enhance: add labeled filesystem metrics observability - #592
Open
tedxu wants to merge 1 commit into
Open
Conversation
Collaborator
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: tedxu The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## main #592 +/- ##
==========================================
+ Coverage 75.95% 76.09% +0.13%
==========================================
Files 168 171 +3
Lines 16672 16912 +240
Branches 2510 2570 +60
==========================================
+ Hits 12664 12869 +205
- Misses 4008 4043 +35
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
tedxu
force-pushed
the
enhance/fs-metrics-observability
branch
from
July 27, 2026 13:07
29cca0b to
abe0059
Compare
Replace the flat 13-counter FilesystemMetrics with a labeled registry keyed by operation type and status. Adds per-op latency and payload-size histograms, error classification, retry counts, and saturation gauges (in-flight, connections, pending multipart). A RAII ScopedOp measures one backend request; ScopedXfer adds payload-size recording for Read, Write, and MultipartUploadPart. Local, S3 (sync and CRT async), and Azure paths are instrumented. The FFI exports a fixed-size snapshot with sum, count, and static bucket bounds, plus a legacy shim so the milvus bridge keeps building. Read-only Get* accessors are retained as derived helpers for in-repo consumers with their prior semantics. Adds design and implementation docs and unit tests for the histogram, registry, error classification, and FFI snapshot. Signed-off-by: Ted Xu <ted.xu@zilliz.com>
tedxu
force-pushed
the
enhance/fs-metrics-observability
branch
from
July 28, 2026 02:16
abe0059 to
cc84e9a
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
See #591
Replace the flat 13-counter FilesystemMetrics with a labeled registry
keyed by operation type and status. Adds per-op latency and payload
size histograms, error classification, retry counts, and saturation
gauges (in-flight, connections, pending multipart).
A RAII ScopedOp measures exactly one backend request; ScopedXfer adds
payload-size recording for the transfer ops (Read, Write,
MultipartUploadPart). Local, S3 (sync and CRT async), and Azure paths
are instrumented, with per-backend error classification.
The C FFI exports a fixed-size snapshot carrying sum, count, and
static bucket bounds so a consumer can rebuild Prometheus histograms,
plus a legacy shim so the milvus bridge keeps building until it
migrates. Read-only Get* accessors are retained as helpers derived
from the registry, preserving the prior counters' semantics for
in-repo consumers (benchmarks and filesystem tests).
Design docs/filesystem-metrics-observability-design.md.