OXYGENHUB

/metrics: database pool, pushes and backups (refs #212) a6d824c5ebf5

Adds to /metrics:
- the database pool's connections, idle and in use;
- reference updates pushed in the last 24 hours, accepted and refused
  (one query, working on SQLite and Postgres);
- for each owner's latest backup run, the seconds since it started and
  whether it succeeded.

Test extended, on SQLite and Postgres. Hub spec §13.5.5.

Refs #212
Author
Andy McCall @andymccall <[email protected]>
via
agent agent:claude@andymccall <[email protected]> · keyless, under your grant for changes,refs,seal · key 1b1a8b9d4684
When
· 2026-10-18 03:50 UTC
Signature
◈ sealed by agent:claude@andymccall
Hash
a6d824c5ebf565d6c29e00acb4f646d80640a7810e3eda0c6953099ec4adb9a0
Depends on
42fa95540468 8a48dd830178 93be9ce4653e b9b74d997cec
Depended on by
9e8cbfc3ad00 85ec74fd82d8 2c47eecfeab6
In bookmarks
api-unknown-fields cancel-landed-runs diffusion-pin-bookmark draft-ready main oh-comment-body-file proposal-list-fast
In tags
v0.6.0 v0.6.1 v0.7.0 v0.8.0 v0.9.0 v0.9.1 v0.9.11 v0.9.15 v0.9.16 v0.9.19 v0.9.2 v0.9.20 v0.9.21 v0.9.22 v0.9.23 v0.9.24 v0.9.3 v0.9.4 v0.9.5 v0.9.6 v0.9.7 v0.9.8 v0.9.9
Edits
6 atoms, 0 tree operations, 3 files

Browse the tree at this change

hide whitespace changes
modified crates/hub/src/metrics.rs +49 −0
… 
223 unchanged lines
224 224 
        ],
225 225 
    );
226 226 
    family(&mut out, "oxyhub_tlog_entries", "Entries in the transparency log.", "gauge", &[(String::new(), app.tlog.size() as f64)]);
227 + 
    family(
228 + 
        &mut out,
229 + 
        "oxyhub_db_connections",
230 + 
        "The database pool's connections, by whether they are idle.",
231 + 
        "gauge",
232 + 
        &[
233 + 
            ("idle=\"true\"".into(), app.db.num_idle() as f64),
234 + 
            ("idle=\"false\"".into(), (app.db.size() as f64 - app.db.num_idle() as f64).max(0.0)),
235 + 
        ],
236 + 
    );
237 + 
    let since = db::now_ms() - 86_400_000;
238 + 
    let (ok, refused): (Option<i64>, Option<i64>) =
239 + 
        sqlx::query_as("SELECT SUM(CASE WHEN ok THEN 1 ELSE 0 END), SUM(CASE WHEN ok THEN 0 ELSE 1 END) FROM pushes WHERE at_ms > $1")
240 + 
            .bind(since)
241 + 
            .fetch_one(&app.db)
242 + 
            .await
243 + 
            .unwrap_or((None, None));
244 + 
    family(
245 + 
        &mut out,
246 + 
        "oxyhub_pushes_last_day",
247 + 
        "Reference updates pushed in the last 24 hours, by whether they were accepted.",
248 + 
        "gauge",
249 + 
        &[("result=\"accepted\"".into(), ok.unwrap_or(0) as f64), ("result=\"refused\"".into(), refused.unwrap_or(0) as f64)],
250 + 
    );
251 + 
    let backups: Vec<(String, String, i64, Option<i64>)> = sqlx::query_as(
252 + 
        "SELECT owner, status, started_ms, finished_ms FROM backup_runs WHERE id IN (SELECT MAX(id) FROM backup_runs GROUP BY owner) ORDER BY owner",
253 + 
    )
254 + 
    .fetch_all(&app.db)
255 + 
    .await
256 + 
    .unwrap_or_default();
257 + 
    let now = db::now_ms();
258 + 
    family(
259 + 
        &mut out,
260 + 
        "oxyhub_backup_last_run_age_seconds",
261 + 
        "Seconds since each owner's latest backup run started.",
262 + 
        "gauge",
263 + 
        &backups.iter().map(|(o, _, s, _)| (format!("owner=\"{}\"", label(o)), (now - s) as f64 / 1000.0)).collect::<Vec<_>>(),
264 + 
    );
265 + 
    family(
266 + 
        &mut out,
267 + 
        "oxyhub_backup_last_run_ok",
268 + 
        "Whether each owner's latest finished backup run succeeded (1), was partial or failed (0); running ones are left out.",
269 + 
        "gauge",
270 + 
        &backups
271 + 
            .iter()
272 + 
            .filter(|(_, st, _, f)| f.is_some() && st != "running")
273 + 
            .map(|(o, st, _, _)| (format!("owner=\"{}\"", label(o)), if st == "ok" { 1.0 } else { 0.0 }))
274 + 
            .collect::<Vec<_>>(),
275 + 
    );
227 276 
    http_families(&mut out);
228 277 
    out
229 278 
}
… 
34 unchanged lines
modified crates/hub/tests/flow.rs +4 −0
… 
5303 unchanged lines
5304 5304 
        "oxyhub_flow_jobs{state=\"queued\"} 0",
5305 5305 
        "oxyhub_bucket_requests_total{kind=\"list\"}",
5306 5306 
        "oxyhub_tlog_entries",
5307 + 
        "oxyhub_db_connections{idle=\"true\"}",
5308 + 
        "oxyhub_pushes_last_day{result=\"refused\"} 0",
5309 + 
        "# TYPE oxyhub_backup_last_run_ok gauge",
5307 5310 
    ] {
5308 5311 
        assert!(text.contains(want), "missing {want}:\n{text}");
5309 5312 
    }
… 
8 unchanged lines
5318 5321 
    assert_eq!(runs_of(&h.app, "andy", "demo").await.len(), 1);
5319 5322 
    let text = oxygenhub::metrics::render(&h.app).await;
5320 5323 
    assert!(!text.contains("oxyhub_flow_jobs{state=\"queued\"} 0"), "the run's jobs are queued:\n{text}");
5324 + 
    assert!(text.contains("oxyhub_pushes_last_day{result=\"accepted\"}"), "{text}");
5321 5325 
    // Requests through the router are counted by route class.
5322 5326 
    call(&h.app, "GET", "/andy/demo", None, None).await;
5323 5327 
    let text = oxygenhub::metrics::render(&h.app).await;
… 
188 unchanged lines
modified docs/spec.md +4 −1
… 
643 unchanged lines
644 644 
- requests to bucket stores by kind (with pack listings sent and kept apart), and the time spent in them;
645 645 
- OxygenFlow jobs queued, waiting for approval and running, and runners online and not;
646 646 
- the transparency log's entries;
647 + 
- the database pool's connections (idle and in use);
648 + 
- reference updates pushed in the last 24 hours, accepted and refused;
649 + 
- each owner's latest backup: seconds since it started, and whether it succeeded;
647 650 
- HTTP requests by route class (`page:<tab>`, `api:<area>`, `protocol:<endpoint>`, `static` and so on, never a user's or repository's name; at most 300 classes), method and status class, with latency histograms (5 ms to 30 s).
648 651 

    

  
    

      649
      
      -
      
Pushes and fetches, backups, Diffusion, mail and the database pool are the next slices; the runner's own metrics too.

    

  
    

      
      652
      +
      
Push and fetch sizes and durations, Diffusion, mail and query latency are the next slices; the runner's own metrics too.

    

  
    

      650
      653
       
      

    

  
    

      651
      654
       
      
### 13.6 Coverage reports (oxygenhub#113, done 2026-10-06)

    

  
    

      652
      655
       
      

    

  
    

      
      
      …
      
168 unchanged lines