:hedgehog: PostHog is the leading platform for building self-driving products. Our developer tools – AI observability, analytics, session replay, flags, experiments, error tracking, logs, and more – capture all the context agents need to diagnose problems, uncover opportunities, and ship fixes. Steer it all from Slack, web, desktop, or the MCP.
Python ab-testingai-analyticsanalyticscdpdata-warehouseexperimentsfeature-flagsjavascriptproduct-analyticspythonreactsession-replaysurveystypescriptweb-analytics
pub const BACKOFF_EVENTS_TOTAL: &str = "batch_import_backoff_events_total";
pub const BACKOFF_DELAY_SECONDS: &str = "batch_import_backoff_delay_seconds";
pub const UNPAUSE_TOTAL: &str = "batch_import_unpause_total";
pub const STAGING_SWEEP_REMOVED: &str = "batch_import_staging_sweep_removed_total";
pub const STAGING_DIR_BYTES: &str = "batch_import_staging_dir_bytes";
pub const STAGING_GUARD_TRIPPED: &str = "batch_import_staging_guard_tripped_total";
pub const ACTIVE_JOBS: &str = "batch_import_active_jobs";
pub const TEMP_BUCKET_STAGE_BYTES: &str = "batch_import_temp_bucket_stage_bytes";
pub const TEMP_BUCKET_STAGE_DURATION_SECONDS: &str =
"batch_import_temp_bucket_stage_duration_seconds";
pub const TEMP_BUCKET_READ_DURATION_SECONDS: &str =
"batch_import_temp_bucket_read_duration_seconds";
pub const STAGED_PLAINTEXT_CEILING_TRIPPED: &str =
"batch_import_staged_plaintext_ceiling_tripped_total";
pub const PART_CLEANUP_TOTAL: &str = "batch_import_part_cleanup_total";
pub const JOB_COMPLETION_TOTAL: &str = "batch_import_job_completion_total";
pub const COMPLETED_JOB_EVENTS: &str = "batch_import_completed_job_events";
pub const COMPLETED_JOB_PARTS: &str = "batch_import_completed_job_parts";
pub const OPERATOR_STOP_TOTAL: &str = "batch_import_operator_stop_total";
pub const INTERRUPTED_COMMIT_PAUSED_TOTAL: &str = "batch_import_interrupted_commit_paused_total";
pub const INTERRUPTED_COMMIT_RESENT_TOTAL: &str = "batch_import_interrupted_commit_resent_total";
use metrics::{counter, gauge, histogram};
pub fn backoff_event(delay_secs: f64) {
counter!(BACKOFF_EVENTS_TOTAL).increment(1);
histogram!(BACKOFF_DELAY_SECONDS).record(delay_secs);
}
pub fn unpause_event() {
counter!(UNPAUSE_TOTAL).increment(1);
}
/// Count jobs a worker stopped because the row left the running status, by commit stage
/// ("before_send" / "after_send").
pub fn operator_stop(stage: &'static str) {
counter!(OPERATOR_STOP_TOTAL, "stage" => stage).increment(1);
}
/// Count jobs paused at claim because an earlier worker stopped mid-commit.
pub fn interrupted_commit_paused() {
counter!(INTERRUPTED_COMMIT_PAUSED_TOTAL).increment(1);
}
/// Count jobs rewound at claim to re-send a chunk an earlier worker stopped sending.
pub fn interrupted_commit_resent() {
counter!(INTERRUPTED_COMMIT_RESENT_TOTAL).increment(1);
}
pub fn staging_sweep_removed(count: u64) {
counter!(STAGING_SWEEP_REMOVED).increment(count);
}
pub fn staging_dir_bytes(bytes: f64) {
gauge!(STAGING_DIR_BYTES).set(bytes);
}
pub fn staging_guard_tripped() {
counter!(STAGING_GUARD_TRIPPED).increment(1);
}
/// Number of batch import jobs currently needing a worker (queued and in-flight).
/// Reported by every replica off the shared Postgres queue; the KEDA autoscaler
/// collapses the per-pod duplicates with `max()` to drive replica count.
pub fn active_jobs(count: f64) {
gauge!(ACTIVE_JOBS).set(count);
}
/// Record a completed temp-bucket part upload: total bytes staged and wall-clock duration.
pub fn temp_bucket_part_staged(bytes: u64, duration_secs: f64) {
histogram!(TEMP_BUCKET_STAGE_BYTES).record(bytes as f64);
histogram!(TEMP_BUCKET_STAGE_DURATION_SECONDS).record(duration_secs);
}
/// Record the latency of a single ranged GET against the temp bucket.
pub fn temp_bucket_read(duration_secs: f64) {
histogram!(TEMP_BUCKET_READ_DURATION_SECONDS).record(duration_secs);
}
/// Count a part that breached STAGED_PLAINTEXT_MAX_BYTES and paused the job.
pub fn staged_plaintext_ceiling_tripped() {
counter!(STAGED_PLAINTEXT_CEILING_TRIPPED).increment(1);
}
/// Count post-commit staging cleanup of a completed part, by outcome ("ok" / "error").
/// A failed cleanup leaks only transient storage (reclaimed by job cleanup / bucket TTL).
pub fn part_cleanup(outcome: &'static str) {
counter!(PART_CLEANUP_TOTAL, "outcome" => outcome).increment(1);
}
/// Count a job that reached the end of its parts, by outcome: "completed",
/// "completed_empty" (no events emitted) or "blocked" (paused instead of completed).
pub fn job_completion(outcome: &'static str) {
counter!(JOB_COMPLETION_TOTAL, "outcome" => outcome).increment(1);
}
/// Record the size of a completed job, so silent partial imports show up as outliers.
pub fn completed_job_size(parts: usize, events: u64) {
histogram!(COMPLETED_JOB_PARTS).record(parts as f64);
histogram!(COMPLETED_JOB_EVENTS).record(events as f64);
}