Add the new metric introduced in #47731
to the reference page. Also performed some minor formatting on the
table, and alphabetized some out of order metrics.

Add the new metric introduced in #47731
to the reference page. Also performed some minor formatting on the
table, and alphabetized some out of order metrics.

Add the new metric introduced in #47731
to the reference page. Also performed some minor formatting on the
table, and alphabetized some out of order metrics.

Add the new metric introduced in #47731
to the reference page. Also performed some minor formatting on the
table, and alphabetized some out of order metrics.

A new `teleport_roles_total` is added to count the number of roles
in the cluster. The metric is intentionally updated infrequently
to avoid putting any additional strain on the backend. Additionally,
the reads performed when calculating the metric are rate limited
to prevent any bursts in reads when the metric timer does fire.

The auth periodic tickers have also been consolidated into a single
multi-interval instead of a ticker per operation. To prevent the
sub-intervals from impacting each other all operations are moved
to their own goroutine once their ticker fires.

* fix: specify a smallish page size

A new `teleport_roles_total` is added to count the number of roles
in the cluster. The metric is intentionally updated infrequently
to avoid putting any additional strain on the backend. Additionally,
the reads performed when calculating the metric are rate limited
to prevent any bursts in reads when the metric timer does fire.

The auth periodic tickers have also been consolidated into a single
multi-interval instead of a ticker per operation. To prevent the
sub-intervals from impacting each other all operations are moved
to their own goroutine once their ticker fires.

A new `teleport_roles_total` is added to count the number of roles
in the cluster. The metric is intentionally updated infrequently
to avoid putting any additional strain on the backend. Additionally,
the reads performed when calculating the metric are rate limited
to prevent any bursts in reads when the metric timer does fire.

The auth periodic tickers have also been consolidated into a single
multi-interval instead of a ticker per operation. To prevent the
sub-intervals from impacting each other all operations are moved
to their own goroutine once their ticker fires.

A new `teleport_roles_total` is added to count the number of roles
in the cluster. The metric is intentionally updated infrequently
to avoid putting any additional strain on the backend. Additionally,
the reads performed when calculating the metric are rate limited
to prevent any bursts in reads when the metric timer does fire.

The auth periodic tickers have also been consolidated into a single
multi-interval instead of a ticker per operation. To prevent the
sub-intervals from impacting each other all operations are moved
to their own goroutine once their ticker fires.

A new `teleport_roles_total` is added to count the number of roles
in the cluster. The metric is intentionally updated infrequently
to avoid putting any additional strain on the backend. Additionally,
the reads performed when calculating the metric are rate limited
to prevent any bursts in reads when the metric timer does fire.

The auth periodic tickers have also been consolidated into a single
multi-interval instead of a ticker per operation. To prevent the
sub-intervals from impacting each other all operations are moved
to their own goroutine once their ticker fires.

A new `teleport_roles_total` is added to count the number of roles
in the cluster. The metric is intentionally updated infrequently
to avoid putting any additional strain on the backend. Additionally,
the reads performed when calculating the metric are rate limited
to prevent any bursts in reads when the metric timer does fire.

The auth periodic tickers have also been consolidated into a single
multi-interval instead of a ticker per operation. To prevent the
sub-intervals from impacting each other all operations are moved
to their own goroutine once their ticker fires.

A new `teleport_roles_total` is added to count the number of roles
in the cluster. The metric is intentionally updated infrequently
to avoid putting any additional strain on the backend. Additionally,
the reads performed when calculating the metric are rate limited
to prevent any bursts in reads when the metric timer does fire.

The auth periodic tickers have also been consolidated into a single
multi-interval instead of a ticker per operation. To prevent the
sub-intervals from impacting each other all operations are moved
to their own goroutine once their ticker fires.

A new `teleport_roles_total` is added to count the number of roles
in the cluster. The metric is intentionally updated infrequently
to avoid putting any additional strain on the backend. Additionally,
the reads performed when calculating the metric are rate limited
to prevent any bursts in reads when the metric timer does fire.

The auth periodic tickers have also been consolidated into a single
multi-interval instead of a ticker per operation. To prevent the
sub-intervals from impacting each other all operations are moved
to their own goroutine once their ticker fires.

#47731 consolidated the auth periodics to use a multi.Interval,
however the upgrade window check called from the periodic launched
it's own periodic, leading to a runaway number of upgrade window
checks. This removes the extra periodic to reduce load upstream.

#47731 consolidated the auth periodics to use a multi.Interval,
however the upgrade window check called from the periodic launched
it's own periodic, leading to a runaway number of upgrade window
checks. This removes the extra periodic to reduce load upstream.

#47731 consolidated the auth periodics to use a multi.Interval,
however the upgrade window check called from the periodic launched
it's own periodic, leading to a runaway number of upgrade window
checks. This removes the extra periodic to reduce load upstream.

#47731 consolidated the auth periodics to use a multi.Interval,
however the upgrade window check called from the periodic launched
it's own periodic, leading to a runaway number of upgrade window
checks. This removes the extra periodic to reduce load upstream.

#47731 consolidated the auth periodics to use a multi.Interval,
however the upgrade window check called from the periodic launched
it's own periodic, leading to a runaway number of upgrade window
checks. This removes the extra periodic to reduce load upstream.

#47731 consolidated the auth periodics to use a multi.Interval,
however the upgrade window check called from the periodic launched
it's own periodic, leading to a runaway number of upgrade window
checks. This removes the extra periodic to reduce load upstream.

#47731 consolidated the auth periodics to use a multi.Interval,
however the upgrade window check called from the periodic launched
it's own periodic, leading to a runaway number of upgrade window
checks. This removes the extra periodic to reduce load upstream.

#47731 consolidated the auth periodics to use a multi.Interval,
however the upgrade window check called from the periodic launched
it's own periodic, leading to a runaway number of upgrade window
checks. This removes the extra periodic to reduce load upstream.

Add the new metric introduced in #47731
to the reference page. Also performed some minor formatting on the
table, and alphabetized some out of order metrics.

Add the new metric introduced in #47731
to the reference page. Also performed some minor formatting on the
table, and alphabetized some out of order metrics.