Vmpooler configuration can be provided via file or environment variable. Some configuration elements are unique to API or manager components. This document aims to describe options for providing vmpooler configuration via environment variables.
These options affect manager and API applications.
Provide the entire configuration as a blob of yaml. Individual parameters passed via environment variable will override any setting in this blob, or a configuration file passed in.
Path to a the file to use when loading the vmpooler configuration. This is only evaluated if VMPOOLER_CONFIG has not been specified.
If set, returns a top-level 'domain' JSON key in POST requests (optional)
The redis server to use for vmpooler. (optional; default: 'localhost')
The port to use when connecting to redis. (optional)
The password to use when connecting to redis. (optional)
How long to keep data when set to expire in hours. (optional; default: 168)
The FQDN hostname of the statsd daemon. (optional)
The prefix to use while storing statsd data. (optional; default: 'vmpooler')
The UDP port to communicate with the statsd daemon. (optional; default: 8125)
The FQDN hostname of the Graphite server. (optional)
The prefix to use while storing Graphite data. (optional; default: 'vmpooler')
The TCP port to communicate with the graphite server. (optional; default: 2003)
The number of concurrent VM creation tasks to perform. Only affects manager. (optional; default: 10)
When set to any value greater than 0 enable VM migration at checkout. When enabled this capability will evaluate a VM for migration to a different host when it is requested in an effort to maintain a more even distribution of load across compute resources. The number of migrations in progress at any one time are constrained by this value. (optional)
How often (in minutes) to check the sanity of VMs in 'ready' queues. (optional; default: 15)
How long (in hours) to keep VMs in 'running' queues before destroying. (optional; default: 24)
Same as vm_lifetime, but applied if a valid authentication token is
included during the request.
(required)
If set, prefixes all created VMs with this string. This should include a separator. (optional; default: '')
The file to use for logging manager operations. (optional; default: '/var/log/vmpooler.log')
The target cluster VMs are cloned into (host with least VMs chosen) (optional; default: same cluster/host as origin template)
How long (in minutes) before marking a clone as 'failed' and retrying. (optional; default: 15)
Set the max number of times a connection should retry in VM providers. This optional setting allows a user to dial in retry limits to suit your environment. (optional; default: 3)
When retrying, each attempt sleeps for the try count * retry_factor. Increase this number to lengthen the delay between retry attempts. This is particularly useful for instances with a large number of pools to prevent a thundering herd when retrying connections. (optional; default: 10)
Create the pool folder specified in the pool configuration. Note: this will only create the last folder when it does not exist. It will not create any parent folders. (optional; default: false)
Create backing delta disks for the specified templates to support creating linked clones. (optional; default: false)
Whether to create linked clone virtual machines when using the vsphere provider. This can also be set per pool. (optional; default: true)
Enable purging of VMs and folders detected within the base folder path that are not configured for the provider Only a single layer of folders and their child VMs are evaluated from detected base folder paths A base folder path for 'vmpooler/redhat-7' would be 'vmpooler' When enabled in the global configuration then purging is enabled for all providers Expects a boolean value (optional; default: false)
Enable shipping of VM usage stats When enabled a metric is emitted when a machine is destroyed. Tags are inspected and used to organize shipped metrics if there is a jenkins_build_url tag set for the VM. Without the jenkins_build_url tag set the metric will be sent as "usage.$user.$pool_name". When the jenkins_build_url tag is set the metric will be sent with additional data. Here is an example based off of the following URL; https://jenkins.example.com/job/platform\_puppet-agent-extra\_puppet-agent-integration-suite\_pr/RMM\_COMPONENT\_TO\_TEST\_NAME=puppet,SLAVE\_LABEL=beaker,TEST\_TARGET=redhat7-64a/824/ "usage.$user.$instance.$value_stream.$branch.$project.$job_name.$component_to_test.$pool_name", which translates to "usage.$user.jenkins_example_com.platform.pr.puppet-agent-extra.puppet-agent-integration-suite.puppet.$pool_name" Expects a boolean value (optional; default: false)
The provider to use for authentication. (optional)
The FQDN hostname of the LDAP server. (optional)
The port used to connect to the LDAP service. (optional; default: 389)
The base DN used for LDAP searches. This can be a string providing a single DN. For multiple DNs please specify the DNs as an array in a configuration file. (optional)
The LDAP object-type used to designate a user object. (optional)
The name of your deployment. (optional; default: 'vmpooler')
Enable experimental API capabilities such as changing pool template and size without application restart Expects a boolean value (optional; default: false)