all lists on lists.proxmox.com
 help / color / mirror / Atom feed
From: Lukas Wagner <l.wagner@proxmox.com>
To: pdm-devel@lists.proxmox.com
Subject: [pdm-devel] [PATCH proxmox-datacenter-manager v4 14/26] metric collection: record remote response time in metric database
Date: Mon, 12 May 2025 15:37:13 +0200	[thread overview]
Message-ID: <20250512133725.262263-15-l.wagner@proxmox.com> (raw)
In-Reply-To: <20250512133725.262263-1-l.wagner@proxmox.com>

This gives us the ability to retrieve max/avg response times for a given
time window.

Signed-off-by: Lukas Wagner <l.wagner@proxmox.com>
Reviewed-by: Maximiliano Sandoval <m.sandoval@proxmox.com>
---
 .../src/metric_collection/collection_task.rs  | 13 +++++++-
 server/src/metric_collection/rrd_task.rs      | 31 +++++++++++++++++++
 2 files changed, 43 insertions(+), 1 deletion(-)

diff --git a/server/src/metric_collection/collection_task.rs b/server/src/metric_collection/collection_task.rs
index dafeaf3b..53def163 100644
--- a/server/src/metric_collection/collection_task.rs
+++ b/server/src/metric_collection/collection_task.rs
@@ -302,6 +302,7 @@ impl MetricCollectionTask {
         let (result_tx, result_rx) = oneshot::channel();
 
         let now = proxmox_time::epoch_i64();
+        let start = Instant::now();
 
         let res: Result<RrdStoreResult, Error> = async {
             match remote.ty {
@@ -315,11 +316,16 @@ impl MetricCollectionTask {
                         )
                         .await?;
 
+                    let duration = start.elapsed();
+
                     sender
                         .send(RrdStoreRequest::Pve {
                             remote: remote.id.clone(),
                             metrics,
                             channel: result_tx,
+                            // TODO: use as_millis_f64 once stabilized
+                            response_time: duration.as_secs_f64() * 1000.,
+                            request_at: now,
                         })
                         .await?;
                 }
@@ -329,15 +335,20 @@ impl MetricCollectionTask {
                         .metrics(Some(true), Some(status.most_recent_datapoint))
                         .await?;
 
+                    let duration = start.elapsed();
+
                     sender
                         .send(RrdStoreRequest::Pbs {
                             remote: remote.id.clone(),
                             metrics,
                             channel: result_tx,
+                            // TODO: use as_millis_f64 once stabilized
+                            response_time: duration.as_secs_f64() * 1000.,
+                            request_at: now,
                         })
                         .await?;
                 }
-            }
+            };
 
             result_rx.await.map_err(Error::from)
         }
diff --git a/server/src/metric_collection/rrd_task.rs b/server/src/metric_collection/rrd_task.rs
index f65a9291..aa1197a0 100644
--- a/server/src/metric_collection/rrd_task.rs
+++ b/server/src/metric_collection/rrd_task.rs
@@ -20,6 +20,10 @@ pub(super) enum RrdStoreRequest {
         metrics: ClusterMetrics,
         /// Oneshot channel to return the [`RrdStoreResult`].
         channel: oneshot::Sender<RrdStoreResult>,
+        /// Reponse time in ms for the API request.
+        response_time: f64,
+        /// Timestamp at which the request was done (UNIX epoch).
+        request_at: i64,
     },
     /// Store PBS metrics.
     Pbs {
@@ -29,6 +33,10 @@ pub(super) enum RrdStoreRequest {
         metrics: Metrics,
         /// Oneshot channel to return the [`RrdStoreResult`].
         channel: oneshot::Sender<RrdStoreResult>,
+        /// Reponse time in ms for the API request.
+        response_time: f64,
+        /// Timestamp at which the request was done (UNIX epoch).
+        request_at: i64,
     },
 }
 
@@ -54,11 +62,14 @@ pub(super) async fn store_in_rrd_task(
                     remote,
                     metrics,
                     channel,
+                    response_time,
+                    request_at,
                 } => {
                     for data_point in metrics.data {
                         most_recent_timestamp = most_recent_timestamp.max(data_point.timestamp);
                         store_metric_pve(&cache_clone, &remote, &data_point);
                     }
+                    store_response_time(&cache_clone, &remote, response_time, request_at);
 
                     channel
                 }
@@ -66,11 +77,14 @@ pub(super) async fn store_in_rrd_task(
                     remote,
                     metrics,
                     channel,
+                    response_time,
+                    request_at,
                 } => {
                     for data_point in metrics.data {
                         most_recent_timestamp = most_recent_timestamp.max(data_point.timestamp);
                         store_metric_pbs(&cache_clone, &remote, &data_point);
                     }
+                    store_response_time(&cache_clone, &remote, response_time, request_at);
 
                     channel
                 }
@@ -137,6 +151,12 @@ fn store_metric_pbs(cache: &RrdCache, remote_name: &str, data_point: &MetricData
     );
 }
 
+fn store_response_time(cache: &RrdCache, remote_name: &str, response_time: f64, timestamp: i64) {
+    let name = format!("local/metric-collection/remotes/{remote_name}/response-time");
+
+    cache.update_value(&name, response_time, timestamp, DataSourceType::Gauge);
+}
+
 #[cfg(test)]
 mod tests {
     use proxmox_rrd_api_types::{RrdMode, RrdTimeframe};
@@ -199,6 +219,8 @@ mod tests {
             remote: "some-remote".into(),
             metrics,
             channel: tx_back,
+            response_time: 10.0,
+            request_at: now,
         };
 
         // Act
@@ -224,6 +246,15 @@ mod tests {
             assert!(data.data.iter().any(Option::is_some));
         }
 
+        if let Some(data) = cache.extract_data(
+            "local/metric-collection/remotes/some-remote",
+            "response-time",
+            RrdTimeframe::Hour,
+            RrdMode::Max,
+        )? {
+            assert!(data.data.iter().any(Option::is_some));
+        }
+
         Ok(())
     }
 }
-- 
2.39.5



_______________________________________________
pdm-devel mailing list
pdm-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pdm-devel


  parent reply	other threads:[~2025-05-12 13:37 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-12 13:36 [pdm-devel] [PATCH proxmox-datacenter-manager v4 00/26] metric collection improvements (concurrency, config, API, CLI) Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 01/26] pdm-api-types: add CollectionSettings type Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 02/26] pdm-config: add functions for reading/writing metric collection settings Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 03/26] metric collection: split top_entities split into separate module Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 04/26] metric collection: save metric data to RRD in separate task Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 05/26] metric collection: rework metric poll task Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 06/26] metric collection: persist state after metric collection Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 07/26] metric collection: skip if last_collection < MIN_COLLECTION_INTERVAL Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 08/26] metric collection: collect overdue metrics on startup/timer change Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 09/26] metric collection: add tests for the fetch_remotes function Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 10/26] metric collection: add test for fetch_overdue Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 11/26] metric collection: pass rrd cache instance as function parameter Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 12/26] metric collection: add test for rrd task Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 13/26] metric collection: wrap rrd_cache::Cache in a struct Lukas Wagner
2025-05-12 13:37 ` Lukas Wagner [this message]
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 15/26] metric collection: save time needed for collection run to RRD Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 16/26] metric collection: periodically clean removed remotes from statefile Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 17/26] api: add endpoint for updating metric collection settings Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 18/26] api: add endpoint to trigger metric collection Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 19/26] api: remotes: trigger immediate metric collection for newly added nodes Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 20/26] api: add api for querying metric collection RRD data Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 21/26] api: metric-collection: add status endpoint Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 22/26] pdm-client: add metric collection API methods Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 23/26] cli: add commands for metric-collection settings, trigger, status Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 24/26] metric collection: factor out handle_tick and handle_control_message fns Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 25/26] metric collection: skip missed timer ticks Lukas Wagner
2025-05-12 13:37 ` [pdm-devel] [PATCH proxmox-datacenter-manager v4 26/26] metric collection: use JoinSet instead of joining from handles in a Vec Lukas Wagner

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20250512133725.262263-15-l.wagner@proxmox.com \
    --to=l.wagner@proxmox.com \
    --cc=pdm-devel@lists.proxmox.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.
Service provided by Proxmox Server Solutions GmbH | Privacy | Legal