rust/neon
1
0
Fork 0
mirror of https://github.com/neondatabase/neon.git synced 2026-01-06 13:02:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat(pageserver): add global timeline count metrics (#12159)

Browse Source 
## Problem

We are getting tenants with a lot of branches and num of timelines is a
good indicator of pageserver loads. I added this metrics to help us
better plan pageserver capacities.

## Summary of changes

Add `pageserver_timeline_states_count` with two labels: active +
offloaded.

Signed-off-by: Alex Chi Z <chi@neon.tech>
This commit is contained in:
Alex Chi Z.
2025-06-09 17:57:36 +08:00
committed by GitHub
parent 7efd4554ab
commit 76f95f06d8
2 changed files with 40 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
pageserver/src/metrics.rs
View File

@@ -1053,6 +1053,15 @@ pub(crate) static TENANT_STATE_METRIC: Lazy<UIntGaugeVec> = Lazy::new(|| {
.expect("Failed to register pageserver_tenant_states_count metric")
});
pub(crate) static TIMELINE_STATE_METRIC: Lazy<UIntGaugeVec> = Lazy::new(|| {
register_uint_gauge_vec!(
"pageserver_timeline_states_count",
"Count of timelines per state",
&["state"]
)
.expect("Failed to register pageserver_timeline_states_count metric")
});
/// A set of broken tenants.
///
/// These are expected to be so rare that a set is fine. Set as in a new timeseries per each broken
@@ -3325,6 +3334,8 @@ impl TimelineMetrics {
&timeline_id,
);
TIMELINE_STATE_METRIC.with_label_values(&["active"]).inc();
TimelineMetrics {
tenant_id,
shard_id,
@@ -3479,6 +3490,8 @@ impl TimelineMetrics {
return;
}
TIMELINE_STATE_METRIC.with_label_values(&["active"]).dec();
let tenant_id = &self.tenant_id;
let timeline_id = &self.timeline_id;
let shard_id = &self.shard_id;

28
pageserver/src/tenant.rs
View File

@@ -89,7 +89,8 @@ use crate::l0_flush::L0FlushGlobalState;
use crate::metrics::{
BROKEN_TENANTS_SET, CIRCUIT_BREAKERS_BROKEN, CIRCUIT_BREAKERS_UNBROKEN, CONCURRENT_INITDBS,
INITDB_RUN_TIME, INITDB_SEMAPHORE_ACQUISITION_TIME, TENANT, TENANT_OFFLOADED_TIMELINES,
TENANT_STATE_METRIC, TENANT_SYNTHETIC_SIZE_METRIC, remove_tenant_metrics,
TENANT_STATE_METRIC, TENANT_SYNTHETIC_SIZE_METRIC, TIMELINE_STATE_METRIC,
remove_tenant_metrics,
};
use crate::task_mgr::TaskKind;
use crate::tenant::config::LocationMode;
@@ -544,6 +545,28 @@ pub struct OffloadedTimeline {
/// Part of the `OffloadedTimeline` object's lifecycle: this needs to be set before we drop it
pub deleted_from_ancestor: AtomicBool,
_metrics_guard: OffloadedTimelineMetricsGuard,
}
/// Increases the offloaded timeline count metric when created, and decreases when dropped.
struct OffloadedTimelineMetricsGuard;
impl OffloadedTimelineMetricsGuard {
fn new() -> Self {
TIMELINE_STATE_METRIC
.with_label_values(&["offloaded"])
.inc();
Self
}
}
impl Drop for OffloadedTimelineMetricsGuard {
fn drop(&mut self) {
TIMELINE_STATE_METRIC
.with_label_values(&["offloaded"])
.dec();
}
}
impl OffloadedTimeline {
@@ -576,6 +599,8 @@ impl OffloadedTimeline {
delete_progress: timeline.delete_progress.clone(),
deleted_from_ancestor: AtomicBool::new(false),
_metrics_guard: OffloadedTimelineMetricsGuard::new(),
})
}
fn from_manifest(tenant_shard_id: TenantShardId, manifest: &OffloadedTimelineManifest) -> Self {
@@ -595,6 +620,7 @@ impl OffloadedTimeline {
archived_at,
delete_progress: TimelineDeleteProgress::default(),
deleted_from_ancestor: AtomicBool::new(false),
_metrics_guard: OffloadedTimelineMetricsGuard::new(),
}
}
fn manifest(&self) -> OffloadedTimelineManifest {