argus-alm 0.15.9__tar.gz → 0.15.11__tar.gz

argus_alm-0.15.11/.github/workflows/call_jira_sync.yml

@@ -0,0 +1,42 @@
+name: Sync Jira Based on PR Events
+
+on:
+  pull_request_target:
+    types:
+      [opened, ready_for_review, review_requested, labeled, unlabeled, closed]
+
+permissions:
+  contents: read
+  pull-requests: write
+  issues: write
+
+jobs:
+  jira-sync-pr-opened:
+    if: github.event.action == 'opened'
+    uses: scylladb/github-automation/.github/workflows/main_jira_sync_pr_opened.yml@main
+    secrets:
+      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
+
+  jira-sync-in-review:
+    if: github.event.action == 'ready_for_review' || github.event.action == 'review_requested'
+    uses: scylladb/github-automation/.github/workflows/main_jira_sync_in_review.yml@main
+    secrets:
+      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
+
+  jira-sync-add-label:
+    if: github.event.action == 'labeled'
+    uses: scylladb/github-automation/.github/workflows/main_jira_sync_add_label.yml@main
+    secrets:
+      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
+
+  jira-status-remove-label:
+    if: github.event.action == 'unlabeled'
+    uses: scylladb/github-automation/.github/workflows/main_jira_sync_remove_label.yml@main
+    secrets:
+      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
+
+  jira-status-pr-closed:
+    if: github.event.action == 'closed'
+    uses: scylladb/github-automation/.github/workflows/main_jira_sync_pr_closed.yml@main
+    secrets:
+      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}

argus_alm-0.15.11/.github/workflows/close_issue_for_scylla_employee.yml

@@ -0,0 +1,62 @@
+name: Close issues created by Scylla employees
+
+on:
+  issues:
+    types: [opened, reopened]
+
+permissions:
+  issues: write
+
+jobs:
+  comment-and-close:
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Comment and close if author email is scylladb.com
+        uses: actions/github-script@v7
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          script: |
+            const issue = context.payload.issue;
+            const actor = context.actor;
+
+            // Get user data (only public email is available)
+            const { data: user } = await github.rest.users.getByUsername({
+              username: actor,
+            });
+
+            const email = user.email || "";
+            console.log(`Actor: ${actor}, public email: ${email || "<none>"}`);
+
+            // Only continue if email exists and ends with @scylladb.com
+            if (!email || !email.toLowerCase().endsWith("@scylladb.com")) {
+              console.log("User is not a scylladb.com email (or email not public); skipping.");
+              return;
+            }
+
+            const owner = context.repo.owner;
+            const repo = context.repo.repo;
+            const issue_number = issue.number;
+
+            const body = "Issues in this repository are closed automatically. Scylla employees should use Jira to manage issues.\nPlease move this issue to Jira https://scylladb.atlassian.net/jira/software/c/projects/ARGUS/list";
+
+            // Add the comment
+            await github.rest.issues.createComment({
+              owner,
+              repo,
+              issue_number,
+              body,
+            });
+
+            console.log(`Comment added to #${issue_number}`);
+
+            // Close the issue
+            await github.rest.issues.update({
+              owner,
+              repo,
+              issue_number,
+              state: "closed",
+              state_reason: "not_planned"
+            });
+
+            console.log(`Issue #${issue_number} closed.`);

{argus_alm-0.15.9 → argus_alm-0.15.11}/.github/workflows/test.yml

@@ -29,7 +29,7 @@ jobs:
           python-version: '3.12'
 
       - name: Install project dependencies
-        run: uv sync --extra web-backend --extra dev
+        run: uv sync --extra web-backend --extra dev --extra ai
 
       - name: Verify Docker installation
         run: docker --version

{argus_alm-0.15.9 → argus_alm-0.15.11}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: argus-alm
-Version: 0.15.9
+Version: 0.15.11
 Summary: Argus
 Author-email: Alexey Kartashov <alexey.kartashov@scylladb.com>, Łukasz Sójka <lukasz.sojka@scylladb.com>
 License-Expression: Apache-2.0
@@ -11,19 +11,22 @@ Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: requests>=2.26.0
 Requires-Dist: click>=8.1.3
+Requires-Dist: lz4>=4.4.4
 Provides-Extra: web-backend
 Requires-Dist: PyYAML~=6.0.0; extra == "web-backend"
-Requires-Dist: scylla-driver>=3.26.8; extra == "web-backend"
+Requires-Dist: scylla-driver>=3.29.4; extra == "web-backend"
 Requires-Dist: Flask~=3.0.0; extra == "web-backend"
 Requires-Dist: Flask-WTF~=1.0.0; extra == "web-backend"
 Requires-Dist: Flask-Login~=0.5.0; extra == "web-backend"
 Requires-Dist: humanize~=3.13.1; extra == "web-backend"
+Requires-Dist: PyJWT[crypto]>=2.10.0; extra == "web-backend"
 Requires-Dist: python-magic~=0.4.24; extra == "web-backend"
 Requires-Dist: uwsgi~=2.0.20; extra == "web-backend"
 Requires-Dist: python-jenkins>=1.7.0; extra == "web-backend"
 Requires-Dist: python-slugify~=6.1.1; extra == "web-backend"
 Requires-Dist: pygithub>=2.6.1; extra == "web-backend"
 Requires-Dist: boto3~=1.38.9; extra == "web-backend"
+Requires-Dist: jira>=3.10.5; extra == "web-backend"
 Requires-Dist: prometheus-flask-exporter>=0.23.2; extra == "web-backend"
 Provides-Extra: docker-image
 Requires-Dist: supervisor~=4.2.4; extra == "docker-image"
@@ -42,6 +45,8 @@ Requires-Dist: nox~=2025.5.1; extra == "dev"
 Requires-Dist: pytest-xdist~=3.7.0; extra == "dev"
 Requires-Dist: pytest-subtests~=0.14.1; extra == "dev"
 Requires-Dist: boto3-stubs~=1.38.9; extra == "dev"
+Provides-Extra: ai
+Requires-Dist: chromadb>=1.0.15; extra == "ai"
 Dynamic: license-file
 
 # Argus

{argus_alm-0.15.9 → argus_alm-0.15.11}/argus/_version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '0.15.9'
-__version_tuple__ = version_tuple = (0, 15, 9)
+__version__ = version = '0.15.11'
+__version_tuple__ = version_tuple = (0, 15, 11)
 
-__commit_id__ = commit_id = 'g18aee71c2'
+__commit_id__ = commit_id = 'g1da64281a'

{argus_alm-0.15.9 → argus_alm-0.15.11}/argus/client/base.py

@@ -5,6 +5,8 @@ from typing import Any, Type
 from uuid import UUID
 
 import requests
+from requests.adapters import HTTPAdapter
+from urllib3.util.retry import Retry
 
 from argus.common.enums import TestStatus
 from argus.client.generic_result import GenericResultTable
@@ -33,11 +35,30 @@ class ArgusClient:
         FETCH_RESULTS = "/testrun/$type/$id/fetch_results"
         FINALIZE = "/testrun/$type/$id/finalize"
 
-    def __init__(self, auth_token: str, base_url: str, api_version="v1", extra_headers: dict | None = None) -> None:
+    def __init__(self, auth_token: str, base_url: str, api_version="v1", extra_headers: dict | None = None,
+                 timeout: int = 60, max_retries: int = 3) -> None:
         self._auth_token = auth_token
         self._base_url = base_url
         self._api_ver = api_version
+        self._timeout = timeout
         self.session = requests.Session()
+
+        # Configure retry strategy
+        retry_strategy = Retry(
+            total=max_retries,
+            connect=max_retries,
+            read=max_retries,
+            status=0,
+            backoff_factor=1,
+            status_forcelist=(),
+            allowed_methods=["GET"],
+        )
+
+        # Mount adapter with retry strategy for both http and https
+        adapter = HTTPAdapter(max_retries=retry_strategy)
+        self.session.mount("http://", adapter)
+        self.session.mount("https://", adapter)
+
         if extra_headers:
             self.session.headers.update(extra_headers)
 
@@ -101,7 +122,8 @@ class ArgusClient:
         response = self.session.get(
             url=url,
             params=params,
-            headers=self.request_headers
+            headers=self.request_headers,
+            timeout=self._timeout
         )
         LOGGER.debug("GET Response: %s %s", response.status_code, response.url)
 
@@ -124,6 +146,7 @@ class ArgusClient:
             params=params,
             json=body,
             headers=self.request_headers,
+            timeout=self._timeout
         )
         LOGGER.debug("POST Response: %s %s", response.status_code, response.url)
 

{argus_alm-0.15.9 → argus_alm-0.15.11}/argus/client/driver_matrix_tests/client.py

@@ -12,8 +12,10 @@ class ArgusDriverMatrixClient(ArgusClient):
         SUBMIT_DRIVER_FAILURE = "/driver_matrix/result/fail"
         SUBMIT_ENV = "/driver_matrix/env/submit"
 
-    def __init__(self, run_id: UUID, auth_token: str, base_url: str, api_version="v1", extra_headers: dict | None = None) -> None:
-        super().__init__(auth_token, base_url, api_version, extra_headers=extra_headers)
+    def __init__(self, run_id: UUID, auth_token: str, base_url: str, api_version="v1", extra_headers: dict | None = None,
+                 timeout: int = 60, max_retries: int = 3) -> None:
+        super().__init__(auth_token, base_url, api_version, extra_headers=extra_headers,
+                         timeout=timeout, max_retries=max_retries)
         self.run_id = run_id
 
     def submit_driver_matrix_run(self, job_name: str, job_url: str) -> None:

{argus_alm-0.15.9 → argus_alm-0.15.11}/argus/client/generic/client.py

@@ -12,8 +12,10 @@ class ArgusGenericClient(ArgusClient):
     class Routes(ArgusClient.Routes):
         TRIGGER_JOBS = "/planning/plan/trigger"
 
-    def __init__(self, auth_token: str, base_url: str, api_version="v1", extra_headers: dict | None = None) -> None:
-        super().__init__(auth_token, base_url, api_version, extra_headers=extra_headers)
+    def __init__(self, auth_token: str, base_url: str, api_version="v1", extra_headers: dict | None = None,
+                 timeout: int = 180, max_retries: int = 3) -> None:
+        super().__init__(auth_token, base_url, api_version, extra_headers=extra_headers,
+                         timeout=timeout, max_retries=max_retries)
 
     def submit_generic_run(self, build_id: str, run_id: str, started_by: str, build_url: str, sub_type: str = None, scylla_version: str | None = None):
         request_body = {

{argus_alm-0.15.9 → argus_alm-0.15.11}/argus/client/sct/client.py

@@ -28,10 +28,13 @@ class ArgusSCTClient(ArgusClient):
         SUBMIT_EVENTS = "/sct/$id/events/submit"
         SUBMIT_EVENT = "/sct/$id/event/submit"
         SUBMIT_JUNIT_REPORT = "/sct/$id/junit/submit"
+        SUBMIT_STRESS_CMD = "/sct/$id//stress_cmd/submit"
         SUBMIT_EMAIL = "/testrun/report/email"
 
-    def __init__(self, run_id: UUID, auth_token: str, base_url: str, api_version="v1", extra_headers: dict | None = None) -> None:
-        super().__init__(auth_token, base_url, api_version, extra_headers=extra_headers)
+    def __init__(self, run_id: UUID, auth_token: str, base_url: str, api_version="v1", extra_headers: dict | None = None,
+                 timeout: int = 60, max_retries: int = 3) -> None:
+        super().__init__(auth_token, base_url, api_version, extra_headers=extra_headers,
+                         timeout=timeout, max_retries=max_retries)
         self.run_id = run_id
 
     def submit_sct_run(self, job_name: str, job_url: str, started_by: str, commit_id: str,
@@ -124,6 +127,22 @@ class ArgusSCTClient(ArgusClient):
         )
         self.check_response(response)
 
+    def add_stress_command(self, command: str, log_name: str, loader_name: str) -> None:
+        """
+            Submits stress command information to be viewed inside Argus.
+        """
+        response = self.post(
+            endpoint=self.Routes.SUBMIT_STRESS_CMD,
+            location_params={"id": str(self.run_id)},
+            body={
+                **self.generic_body,
+                "cmd": command,
+                "log_name": log_name,
+                "loader_name": loader_name,
+            }
+        )
+        self.check_response(response)
+
     def submit_screenshots(self, screenshot_links: list[str]) -> None:
         """
             Submits links to the screenshots from grafana, taken at the end of the test.
@@ -258,7 +277,8 @@ class ArgusSCTClient(ArgusClient):
         self.check_response(response)
 
     def submit_nemesis(self, name: str, class_name: str, start_time: int,
-                       target_name: str, target_ip: str, target_shards: int) -> None:
+                       target_name: str, target_ip: str, target_shards: int,
+                       description: str | None = None) -> None:
         """
             Submits a nemesis record. Should then be finalized by
             .finalize_nemesis method on nemesis completion.
@@ -275,6 +295,7 @@ class ArgusSCTClient(ArgusClient):
                     "node_name": target_name,
                     "node_ip": target_ip,
                     "node_shards": target_shards,
+                    "description": description,
                 }
             }
         )

{argus_alm-0.15.9 → argus_alm-0.15.11}/argus/client/sirenada/client.py

@@ -45,9 +45,11 @@ class ArgusSirenadaClient(ArgusClient):
         "skipped": "skipped"
     }
 
-    def __init__(self, auth_token: str, base_url: str, api_version="v1", extra_headers: dict | None = None) -> None:
+    def __init__(self, auth_token: str, base_url: str, api_version="v1", extra_headers: dict | None = None,
+                 timeout: int = 60, max_retries: int = 3) -> None:
         self.results_path: Path | None = None
-        super().__init__(auth_token, base_url, api_version, extra_headers=extra_headers)
+        super().__init__(auth_token, base_url, api_version, extra_headers=extra_headers,
+                         timeout=timeout, max_retries=max_retries)
 
     def _verify_required_files_exist(self, results_path: Path):
         assert (results_path / self._junit_xml_filename).exists(), "Missing jUnit XML results file!"

argus_alm-0.15.11/argusAI/README.md

@@ -0,0 +1,149 @@
+# Event Similarity Processor V2
+
+## Overview
+
+This directory contains the Event Similarity Processor V2 — the new system that generates and stores embeddings for SCT events. It supersedes the legacy `event_similarity_processor.py` that worked with the deprecated `sct_run` table, but for now both systems are allowed to run in parallel (see "Co‑existence with V1").
+
+## Architecture
+
+The processor follows a clean, decoupled flow:
+
+1. Event creation → SCT events are submitted via `SCTService.submit_event()`
+2. Queue → ERROR and CRITICAL events are added to `sct_unprocessed_events`
+3. Processing loop → `event_similarity_processor_v2.py` continuously reads from the queue
+4. Embedding generation → Message is sanitized, then embedded via the BGE‑Small‑EN model
+5. Storage → Embeddings are written into severity‑specific tables
+6. Cleanup → Processed events are removed from the queue
+
+## Database Tables
+
+### sct_unprocessed_events
+Tracks events that need embedding generation:
+- `run_id` (UUID, partition key)
+- `severity` (text, clustering key)
+- `ts` (timestamp, clustering key)
+
+### sct_error_event_embedding
+Stores ERROR event embeddings (separate table for performance):
+- `run_id` (UUID, partition key)
+- `ts` (timestamp, clustering key)
+- `embedding` (list<float>)
+
+### sct_critical_event_embedding
+Stores CRITICAL event embeddings (separate table for performance):
+- `run_id` (UUID, partition key)
+- `ts` (timestamp, clustering key)
+- `embedding` (list<float>)
+
+Note: Using separate tables for ERROR and CRITICAL events keeps similarity searches scoped to a single severity class and improves query performance by avoiding mixed‑severity comparisons.
+
+## Running the Processor
+
+### Standalone
+```bash
+cd argusAI
+PYTHONPATH=.. uv run event_similarity_processor_v2.py
+```
+
+### As a Service
+The processor runs in an infinite loop and every ~1s:
+- Reads up to 100 unprocessed events per batch
+- Routes embeddings to the table matching the event severity
+- Handles errors gracefully (logs and continues)
+- Can be stopped with Ctrl+C
+
+## Key Features
+
+### Severity‑specific storage
+- ERROR → `sct_error_event_embedding`
+- CRITICAL → `sct_critical_event_embedding`
+- Faster similarity searches within the same severity
+
+### Message sanitization
+Uses `MessageSanitizer` from `argusAI/utils/event_message_sanitizer.py` to:
+- Remove IPs, URLs, file paths
+- Normalize timestamps and memory addresses
+- Clean up stack traces and backtraces
+- Remove redundant/noisy details
+
+### Embedding model
+`BgeSmallEnEmbeddingModel` (BGE‑Small‑EN‑v1.5):
+- 384‑dimensional embeddings
+- Optimized for semantic similarity search
+- Same model family used by the legacy processor for consistency
+
+### Error handling
+- Event not found → logged and skipped
+- Empty message → logged and skipped
+- Sanitization failure → logged and removed from queue
+- Embedding failure → logged and removed from queue
+- Failures are removed from the queue to avoid infinite retries
+
+## Testing
+
+### Unit tests
+```bash
+PYTHONPATH=. uv run pytest argusAI
+```
+
+Covers initialization, single/batch processing, error handling, and shutdown behavior.
+
+### Integration tests
+```bash
+uv run pytest argus/backend/tests/sct_events/test_event_embedding_integration.py -v
+```
+
+Covers the end‑to‑end flow from event creation to storage, proper routing to severity tables, verifying that non‑ERROR/CRITICAL events don’t create queue entries, multi‑event processing, and error recovery.
+
+## Monitoring & Metrics
+
+Logging:
+- INFO: processing cycles, batch counts, startup/shutdown, table routing
+- DEBUG: per‑event processing, chosen table names
+- WARNING: missing events, empty messages
+- ERROR: processing/database failures
+
+Metrics:
+- `processed_count`: total successfully processed events
+- `error_count`: total errors encountered
+
+## Differences from V1
+
+Improvements in V2:
+1. Cleaner architecture: no Vector DB dependency for processing
+2. Better error handling: failures don’t block processing
+3. Simpler flow: queue → process → store
+4. Severity‑specific tables for similarity performance
+5. Comprehensive unit and integration tests
+
+What’s not included in V2:
+- Similarity search (removed from the processor itself)
+- ChromaDB collections (not required for embedding storage)
+- Historical event backfill (processes only new events)
+
+## Performance Benefits
+
+By separating ERROR and CRITICAL embeddings:
+- Faster similarity searches (smaller, severity‑scoped sets)
+- Better data locality
+- Reduced query complexity (no severity filters inside similarity search)
+- Improved scalability (per‑table tuning)
+
+## Configuration
+
+No dedicated config file is required for the processor itself. It uses the standard Argus configuration for database connectivity (e.g., ScyllaDB settings from your environment or YAML config such as `argus.yaml`/`argus.local.yaml`, depending on your deployment). The embedding model is auto‑downloaded to `~/.cache/chroma/onnx_models/`.
+
+## Co‑existence with V1 (temporary)
+
+- For now, both V1 (`event_similarity_processor.py`) and V2 (`event_similarity_processor_v2.py`) may run in parallel.
+- There is no data migration at this time; V1 will be disabled once the team fully moves to V2.
+- If you still need to run V1 locally for comparison or fallback:
+  ```bash
+  cd argusAI
+  PYTHONPATH=.. uv run event_similarity_processor.py
+  ```
+
+## Notes
+
+- Avoid mixing severities in similarity queries; use the dedicated tables for the best performance.
+- If you encounter repeated failures for the same event, verify the event content after sanitization and check connectivity to the database and model files.