Adds synchronous and asynchronous OpenTelemetry metrics #43
Conversation
This file contains 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
obitech
reviewed
Dec 10, 2024
costela
requested changes
Jan 7, 2025
| // WithAttributes specifies additional attributes to be added to the span. | ||
| func WithAttributes(attrs ...attribute.KeyValue) Option { | ||
| // WithTracerAttributes specifies additional attributes to be added to spans. | ||
| func WithTracerAttributes(attrs ...attribute.KeyValue) Option { |
There was a problem hiding this comment.
This is a breaking change, right? Can we maybe add a wrapper function with the old name and a Deprecated marker to avoid it?
| pgxPoolTotalConnections = "pgxpool.total_connections" | ||
| ) | ||
|
|
||
| // RecordStats records database statistics for provided pgxpool.Pool at the provided interval. |
There was a problem hiding this comment.
Suggested change
| // RecordStats records database statistics for provided pgxpool.Pool at the provided interval. | |
| // RecordStats records database statistics for provided pgxpool.Pool. |
the interval is now AFAICT handled via WithMinimumReadDBStatsInterval?
| } | ||
|
|
||
| // WithStatsMeterProvider sets meter provider to use for pgx stat metric collection. | ||
| func WithStatsMeterProvider(provider metric.MeterProvider) StatsOption { |
There was a problem hiding this comment.
sorry, I'm a bit confused by this vs WithMeterProvider: is it worth it to have two ways of setting it? 🤔
| // recordSpanError handles all error handling to be applied on the provided span. | ||
| // The provided error must be non-nil and not a sql.ErrNoRows error. | ||
| // Otherwise, recordSpanError will be a no-op. | ||
| func (t *Tracer) recordSpanError(span trace.Span, err error) { |
There was a problem hiding this comment.
AFAICT we're not using the Tracer here? So why make it a method? What am I missing?
Comment on lines
+62
to
+83
| err error | ||
|
|
||
| // Asynchronous Observable Metrics | ||
| acquireCount metric.Int64ObservableCounter | ||
| acquireDuration metric.Int64ObservableCounter | ||
| acquiredConns metric.Int64ObservableUpDownCounter | ||
| cancelledAcquires metric.Int64ObservableCounter | ||
| constructingConns metric.Int64ObservableUpDownCounter | ||
| emptyAcquires metric.Int64ObservableCounter | ||
| idleConns metric.Int64ObservableUpDownCounter | ||
| maxConns metric.Int64ObservableGauge | ||
| maxIdleDestroyCount metric.Int64ObservableCounter | ||
| maxLifetimeDestroyCount metric.Int64ObservableCounter | ||
| newConnsCount metric.Int64ObservableCounter | ||
| totalConns metric.Int64ObservableUpDownCounter | ||
|
|
||
| observeOptions []metric.ObserveOption | ||
| serverAddress = semconv.ServerAddress(db.Config().ConnConfig.Host) | ||
| serverPort = semconv.ServerPort(int(db.Config().ConnConfig.Port)) | ||
| dbNamespace = semconv.DBNamespace(db.Config().ConnConfig.Database) | ||
| poolName = fmt.Sprintf("%s:%d/%s", serverAddress.Value.AsString(), serverPort.Value.AsInt64(), dbNamespace.Value.AsString()) | ||
| dbClientConnectionPoolName = semconv.DBClientConnectionPoolName(poolName) |
There was a problem hiding this comment.
nit: can we just refactor the instantiations below into := and get rid of this big block? This smells a bit like C 😅
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Problem Statement:
otelpgx, as of today, is only responsible for tracing observability.Clients of
pgxandotelpgxare forced to manage their own metric observability which can lead to:pgxbecause of implementation redundancyCurrently, PR #18 exists to add PGX Stats as Asynchronous Observable OpenTelemetry metrics.
This PR, while also instruments the same asynchronous metrics, aims to do so with a few changes and alongside additional features:
This PR also implicitly resolves issue #40.
PR #18 seems to have left off awaiting information for how to have OTLP metrics scraped by Prometheus.
Those instructions, at least to me, seem outside the scope of
otelpgxsince that is more infrastructure-setup specifics, which OpenTelemetry-ecosystem tends to be agnostic to, and could become a chore to maintain long-term.Because of that, I haven't included any specifics in the README as part of this PR; though if felt strongly otherwise, I can of course do so.
In my own environment (Application -> Collector <- Prometheus), I am able to verify metrics being reported.
See attached screenshots:
Screenshots