admin/mcp-nextcloud
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(vector): don't inflate qdrant-error metric on embed drops (#893 r3)

Browse Source 
Round-3 review on PR #893:
- record_qdrant_operation("upsert","error") now fires only when the exhausted
  retry was actually a Qdrant failure (reason=="qdrant"); an embed/connection
  failure exhausts retries before Qdrant is called, so attributing it to
  mcp_qdrant_operations_total{error} inflated that signal. The cause is still
  captured by record_ingest_dropped.
- Add test_mistral_embed_retries_on_5xx: exercises the full Mistral retry path
  (5xx SDKError then success), not just the predicate.
- Add test_generate_does_not_retry_on_bad_request: generate() fast-fails on a
  permanent 4xx.
- Move astrolabe_vector_ingest_dropped_total's definition into the astrolabe_
  pipeline-metrics block (was in the mcp_ section).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
Chris Coutinho
2026-06-11 06:17:59 +02:00
co-authored by Claude Opus 4.8
parent 8f7a8432f5
commit c4b6d4a017
4 changed files with 66 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files
nextcloud_mcp_server/observability/metrics.py
+12 -12
View File
@@ -161,18 +161,6 @@ vector_sync_processing_duration_seconds = Histogram(
buckets=(0.1, 0.5, 1.0, 2.5, 5.0, 10.0, 30.0, 60.0),
)
# Documents dropped after exhausting in-process indexing retries (the scanner
# re-picks them on a later full scan, so this is "dropped for this cycle", not
# "lost forever"). Labelled by classified cause so the embed-drop rate from a
# transient backend-pod rollover (connection/timeout) is alertable distinctly
# from a persistent fault (card 309). astrolabe_ prefix: pipeline metric.
vector_ingest_dropped_total = Counter(
"astrolabe_vector_ingest_dropped_total",
"Documents dropped after exhausting indexing retries, by cause",
# reason: connection | timeout | rate_limit | server | qdrant | other
["reason"],
)
vector_sync_queue_size = Gauge(
"mcp_vector_sync_queue_size",
"Current number of documents in processing queue",
@@ -284,6 +272,18 @@ document_parse_failed_total = Counter(
["reason"], # reason: timeout | oom | error
)
# Documents dropped after exhausting in-process indexing retries (the scanner
# re-picks them on a later full scan, so this is "dropped for this cycle", not
# "lost forever"). Labelled by classified cause so the embed-drop rate from a
# transient backend-pod rollover (connection/timeout) is alertable distinctly
# from a persistent fault (card 309).
vector_ingest_dropped_total = Counter(
"astrolabe_vector_ingest_dropped_total",
"Documents dropped after exhausting indexing retries, by cause",
# reason: connection | timeout | rate_limit | server | qdrant | other
["reason"],
)
# --- Tier-0 classifier (shadow mode) -----------------------------------------
#
# The classifier runs a cheap pre-pass per PDF and recommends a starting tier.