -f vmnamecommand.
For major Anka releases, it maybe required to upgrade guest addons in existing Anka VMs.
Check the release notes to identify if this step is required or not.
anka start -u vmnameto upgrade guest addons in existing VMs to the current release. If you are not able to upgrade the guest add-ons tool using the
anka start -u vmnamecommand, then you have a very old version of guest addon tools on your VM. You will first need to manually update them. Contact Veertu support.
anka stop -f vmname anka start -u vmname Preparing update configuration Installing updates Suspending update succeeded
Was this page helpful?
Glad to hear it! Please tell us how we can improve.
Sorry to hear that. Please tell us how we can improve.