[pdm-devel] [PATCH proxmox-datacenter-manager 17/25] metric collection: save time needed for collection run to RRD

[Lukas Wagner]

For large setups, it might be useful to know how much time was needed to
collect metrics for *all* remotes together, e.g. for making sure that
the collection interval is not exceeded.

Signed-off-by: Lukas Wagner <l.wagner at proxmox.com>
---
 .../src/metric_collection/collection_task.rs  | 15 ++++++
 server/src/metric_collection/rrd_task.rs      | 50 +++++++++++++++----
 2 files changed, 56 insertions(+), 9 deletions(-)

diff --git a/server/src/metric_collection/collection_task.rs b/server/src/metric_collection/collection_task.rs
index 27a9987..501df9b 100644
--- a/server/src/metric_collection/collection_task.rs
+++ b/server/src/metric_collection/collection_task.rs
@@ -22,6 +22,7 @@ use pdm_api_types::{
 };
 use pdm_config::metric_collection::COLLECTION_SETTINGS_TYPE;
 
+use crate::metric_collection::rrd_task::CollectionStats;
 use crate::{connection, task_utils};
 
 use super::{
@@ -97,6 +98,20 @@ impl MetricCollectionTask {
                     if let Some(remotes) = Self::load_remote_config() {
                         let to_fetch = remotes.order.as_slice();
                         self.fetch_remotes(&remotes, to_fetch).await;
+
+                        let now = Instant::now();
+                        self.fetch_remotes(&remotes, &remotes.order).await;
+                        let elapsed = now.elapsed();
+
+                        if let Err(err) = self.metric_data_tx.send(
+                            RrdStoreRequest::CollectionStats {
+                                timestamp: proxmox_time::epoch_i64(),
+                                stats: CollectionStats {
+                                    total_time: elapsed.as_secs_f64() * 1000.
+                                }
+                            }).await {
+                            log::error!("could not send collection stats to rrd task: {err}");
+                        }
                     }
                 }
 
diff --git a/server/src/metric_collection/rrd_task.rs b/server/src/metric_collection/rrd_task.rs
index 7f20d2d..6a4bacf 100644
--- a/server/src/metric_collection/rrd_task.rs
+++ b/server/src/metric_collection/rrd_task.rs
@@ -38,6 +38,13 @@ pub(super) enum RrdStoreRequest {
         /// Timestamp at which the request was done.
         request_at: i64,
     },
+    /// Store collection stats.
+    CollectionStats {
+        /// Timestamp at which the collection took place.
+        timestamp: i64,
+        /// Statistics.
+        stats: CollectionStats,
+    },
 }
 
 /// Result for a [`RrdStoreRequest`].
@@ -46,6 +53,12 @@ pub(super) struct RrdStoreResult {
     pub(super) most_recent_timestamp: i64,
 }
 
+/// Statistics for a (full) metric collection run.
+pub(super) struct CollectionStats {
+    /// Total time in ms
+    pub(super) total_time: f64,
+}
+
 /// Task which stores received metrics in the RRD. Metric data is fed into
 /// this task via a MPSC channel.
 pub(super) async fn store_in_rrd_task(
@@ -57,7 +70,8 @@ pub(super) async fn store_in_rrd_task(
         //// Involves some blocking file IO
         tokio::task::spawn_blocking(move || {
             let mut most_recent_timestamp = 0;
-            let channel = match msg {
+
+            match msg {
                 RrdStoreRequest::Pve {
                     remote,
                     metrics,
@@ -71,7 +85,13 @@ pub(super) async fn store_in_rrd_task(
                     }
                     store_response_time(&cache_clone, &remote, response_time, request_at);
 
-                    channel
+                    let result = RrdStoreResult {
+                        most_recent_timestamp,
+                    };
+
+                    if let Err(_err) = channel.send(result) {
+                        log::error!("could not send RrdStoreStoreResult to metric collection task");
+                    };
                 }
                 RrdStoreRequest::Pbs {
                     remote,
@@ -86,14 +106,17 @@ pub(super) async fn store_in_rrd_task(
                     }
                     store_response_time(&cache_clone, &remote, response_time, request_at);
 
-                    channel
-                }
-            };
+                    let result = RrdStoreResult {
+                        most_recent_timestamp,
+                    };
 
-            if let Err(_err) = channel.send(RrdStoreResult {
-                most_recent_timestamp,
-            }) {
-                log::error!("could not send RrdStoreStoreResult to metric collection task");
+                    if let Err(_err) = channel.send(result) {
+                        log::error!("could not send RrdStoreStoreResult to metric collection task");
+                    };
+                }
+                RrdStoreRequest::CollectionStats { timestamp, stats } => {
+                    store_stats(&cache_clone, &stats, timestamp)
+                }
             };
         })
         .await?;
@@ -150,6 +173,15 @@ fn store_response_time(cache: &RrdCache, remote_name: &str, response_time: f64,
     cache.update_value(&name, response_time, timestamp, DataSourceType::Gauge);
 }
 
+fn store_stats(cache: &RrdCache, stats: &CollectionStats, timestamp: i64) {
+    cache.update_value(
+        "local/metric-collection/total-time",
+        stats.total_time,
+        timestamp,
+        DataSourceType::Gauge,
+    );
+}
+
 #[cfg(test)]
 mod tests {
     use proxmox_rrd_api_types::{RrdMode, RrdTimeframe};
-- 
2.39.5