Xray
1h 6h 24h 48h 3d 5d 7d 14d 30d ‹
Last 6 hours▾
Ends now. Anything up to this cluster's 30-day retention.
Start and end times
Cancel
Interpreted in UTC (change it at the right of this bar). Earliest available: 2026-10-19 13:17:14.
› →|
updated 1d ago
Sample data. This database was generated by scripts/seed_dev_db.py — every service, deploy, person and figure on this page is synthetic. Names are sample.*@seed.invalid and image tags are prefixed SEED-. Nothing here happened.
← findings

test27 (whole cluster) p95 5.10 s vs a normal 20 ms ± 15 ms for Tue 07:00 UTC (4 of the last 4 checks)

info Latency anomaly test27 resolved impact 22 opened 84d ago · last observed 1d ago · observed 348× · #38 · c121f0e3aadd97f0b394

What happened template

the whole test27 cluster's p95 is 5.1 s on test27, above a normal of 20 ms ± 15 ms for this hour of the week (4 of the last 4 five-minute buckets, z = 341.3, 5 weeks of history). Whether users noticed depends on which endpoints carried it — the service page shows the split. A p95 that stays under a second is never reported — below that nobody notices, whatever the deviation.

p95 far above its hour-of-week normal — and at least 1 s; under a second is noise.

The facts

measureapi_p95_ms
unitms
value5,103
expected_median19.84
expected_mad14.89
z341.29
rel256.16
buckets_hit4
n55
weeks5
slotTue 07:00 UTC
bucket2026-11-17 07:40:00 UTC (1d ago)
floor_ms1,000
discriminatorapi_p95_ms
window2026-11-17 07:25:00 UTC → 2026-11-17 07:45:00 UTC

History

  • 2026-11-15 22:47:29 UTCreopened x-ray · severity=info was=info title=test27 (whole cluster) p95 1.32 s vs a normal 20 ms ± <1 ms for Sun 22:00 UTC (3 of the last 4 checks) level=low was_level=low finding={'id': 38, 'fingerprint': 'c121f0e3aadd97f0b394', 'kind': 'anomaly.latency', 'severity': 'info', 'cluster': 'test27', 'subject': '*', 'title': 'test27 (whole cluster) p95 1.32 s vs a normal 20 ms ± <1 ms for Sun 22:00 UTC (3 of the last 4 checks)', 'facts': {'measure': 'api_p95_ms', 'unit': 'ms', 'value': 1316.6666666666629, 'expected_median': 19.791237, 'expected_mad': 0.5940445, 'z': 2183.1284182694444, 'rel': 65.52776007213005, 'buckets_hit': 3, 'n': 48, 'weeks': 4, 'slot': 'Sun 22:00 UTC', 'bucket': 1794781800, 'window': [1794781200, 1794782400], 'floor_ms': 1000, 'recent_deploys': [], 'discriminator': 'api_p95_ms'}, 'evidence': [{'label': 'Cluster charts', 'href': '/charts?cluster=test27&from=1794759600&to=1794782400'}, {'label': 'Fleet vs normal', 'href': '/fleet?cluster=test27&from=1794759600&to=1794782400'}], 'provenance': {'source': 'rollup', 'collected_at': 1794782847, 'age_seconds': 15, 'stale': False, 'shown': None, 'total': None, 'truncated': False, 'coverage_from': None, 'retain_days': None, 'note': None}, 'impact': 17.5, 'status': 'open', 'first_seen': 1787694095, 'last_seen': 1794782849, 'resolved_at': None, 'acked_by': None, 'acked_at': None, 'snoozed_until': None, 'occurrences': 347, 'narrative': "the whole test27 cluster's p95 is 1.2 s on test27, above a normal of 181 ms ± 239 ms for this hour of the week (3 of the last 4 five-minute buckets, z = 4.5, 4 weeks of history). Whether users noticed depends on which endpoints carried it — the service page shows the split. A p95 that stays under a second is never reported — below that nobody notices, whatever the deviation.", 'narrative_kind': 'template', 'discriminator': 'api_p95_ms', 'level': 'low', 'told_level': None, 'episode_at': 1794782849, 'chat_thread': None, 'kind_label': 'Latency anomaly', 'group': 'Anomalies', 'age_s': 7088766, 'stale_s': 12}
  • 2026-11-15 23:11:07 UTCresolved x-ray · reason=not observed grace_s=900 finding={'id': 38, 'fingerprint': 'c121f0e3aadd97f0b394', 'kind': 'anomaly.latency', 'severity': 'info', 'cluster': 'test27', 'subject': '*', 'title': 'test27 (whole cluster) p95 1.32 s vs a normal 20 ms ± <1 ms for Sun 22:00 UTC (3 of the last 4 checks)', 'facts': {'measure': 'api_p95_ms', 'unit': 'ms', 'value': 1316.6666666666629, 'expected_median': 19.791237, 'expected_mad': 0.5940445, 'z': 2183.1284182694444, 'rel': 65.52776007213005, 'buckets_hit': 3, 'n': 48, 'weeks': 4, 'slot': 'Sun 22:00 UTC', 'bucket': 1794781800, 'window': [1794781200, 1794782400], 'floor_ms': 1000, 'recent_deploys': [], 'discriminator': 'api_p95_ms'}, 'evidence': [{'label': 'Cluster charts', 'href': '/charts?cluster=test27&from=1794759600&to=1794782400'}, {'label': 'Fleet vs normal', 'href': '/fleet?cluster=test27&from=1794759600&to=1794782400'}], 'provenance': {'source': 'rollup', 'collected_at': 1794782847, 'age_seconds': 15, 'stale': False, 'shown': None, 'total': None, 'truncated': False, 'coverage_from': None, 'retain_days': None, 'note': None}, 'impact': 22.5, 'status': 'resolved', 'first_seen': 1787694095, 'last_seen': 1794782849, 'resolved_at': 1794784267, 'acked_by': None, 'acked_at': None, 'snoozed_until': None, 'occurrences': 347, 'narrative': "the whole test27 cluster's p95 is 1.3 s on test27, above a normal of 20 ms ± 1 ms for this hour of the week (3 of the last 4 five-minute buckets, z = 2183.1, 4 weeks of history). Whether users noticed depends on which endpoints carried it — the service page shows the split. A p95 that stays under a second is never reported — below that nobody notices, whatever the deviation.", 'narrative_kind': 'template', 'discriminator': 'api_p95_ms', 'level': 'low', 'told_level': None, 'episode_at': 1794782849, 'chat_thread': None, 'kind_label': 'Latency anomaly', 'group': 'Anomalies', 'age_s': 7090186, 'stale_s': 1432}
  • 2026-11-17 08:04:09 UTCreopened x-ray · severity=info was=info title=test27 (whole cluster) p95 5.10 s vs a normal 20 ms ± 15 ms for Tue 07:00 UTC (4 of the last 4 checks) level=low was_level=low finding={'id': 38, 'fingerprint': 'c121f0e3aadd97f0b394', 'kind': 'anomaly.latency', 'severity': 'info', 'cluster': 'test27', 'subject': '*', 'title': 'test27 (whole cluster) p95 5.10 s vs a normal 20 ms ± 15 ms for Tue 07:00 UTC (4 of the last 4 checks)', 'facts': {'measure': 'api_p95_ms', 'unit': 'ms', 'value': 5102.941176470578, 'expected_median': 19.84375, 'expected_mad': 14.893859, 'z': 341.28813939158266, 'rel': 256.1560907827693, 'buckets_hit': 4, 'n': 55, 'weeks': 5, 'slot': 'Tue 07:00 UTC', 'bucket': 1794901200, 'window': [1794900300, 1794901500], 'floor_ms': 1000, 'recent_deploys': [], 'discriminator': 'api_p95_ms'}, 'evidence': [{'label': 'Cluster charts', 'href': '/charts?cluster=test27&from=1794878700&to=1794901500'}, {'label': 'Fleet vs normal', 'href': '/fleet?cluster=test27&from=1794878700&to=1794901500'}], 'provenance': {'source': 'rollup', 'collected_at': 1794902361, 'age_seconds': 306, 'stale': False, 'shown': None, 'total': None, 'truncated': False, 'coverage_from': None, 'retain_days': None, 'note': None}, 'impact': 17.5, 'status': 'open', 'first_seen': 1787694095, 'last_seen': 1794902649, 'resolved_at': None, 'acked_by': None, 'acked_at': None, 'snoozed_until': None, 'occurrences': 348, 'narrative': "the whole test27 cluster's p95 is 1.3 s on test27, above a normal of 20 ms ± 1 ms for this hour of the week (3 of the last 4 five-minute buckets, z = 2183.1, 4 weeks of history). Whether users noticed depends on which endpoints carried it — the service page shows the split. A p95 that stays under a second is never reported — below that nobody notices, whatever the deviation.", 'narrative_kind': 'template', 'discriminator': 'api_p95_ms', 'level': 'low', 'told_level': None, 'episode_at': 1794902649, 'chat_thread': None, 'kind_label': 'Latency anomaly', 'group': 'Anomalies', 'age_s': 7208572, 'stale_s': 18}
  • 2026-11-17 08:26:41 UTCresolved x-ray · reason=not observed grace_s=900 finding={'id': 38, 'fingerprint': 'c121f0e3aadd97f0b394', 'kind': 'anomaly.latency', 'severity': 'info', 'cluster': 'test27', 'subject': '*', 'title': 'test27 (whole cluster) p95 5.10 s vs a normal 20 ms ± 15 ms for Tue 07:00 UTC (4 of the last 4 checks)', 'facts': {'measure': 'api_p95_ms', 'unit': 'ms', 'value': 5102.941176470578, 'expected_median': 19.84375, 'expected_mad': 14.893859, 'z': 341.28813939158266, 'rel': 256.1560907827693, 'buckets_hit': 4, 'n': 55, 'weeks': 5, 'slot': 'Tue 07:00 UTC', 'bucket': 1794901200, 'window': [1794900300, 1794901500], 'floor_ms': 1000, 'recent_deploys': [], 'discriminator': 'api_p95_ms'}, 'evidence': [{'label': 'Cluster charts', 'href': '/charts?cluster=test27&from=1794878700&to=1794901500'}, {'label': 'Fleet vs normal', 'href': '/fleet?cluster=test27&from=1794878700&to=1794901500'}], 'provenance': {'source': 'rollup', 'collected_at': 1794902361, 'age_seconds': 306, 'stale': False, 'shown': None, 'total': None, 'truncated': False, 'coverage_from': None, 'retain_days': None, 'note': None}, 'impact': 22.5, 'status': 'resolved', 'first_seen': 1787694095, 'last_seen': 1794902649, 'resolved_at': 1794904001, 'acked_by': None, 'acked_at': None, 'snoozed_until': None, 'occurrences': 348, 'narrative': "the whole test27 cluster's p95 is 5.1 s on test27, above a normal of 20 ms ± 15 ms for this hour of the week (4 of the last 4 five-minute buckets, z = 341.3, 5 weeks of history). Whether users noticed depends on which endpoints carried it — the service page shows the split. A p95 that stays under a second is never reported — below that nobody notices, whatever the deviation.", 'narrative_kind': 'template', 'discriminator': 'api_p95_ms', 'level': 'low', 'told_level': None, 'episode_at': 1794902649, 'chat_thread': None, 'kind_label': 'Latency anomaly', 'group': 'Anomalies', 'age_s': 7209926, 'stale_s': 1372}

Check it yourself

Every link carries the finding's own time window, so it shows tomorrow what it shows now.

Actions

Sign in (IAP) to acknowledge, snooze or resolve.

Where the numbers come from

sourcerollup
collected_at1794902361
age_seconds306
staleFalse
shownNone
totalNone
truncatedFalse
coverage_fromNone
retain_daysNone
noteNone
detectorseasonal