docs: remove references to dead metric tcp_connection_duration_ms (#…

…1905)

* docs: remove references to dead metric `tcp_connection_duration_ms`

This metric was dropped in linkerd/linkerd2-proxy#930

Signed-off-by: Nicklas Møller <[email protected]>

* docs: remove `tcp_connection_duration_ms` reference from 2.10

Signed-off-by: Nicklas Møller <[email protected]>

* docs: remove references to tcp_connection_duration_ms altogether

This metrics was removed in 2.10, but was incorrectly attributed to 2.11
in the upgrade docs, as of e9ae777#diff-f4328e2dffc88e1711b12c9719e0f948d4bdd73be957179eeb3ceacc8391fe29

It was never incorporated into the original 2.10 docs: https://github.com/linkerd/website/blob/e9ae7770ea685336aa9b2aa0b9f802b028f9aa76/linkerd.io/content/2.10/tasks/upgrade.md

Opting to just leave it out - it's been years on end, and the metric
itself was taken out due to a memory leak: it's not super likely that
someone will come around in the future, missing the metric, after
potentially upgrading from 2.10.

Signed-off-by: Nicklas Møller <[email protected]>

---------

Signed-off-by: Nicklas Møller <[email protected]>
Co-authored-by: Flynn <[email protected]>

linkerd.io/content/2-edge/reference/proxy-metrics.md

@@ -192,8 +192,6 @@ layer.
   updated when the connection closes.
 * `tcp_read_bytes_total`: A counter of the total number of received bytes. This
   is updated when the connection closes.
-* `tcp_connection_duration_ms`: A histogram of the duration of the lifetime of a
-  connection, in milliseconds. This is updated when the connection closes.
 * `inbound_tcp_errors_total`: A counter of the total number of inbound TCP
   connections that could not be processed due to a proxy error.
 * `outbound_tcp_errors_total`: A counter of the total number of outbound TCP
@@ -220,7 +218,7 @@ are also added to transport-level metrics, when applicable.
 #### Connection Close Labels
 
 The following labels are added only to metrics which are updated when a
-connection closes (`tcp_close_total` and `tcp_connection_duration_ms`):
+connection closes (`tcp_close_total`):
 
 * `classification`: `success` if the connection terminated cleanly, `failure` if
   the connection closed due to a connection failure.

linkerd.io/content/2-edge/tasks/upgrade.md

@@ -591,7 +591,6 @@ be aware of when upgrading from `stable-2.10.x`:
  result of their deprecation in Kubernetes v1.21 and above. The control plane
  and core extensions will now be shipped without PSPs; they can be enabled
  through a new install option `enablePSP: true`.
-- The `tcp_connection_duration_ms` metric has been removed.
 - Opaque ports changes: `443` is no longer included in the default opaque ports
  list. Ports `4444`, `6379` and `9300` corresponding to Galera, Redis and
  ElasticSearch respectively (all server speak first protocols) have been added

linkerd.io/content/2.10/reference/proxy-metrics.md

@@ -164,8 +164,6 @@ layer.
   updated when the connection closes.
 * `tcp_read_bytes_total`: A counter of the total number of received bytes. This
   is updated when the connection closes.
-* `tcp_connection_duration_ms`: A histogram of the duration of the lifetime of a
-  connection, in milliseconds. This is updated when the connection closes.
 
 ### Labels
 
@@ -184,7 +182,7 @@ are also added to transport-level metrics, when applicable.
 #### Connection Close Labels
 
 The following labels are added only to metrics which are updated when a
-connection closes (`tcp_close_total` and `tcp_connection_duration_ms`):
+connection closes (`tcp_close_total`):
 
 * `classification`: `success` if the connection terminated cleanly, `failure` if
   the connection closed due to a connection failure.

linkerd.io/content/2.11/reference/proxy-metrics.md

@@ -164,8 +164,6 @@ layer.
   updated when the connection closes.
 * `tcp_read_bytes_total`: A counter of the total number of received bytes. This
   is updated when the connection closes.
-* `tcp_connection_duration_ms`: A histogram of the duration of the lifetime of a
-  connection, in milliseconds. This is updated when the connection closes.
 
 ### Labels
 
@@ -184,7 +182,7 @@ are also added to transport-level metrics, when applicable.
 #### Connection Close Labels
 
 The following labels are added only to metrics which are updated when a
-connection closes (`tcp_close_total` and `tcp_connection_duration_ms`):
+connection closes (`tcp_close_total`):
 
 * `classification`: `success` if the connection terminated cleanly, `failure` if
   the connection closed due to a connection failure.

linkerd.io/content/2.11/tasks/upgrade.md

@@ -294,7 +294,6 @@ be aware of when upgrading from `stable-2.10.x`:
  result of their deprecation in Kubernetes v1.21 and above. The control plane
  and core extensions will now be shipped without PSPs; they can be enabled
  through a new install option `enablePSP: true`.
-- The `tcp_connection_duration_ms` metric has been removed.
 - Opaque ports changes: `443` is no longer included in the default opaque ports
  list. Ports `4444`, `6379` and `9300` corresponding to Galera, Redis and
  ElasticSearch respectively (all server speak first protocols) have been added

linkerd.io/content/2.12/reference/proxy-metrics.md

@@ -192,8 +192,6 @@ layer.
   updated when the connection closes.
 * `tcp_read_bytes_total`: A counter of the total number of received bytes. This
   is updated when the connection closes.
-* `tcp_connection_duration_ms`: A histogram of the duration of the lifetime of a
-  connection, in milliseconds. This is updated when the connection closes.
 * `inbound_tcp_errors_total`: A counter of the total number of inbound TCP
   connections that could not be processed due to a proxy error.
 * `outbound_tcp_errors_total`: A counter of the total number of outbound TCP
@@ -220,7 +218,7 @@ are also added to transport-level metrics, when applicable.
 #### Connection Close Labels
 
 The following labels are added only to metrics which are updated when a
-connection closes (`tcp_close_total` and `tcp_connection_duration_ms`):
+connection closes (`tcp_close_total`):
 
 * `classification`: `success` if the connection terminated cleanly, `failure` if
   the connection closed due to a connection failure.

linkerd.io/content/2.12/tasks/upgrade.md

@@ -501,7 +501,6 @@ be aware of when upgrading from `stable-2.10.x`:
  result of their deprecation in Kubernetes v1.21 and above. The control plane
  and core extensions will now be shipped without PSPs; they can be enabled
  through a new install option `enablePSP: true`.
-- The `tcp_connection_duration_ms` metric has been removed.
 - Opaque ports changes: `443` is no longer included in the default opaque ports
  list. Ports `4444`, `6379` and `9300` corresponding to Galera, Redis and
  ElasticSearch respectively (all server speak first protocols) have been added

linkerd.io/content/2.13/reference/proxy-metrics.md

@@ -192,8 +192,6 @@ layer.
   updated when the connection closes.
 * `tcp_read_bytes_total`: A counter of the total number of received bytes. This
   is updated when the connection closes.
-* `tcp_connection_duration_ms`: A histogram of the duration of the lifetime of a
-  connection, in milliseconds. This is updated when the connection closes.
 * `inbound_tcp_errors_total`: A counter of the total number of inbound TCP
   connections that could not be processed due to a proxy error.
 * `outbound_tcp_errors_total`: A counter of the total number of outbound TCP
@@ -220,7 +218,7 @@ are also added to transport-level metrics, when applicable.
 #### Connection Close Labels
 
 The following labels are added only to metrics which are updated when a
-connection closes (`tcp_close_total` and `tcp_connection_duration_ms`):
+connection closes (`tcp_close_total`):
 
 * `classification`: `success` if the connection terminated cleanly, `failure` if
   the connection closed due to a connection failure.

linkerd.io/content/2.13/tasks/upgrade.md

@@ -514,7 +514,6 @@ be aware of when upgrading from `stable-2.10.x`:
  result of their deprecation in Kubernetes v1.21 and above. The control plane
  and core extensions will now be shipped without PSPs; they can be enabled
  through a new install option `enablePSP: true`.
-- The `tcp_connection_duration_ms` metric has been removed.
 - Opaque ports changes: `443` is no longer included in the default opaque ports
  list. Ports `4444`, `6379` and `9300` corresponding to Galera, Redis and
  ElasticSearch respectively (all server speak first protocols) have been added

linkerd.io/content/2.14/reference/proxy-metrics.md

@@ -192,8 +192,6 @@ layer.
   updated when the connection closes.
 * `tcp_read_bytes_total`: A counter of the total number of received bytes. This
   is updated when the connection closes.
-* `tcp_connection_duration_ms`: A histogram of the duration of the lifetime of a
-  connection, in milliseconds. This is updated when the connection closes.
 * `inbound_tcp_errors_total`: A counter of the total number of inbound TCP
   connections that could not be processed due to a proxy error.
 * `outbound_tcp_errors_total`: A counter of the total number of outbound TCP
@@ -220,7 +218,7 @@ are also added to transport-level metrics, when applicable.
 #### Connection Close Labels
 
 The following labels are added only to metrics which are updated when a
-connection closes (`tcp_close_total` and `tcp_connection_duration_ms`):
+connection closes (`tcp_close_total`):
 
 * `classification`: `success` if the connection terminated cleanly, `failure` if
   the connection closed due to a connection failure.

linkerd.io/content/2.14/tasks/upgrade.md

@@ -529,7 +529,6 @@ be aware of when upgrading from `stable-2.10.x`:
  result of their deprecation in Kubernetes v1.21 and above. The control plane
  and core extensions will now be shipped without PSPs; they can be enabled
  through a new install option `enablePSP: true`.
-- The `tcp_connection_duration_ms` metric has been removed.
 - Opaque ports changes: `443` is no longer included in the default opaque ports
  list. Ports `4444`, `6379` and `9300` corresponding to Galera, Redis and
  ElasticSearch respectively (all server speak first protocols) have been added

linkerd.io/content/2.15/reference/proxy-metrics.md

@@ -192,8 +192,6 @@ layer.
   updated when the connection closes.
 * `tcp_read_bytes_total`: A counter of the total number of received bytes. This
   is updated when the connection closes.
-* `tcp_connection_duration_ms`: A histogram of the duration of the lifetime of a
-  connection, in milliseconds. This is updated when the connection closes.
 * `inbound_tcp_errors_total`: A counter of the total number of inbound TCP
   connections that could not be processed due to a proxy error.
 * `outbound_tcp_errors_total`: A counter of the total number of outbound TCP
@@ -220,7 +218,7 @@ are also added to transport-level metrics, when applicable.
 #### Connection Close Labels
 
 The following labels are added only to metrics which are updated when a
-connection closes (`tcp_close_total` and `tcp_connection_duration_ms`):
+connection closes (`tcp_close_total`):
 
 * `classification`: `success` if the connection terminated cleanly, `failure` if
   the connection closed due to a connection failure.

linkerd.io/content/2.15/tasks/upgrade.md

@@ -591,7 +591,6 @@ be aware of when upgrading from `stable-2.10.x`:
  result of their deprecation in Kubernetes v1.21 and above. The control plane
  and core extensions will now be shipped without PSPs; they can be enabled
  through a new install option `enablePSP: true`.
-- The `tcp_connection_duration_ms` metric has been removed.
 - Opaque ports changes: `443` is no longer included in the default opaque ports
  list. Ports `4444`, `6379` and `9300` corresponding to Galera, Redis and
  ElasticSearch respectively (all server speak first protocols) have been added

linkerd.io/content/2.16/reference/proxy-metrics.md

@@ -192,8 +192,6 @@ layer.
   updated when the connection closes.
 * `tcp_read_bytes_total`: A counter of the total number of received bytes. This
   is updated when the connection closes.
-* `tcp_connection_duration_ms`: A histogram of the duration of the lifetime of a
-  connection, in milliseconds. This is updated when the connection closes.
 * `inbound_tcp_errors_total`: A counter of the total number of inbound TCP
   connections that could not be processed due to a proxy error.
 * `outbound_tcp_errors_total`: A counter of the total number of outbound TCP
@@ -220,7 +218,7 @@ are also added to transport-level metrics, when applicable.
 #### Connection Close Labels
 
 The following labels are added only to metrics which are updated when a
-connection closes (`tcp_close_total` and `tcp_connection_duration_ms`):
+connection closes (`tcp_close_total`):
 
 * `classification`: `success` if the connection terminated cleanly, `failure` if
   the connection closed due to a connection failure.

linkerd.io/content/2.16/tasks/upgrade.md

@@ -591,7 +591,6 @@ be aware of when upgrading from `stable-2.10.x`:
  result of their deprecation in Kubernetes v1.21 and above. The control plane
  and core extensions will now be shipped without PSPs; they can be enabled
  through a new install option `enablePSP: true`.
-- The `tcp_connection_duration_ms` metric has been removed.
 - Opaque ports changes: `443` is no longer included in the default opaque ports
  list. Ports `4444`, `6379` and `9300` corresponding to Galera, Redis and
  ElasticSearch respectively (all server speak first protocols) have been added

linkerd.io/content/2.17/reference/proxy-metrics.md

@@ -192,8 +192,6 @@ layer.
   updated when the connection closes.
 * `tcp_read_bytes_total`: A counter of the total number of received bytes. This
   is updated when the connection closes.
-* `tcp_connection_duration_ms`: A histogram of the duration of the lifetime of a
-  connection, in milliseconds. This is updated when the connection closes.
 * `inbound_tcp_errors_total`: A counter of the total number of inbound TCP
   connections that could not be processed due to a proxy error.
 * `outbound_tcp_errors_total`: A counter of the total number of outbound TCP
@@ -220,7 +218,7 @@ are also added to transport-level metrics, when applicable.
 #### Connection Close Labels
 
 The following labels are added only to metrics which are updated when a
-connection closes (`tcp_close_total` and `tcp_connection_duration_ms`):
+connection closes (`tcp_close_total`):
 
 * `classification`: `success` if the connection terminated cleanly, `failure` if
   the connection closed due to a connection failure.

linkerd.io/content/2.17/tasks/upgrade.md

@@ -591,7 +591,6 @@ be aware of when upgrading from `stable-2.10.x`:
  result of their deprecation in Kubernetes v1.21 and above. The control plane
  and core extensions will now be shipped without PSPs; they can be enabled
  through a new install option `enablePSP: true`.
-- The `tcp_connection_duration_ms` metric has been removed.
 - Opaque ports changes: `443` is no longer included in the default opaque ports
  list. Ports `4444`, `6379` and `9300` corresponding to Galera, Redis and
  ElasticSearch respectively (all server speak first protocols) have been added