Load Balancer Statuses

Provisioning statuses

The provisioning status is a status that indicates the lifecycle stage the load balancer is currently in. It shows the state and success of creating, updating, or deleting the load balancer.

Possible provisioning statuses:

Name in the management console	Technical name	Description
Active	`ACTIVE`	The load balancer has been successfully created and is ready for operation; the configuration has been applied
Error	`ERROR`	An error occurred while creating or changing the load balancer settings. Its exact state is unknown. Contact technical support
Creating	`PENDING_CREATE`	The load balancer is being created
Updating	`PENDING_UPDATE`	The load balancer is being updated
Deleting	`PENDING_DELETE`	The load balancer is being deleted

Name in the management console

Technical name

Description

Active

ACTIVE

The load balancer has been successfully created and is ready for operation; the configuration has been applied

Error

ERROR

An error occurred while creating or changing the load balancer settings. Its exact state is unknown. Contact technical support

Creating

PENDING_CREATE

The load balancer is being created

Updating

PENDING_UPDATE

The load balancer is being updated

Deleting

PENDING_DELETE

The load balancer is being deleted

Operating statuses

Operating statuses are of two types:

Common — displayed at the top level, for example, in the management console.
Special — displayed only at the pool and backend level; can be viewed using the OpenStack CLI.

Common operating statuses

The load balancer’s common status may change depending on the combination of special operating statuses specific to an individual backend. A backend is one or more servers that accept requests and are located behind the load balancer.

Possible common operating statuses:

Name	Description
`ONLINE`	The status is assigned if at least one of the following conditions is met: The load balancer is operating normally (all system processes of the load balancer are functioning correctly and accepting connections). This condition is checked even if no backends have been added to the load balancing pool. All pool members (backends, load balancing rules) are healthy and respond to the health check
`DEGRADED`	The status is assigned if at least one of the following conditions is met: One or more pool members (backend) are in the `ERROR` state (do not respond to the health check). One of the Amphoras in the `Active`-`Standby` high-availability cluster stopped responding to the health check; the service continues to handle requests on the node operating in `Active` mode
`ERROR`	The status is assigned if at least one of the following conditions is met: All pool members (backends) do not respond to the health check. All pool members have a weight coefficient of 0 (the administrator enabled `DRAINING` on all backends). An infrastructure failure occurred, as a result of which both nodes of the Amphora cluster do not respond to health checks. All pool members (backends) are turned off (the administrator switched all backends to the `OFFLINE` state). An infrastructure failure occurred because the HAProxy daemon failed to apply the configuration

Name

Description

ONLINE

The status is assigned if at least one of the following conditions is met:

The load balancer is operating normally (all system processes of the load balancer are functioning correctly and accepting connections). This condition is checked even if no backends have been added to the load balancing pool.
All pool members (backends, load balancing rules) are healthy and respond to the health check

DEGRADED

The status is assigned if at least one of the following conditions is met:

One or more pool members (backend) are in the ERROR state (do not respond to the health check).
One of the Amphoras in the Active-Standby high-availability cluster stopped responding to the health check; the service continues to handle requests on the node operating in Active mode

ERROR

The status is assigned if at least one of the following conditions is met:

All pool members (backends) do not respond to the health check.
All pool members have a weight coefficient of 0 (the administrator enabled DRAINING on all backends).
An infrastructure failure occurred, as a result of which both nodes of the Amphora cluster do not respond to health checks.
All pool members (backends) are turned off (the administrator switched all backends to the OFFLINE state).
An infrastructure failure occurred because the HAProxy daemon failed to apply the configuration

Special operating statuses

Special statuses reflect the state of the load balancing pool or its members. A pool or its member can also take one of the common statuses; in this case, such a status participates in forming the common status displayed at the top level. For details, see the status mapping matrix.

Possible special operating statuses:

Name	Description
`DRAINING`	The administrator has determined that one or more pool members do not accept new connections
`OFFLINE`	The administrator switched the pool or backend to an inactive state
`NO_MONITOR`	No mechanism for checking backend health is configured for the pool

Name

Description

DRAINING

The administrator has determined that one or more pool members do not accept new connections

OFFLINE

The administrator switched the pool or backend to an inactive state

NO_MONITOR

No mechanism for checking backend health is configured for the pool

Operating status mapping matrix

Below is an example of how the load balancer’s common operating statuses depend on the operating statuses of members of a load balancing pool consisting of two backends.

In this example, a backend in the ERROR state indicates a failure at the web application level when the HTTP health check fails. A backend in the ONLINE state indicates a normally operating backend that responds to the health check.

Backend-1 status	Backend-2 status	Common status
`ONLINE`	`ONLINE`	`ONLINE`
`ONLINE`	`ERROR`	`DEGRADED`
`ERROR`	`ERROR`	`ERROR`
`ONLINE`	`DRAINING`	`ONLINE`
`DRAINING`	`DRAINING`	`ERROR`
`ONLINE`	`OFFLINE`	`ONLINE`
`OFFLINE`	`OFFLINE`	`ERROR`

Was this article helpful?

Table of Contents
Provisioning statuses
Operating statuses
Was this article helpful?