feat(swarm): expose ConnectionId and add conn duration metric
#3927
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -18,20 +18,25 @@ | |
| // FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER | ||
| // DEALINGS IN THE SOFTWARE. | ||
|
|
||
| use std::collections::HashMap; | ||
| use std::sync::{Arc, Mutex}; | ||
|
|
||
| use crate::protocol_stack; | ||
| use instant::Instant; | ||
| use libp2p_swarm::ConnectionId; | ||
| use prometheus_client::encoding::{EncodeLabelSet, EncodeLabelValue}; | ||
| use prometheus_client::metrics::counter::Counter; | ||
| use prometheus_client::metrics::family::Family; | ||
| use prometheus_client::metrics::histogram::{exponential_buckets, Histogram}; | ||
| use prometheus_client::registry::{Registry, Unit}; | ||
|
|
||
| pub(crate) struct Metrics { | ||
| connections_incoming: Family<AddressLabels, Counter>, | ||
| connections_incoming_error: Family<IncomingConnectionErrorLabels, Counter>, | ||
|
|
||
| connections_established: Family<ConnectionLabels, Counter>, | ||
| connections_establishment_duration: Family<ConnectionLabels, Histogram>, | ||
| connections_duration: Family<ConnectionClosedLabels, Histogram>, | ||
|
|
||
| new_listen_addr: Family<AddressLabels, Counter>, | ||
| expired_listen_addr: Family<AddressLabels, Counter>, | ||
|
|
@@ -41,6 +46,8 @@ pub(crate) struct Metrics { | |
|
|
||
| dial_attempt: Counter, | ||
| outgoing_connection_error: Family<OutgoingConnectionErrorLabels, Counter>, | ||
|
|
||
| connections: Arc<Mutex<HashMap<ConnectionId, Instant>>>, | ||
| } | ||
|
|
||
| impl Metrics { | ||
|
|
@@ -110,34 +117,42 @@ impl Metrics { | |
| connections_established.clone(), | ||
| ); | ||
|
|
||
| let connections_establishment_duration = { | ||
| let constructor: fn() -> Histogram = | ||
| || Histogram::new(exponential_buckets(0.01, 1.5, 20)); | ||
| Family::new_with_constructor(constructor) | ||
| }; | ||
| sub_registry.register( | ||
| "connections_establishment_duration", | ||
| "Time it took (locally) to establish connections", | ||
| connections_establishment_duration.clone(), | ||
| ); | ||
|
|
||
| let connections_duration = { | ||
| let constructor: fn() -> Histogram = | ||
| || Histogram::new(exponential_buckets(0.01, 3.0, 20)); | ||
| Family::new_with_constructor(constructor) | ||
| }; | ||
| sub_registry.register_with_unit( | ||
| "connections_establishment_duration", | ||
| "Time it took (locally) to establish connections", | ||
| Unit::Seconds, | ||
| connections_establishment_duration.clone(), | ||
| ); | ||
|
|
||
| Self { | ||
| connections_incoming, | ||
| connections_incoming_error, | ||
| connections_established, | ||
| new_listen_addr, | ||
| expired_listen_addr, | ||
| listener_closed, | ||
| listener_error, | ||
| dial_attempt, | ||
| outgoing_connection_error, | ||
| connections_establishment_duration, | ||
| connections_duration, | ||
| connections: Default::default(), | ||
| } | ||
| } | ||
| } | ||
|
|
@@ -149,24 +164,44 @@ impl<TBvEv, THandleErr> super::Recorder<libp2p_swarm::SwarmEvent<TBvEv, THandleE | |
| libp2p_swarm::SwarmEvent::ConnectionEstablished { | ||
| endpoint, | ||
| established_in: time_taken, | ||
| connection_id, | ||
| .. | ||
| } => { | ||
| let labels = ConnectionLabels { | ||
| role: endpoint.into(), | ||
| protocols: protocol_stack::as_string(endpoint.get_remote_address()), | ||
| }; | ||
| self.connections_established.get_or_create(&labels).inc(); | ||
| self.connections_establishment_duration | ||
| .get_or_create(&labels) | ||
| .observe(time_taken.as_secs_f64()); | ||
| self.connections | ||
| .lock() | ||
| .expect("lock not to be poisoned") | ||
| .insert(*connection_id, Instant::now()); | ||
| } | ||
| libp2p_swarm::SwarmEvent::ConnectionClosed { | ||
| endpoint, | ||
| connection_id, | ||
| cause, | ||
| .. | ||
| } => { | ||
| let labels = ConnectionClosedLabels { | ||
| connection: ConnectionLabels { | ||
| role: endpoint.into(), | ||
| protocols: protocol_stack::as_string(endpoint.get_remote_address()), | ||
| }, | ||
| cause: cause.as_ref().expect("TODO remove see definition").into(), | ||
| }; | ||
| self.connections_duration.get_or_create(&labels).observe( | ||
| self.connections | ||
| .lock() | ||
| .expect("lock not to be poisoned") | ||
| .remove(connection_id) | ||
| .expect("closed connection to previously be established") | ||
| .elapsed() | ||
| .as_secs_f64(), | ||
| ); | ||
| } | ||
| libp2p_swarm::SwarmEvent::IncomingConnection { send_back_addr, .. } => { | ||
| self.connections_incoming | ||
|
|
@@ -187,7 +222,7 @@ impl<TBvEv, THandleErr> super::Recorder<libp2p_swarm::SwarmEvent<TBvEv, THandleE | |
| }) | ||
| .inc(); | ||
| } | ||
| libp2p_swarm::SwarmEvent::OutgoingConnectionError { error, peer_id, .. } => { | ||
| let peer = match peer_id { | ||
| Some(_) => PeerStatus::Known, | ||
| None => PeerStatus::Unknown, | ||
|
|
@@ -261,25 +296,42 @@ impl<TBvEv, THandleErr> super::Recorder<libp2p_swarm::SwarmEvent<TBvEv, THandleE | |
| libp2p_swarm::SwarmEvent::ListenerError { .. } => { | ||
| self.listener_error.inc(); | ||
| } | ||
| libp2p_swarm::SwarmEvent::Dialing { .. } => { | ||
| self.dial_attempt.inc(); | ||
| } | ||
| } | ||
| } | ||
| } | ||
|
|
||
| #[derive(EncodeLabelSet, Hash, Clone, Eq, PartialEq, Debug)] | ||
| struct ConnectionLabels { | ||
| role: Role, | ||
| protocols: String, | ||
| } | ||
|
|
||
| #[derive(EncodeLabelSet, Hash, Clone, Eq, PartialEq, Debug)] | ||
| struct ConnectionClosedLabels { | ||
| // TODO: Should be Option<ConnectionError>. Needs https://github.com/prometheus/client_rust/pull/137 | ||
|
Contributor
There was a problem hiding this comment. Do we wait for this before merging here?
Member
Author
There was a problem hiding this comment. Oversight. Fixed with 9578315. Thanks for the catch Thomas! |
||
| cause: ConnectionError, | ||
| #[prometheus(flatten)] | ||
| connection: ConnectionLabels, | ||
| } | ||
|
|
||
| #[derive(EncodeLabelValue, Hash, Clone, Eq, PartialEq, Debug)] | ||
| enum ConnectionError { | ||
| Io, | ||
| KeepAliveTimeout, | ||
| Handler, | ||
| } | ||
|
|
||
| impl<E> From<&libp2p_swarm::ConnectionError<E>> for ConnectionError { | ||
| fn from(value: &libp2p_swarm::ConnectionError<E>) -> Self { | ||
| match value { | ||
| libp2p_swarm::ConnectionError::IO(_) => ConnectionError::Io, | ||
| libp2p_swarm::ConnectionError::KeepAliveTimeout => ConnectionError::KeepAliveTimeout, | ||
| libp2p_swarm::ConnectionError::Handler(_) => ConnectionError::Handler, | ||
| } | ||
| } | ||
| } | ||
|
|
||
| #[derive(EncodeLabelSet, Hash, Clone, Eq, PartialEq, Debug)] | ||
|
|
@@ -359,7 +411,3 @@ impl From<&libp2p_swarm::ListenError> for IncomingConnectionError { | |
| } | ||
| } | ||
| } | ||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Uh oh!
There was an error while loading. Please reload this page.