chore(mcp): add server info metadata on client session "root" spans #14828
Conversation
sabrenner
added
the
changelog/no-changelog
|
|
Bootstrap import analysisComparison of import times between this PR and base. SummaryThe average import time from this PR is: 246 ± 3 ms. The average import time from base is: 245 ± 6 ms. The import time difference between this PR and base is: 0.4 ± 0.2 ms. The difference is not statistically significant (z = 2.00). Import time breakdownThe following import paths have grown:
|
Performance SLOsComparing candidate sabrenner/mcp-server-info (9ee52d3) with baseline 3.19 (015c84a) 🟡 Near SLO Breach (3 suites)🟡 flasksimple - 18/18✅ appsec-getTime: ✅ 4.593ms (SLO: <4.750ms -3.3%) vs baseline: +0.4% Memory: ✅ 64.383MB (SLO: <66.500MB -3.2%) vs baseline: +4.9% ✅ appsec-postTime: ✅ 6.637ms (SLO: <6.750ms 🟡 -1.7%) vs baseline: ~same Memory: ✅ 64.372MB (SLO: <66.500MB -3.2%) vs baseline: +5.0% ✅ appsec-telemetryTime: ✅ 4.587ms (SLO: <4.750ms -3.4%) vs baseline: -0.6% Memory: ✅ 64.366MB (SLO: <66.500MB -3.2%) vs baseline: +5.0% ✅ debuggerTime: ✅ 1.857ms (SLO: <2.000ms -7.2%) vs baseline: -0.2% Memory: ✅ 47.813MB (SLO: <49.500MB -3.4%) vs baseline: +4.8% ✅ iast-getTime: ✅ 1.855ms (SLO: <2.000ms -7.2%) vs baseline: -0.6% Memory: ✅ 44.620MB (SLO: <49.000MB -8.9%) vs baseline: +4.8% ✅ profilerTime: ✅ 1.915ms (SLO: <2.100ms -8.8%) vs baseline: -0.6% Memory: ✅ 48.807MB (SLO: <50.000MB -2.4%) vs baseline: +4.8% ✅ resource-renamingTime: ✅ 3.363ms (SLO: <3.650ms -7.9%) vs baseline: ~same Memory: ✅ 54.638MB (SLO: <56.000MB -2.4%) vs baseline: +4.8% ✅ tracerTime: ✅ 3.353ms (SLO: <3.650ms -8.1%) vs baseline: -0.3% Memory: ✅ 54.583MB (SLO: <56.500MB -3.4%) vs baseline: +4.8% ✅ tracer-nativeTime: ✅ 3.350ms (SLO: <3.650ms -8.2%) vs baseline: ~same Memory: ✅ 54.603MB (SLO: <60.000MB -9.0%) vs baseline: +4.9% 🟡 span - 26/26✅ add-eventTime: ✅ 17.410ms (SLO: <22.500ms 📉 -22.6%) vs baseline: +1.0% Memory: ✅ 36.372MB (SLO: <53.000MB 📉 -31.4%) vs baseline: +4.8% ✅ add-metricsTime: ✅ 86.189ms (SLO: <93.500ms -7.8%) vs baseline: -1.4% Memory: ✅ 40.760MB (SLO: <53.000MB 📉 -23.1%) vs baseline: +4.7% ✅ add-tagsTime: ✅ 145.058ms (SLO: <155.000ms -6.4%) vs baseline: +0.3% Memory: ✅ 40.730MB (SLO: <53.000MB 📉 -23.2%) vs baseline: +4.6% ✅ get-contextTime: ✅ 16.068ms (SLO: <20.500ms 📉 -21.6%) vs baseline: -0.7% Memory: ✅ 36.288MB (SLO: <53.000MB 📉 -31.5%) vs baseline: +4.2% ✅ is-recordingTime: ✅ 16.513ms (SLO: <20.500ms 📉 -19.4%) vs baseline: +1.1% Memory: ✅ 36.468MB (SLO: <53.000MB 📉 -31.2%) vs baseline: +5.2% ✅ record-exceptionTime: ✅ 35.541ms (SLO: <40.000ms 📉 -11.1%) vs baseline: +0.7% Memory: ✅ 36.994MB (SLO: <53.000MB 📉 -30.2%) vs baseline: +5.2% ✅ set-statusTime: ✅ 18.237ms (SLO: <22.000ms 📉 -17.1%) vs baseline: +1.4% Memory: ✅ 36.432MB (SLO: <53.000MB 📉 -31.3%) vs baseline: +5.2% ✅ startTime: ✅ 16.278ms (SLO: <20.500ms 📉 -20.6%) vs baseline: +2.5% Memory: ✅ 36.390MB (SLO: <53.000MB 📉 -31.3%) vs baseline: +4.8% ✅ start-finishTime: ✅ 51.731ms (SLO: <52.500ms 🟡 -1.5%) vs baseline: +1.6% Memory: ✅ 34.524MB (SLO: <35.500MB -2.7%) vs baseline: +4.9% ✅ start-finish-telemetryTime: ✅ 52.043ms (SLO: <54.500ms -4.5%) vs baseline: -0.6% Memory: ✅ 34.465MB (SLO: <35.500MB -2.9%) vs baseline: +4.4% ✅ start-finish-traceid128Time: ✅ 53.851ms (SLO: <57.000ms -5.5%) vs baseline: ~same Memory: ✅ 34.465MB (SLO: <35.500MB -2.9%) vs baseline: +5.0% ✅ start-traceid128Time: ✅ 16.346ms (SLO: <22.500ms 📉 -27.3%) vs baseline: -0.3% Memory: ✅ 36.408MB (SLO: <53.000MB 📉 -31.3%) vs baseline: +5.0% ✅ update-nameTime: ✅ 16.347ms (SLO: <22.000ms 📉 -25.7%) vs baseline: -0.9% Memory: ✅ 36.442MB (SLO: <53.000MB 📉 -31.2%) vs baseline: +4.7% 🟡 telemetryaddmetric - 30/30✅ 1-count-metric-1-timesTime: ✅ 3.045µs (SLO: <20.000µs 📉 -84.8%) vs baseline: +1.6% Memory: ✅ 34.426MB (SLO: <35.500MB -3.0%) vs baseline: +4.9% ✅ 1-count-metrics-100-timesTime: ✅ 203.800µs (SLO: <220.000µs -7.4%) vs baseline: -0.6% Memory: ✅ 34.485MB (SLO: <35.500MB -2.9%) vs baseline: +4.7% ✅ 1-distribution-metric-1-timesTime: ✅ 3.623µs (SLO: <20.000µs 📉 -81.9%) vs baseline: +9.6% Memory: ✅ 34.544MB (SLO: <35.500MB -2.7%) vs baseline: +5.0% ✅ 1-distribution-metrics-100-timesTime: ✅ 213.546µs (SLO: <220.000µs -2.9%) vs baseline: -0.9% Memory: ✅ 34.485MB (SLO: <35.500MB -2.9%) vs baseline: +4.9% ✅ 1-gauge-metric-1-timesTime: ✅ 2.173µs (SLO: <20.000µs 📉 -89.1%) vs baseline: -0.5% Memory: ✅ 34.406MB (SLO: <35.500MB -3.1%) vs baseline: +4.7% ✅ 1-gauge-metrics-100-timesTime: ✅ 138.379µs (SLO: <150.000µs -7.7%) vs baseline: +1.1% Memory: ✅ 34.505MB (SLO: <35.500MB -2.8%) vs baseline: +5.0% ✅ 1-rate-metric-1-timesTime: ✅ 3.450µs (SLO: <20.000µs 📉 -82.8%) vs baseline: +9.3% Memory: ✅ 34.505MB (SLO: <35.500MB -2.8%) vs baseline: +5.0% ✅ 1-rate-metrics-100-timesTime: ✅ 217.068µs (SLO: <250.000µs 📉 -13.2%) vs baseline: -0.9% Memory: ✅ 34.367MB (SLO: <35.500MB -3.2%) vs baseline: +4.7% ✅ 100-count-metrics-100-timesTime: ✅ 20.476ms (SLO: <22.000ms -6.9%) vs baseline: +1.4% Memory: ✅ 34.446MB (SLO: <35.500MB -3.0%) vs baseline: +4.8% ✅ 100-distribution-metrics-100-timesTime: ✅ 2.224ms (SLO: <2.300ms -3.3%) vs baseline: -0.3% Memory: ✅ 34.406MB (SLO: <35.500MB -3.1%) vs baseline: +4.8% ✅ 100-gauge-metrics-100-timesTime: ✅ 1.410ms (SLO: <1.550ms -9.1%) vs baseline: ~same Memory: ✅ 34.505MB (SLO: <35.500MB -2.8%) vs baseline: +4.7% ✅ 100-rate-metrics-100-timesTime: ✅ 2.222ms (SLO: <2.550ms 📉 -12.9%) vs baseline: -0.1% Memory: ✅ 34.505MB (SLO: <35.500MB -2.8%) vs baseline: +4.9% ✅ flush-1-metricTime: ✅ 4.451µs (SLO: <20.000µs 📉 -77.7%) vs baseline: -1.3% Memory: ✅ 34.347MB (SLO: <35.500MB -3.2%) vs baseline: +4.5% ✅ flush-100-metricsTime: ✅ 175.659µs (SLO: <250.000µs 📉 -29.7%) vs baseline: -0.5% Memory: ✅ 34.839MB (SLO: <35.500MB 🟡 -1.9%) vs baseline: +4.7% ✅ flush-1000-metricsTime: ✅ 2.126ms (SLO: <2.500ms 📉 -15.0%) vs baseline: ~same Memory: ✅ 35.252MB (SLO: <36.500MB -3.4%) vs baseline: +4.8%
|
…ce-py into sabrenner/mcp-server-info
…ce-py into sabrenner/mcp-server-info
…14828) ## Description Adds server info metadata to llm observability mcp client session "root" spans. This information includes - server name - server version - server title Additionally, adds similar tags to client tool spans: - For both client and server tool calls, mark which kind of tool call they are on a tag (as besides the name, which could vary for manual instrumentation, it is not apparent as they are just normal tool calls otherwise) - adds the mcp server name of the session the client tool call was used for Lastly, generally for the `mcp` integration, we now mark client tool calls as errors when the tool result returned from the mcp server was an error as well. MLOB-4147 ## Testing Add unit tests, and additionally, verified through internal testing against a staging hash in our frontend. ## Risks None
…15306) …#14828) ## Description Adds server info metadata to llm observability mcp client session "root" spans. This information includes - server name - server version - server title Additionally, adds similar tags to client tool spans: - For both client and server tool calls, mark which kind of tool call they are on a tag (as besides the name, which could vary for manual instrumentation, it is not apparent as they are just normal tool calls otherwise) - adds the mcp server name of the session the client tool call was used for Lastly, generally for the `mcp` integration, we now mark client tool calls as errors when the tool result returned from the mcp server was an error as well. MLOB-4147 ## Testing Add unit tests, and additionally, verified through internal testing against a staging hash in our frontend. ## Risks None ## Description <!-- Provide an overview of the change and motivation for the change --> ## Testing <!-- Describe your testing strategy or note what tests are included --> ## Risks <!-- Note any risks associated with this change, or "None" if no risks --> ## Additional Notes <!-- Any other information that would be helpful for reviewers -->
Description
Adds server info metadata to llm observability mcp client session "root" spans. This information includes
Additionally, adds similar tags to client tool spans:
Lastly, generally for the
mcpintegration, we now mark client tool calls as errors when the tool result returned from the mcp server was an error as well.MLOB-4147
Testing
Add unit tests, and additionally, verified through internal testing against a staging hash in our frontend.
Risks
None