k2safe/orbit
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

test: capture transaction retry counters in backlog samples

Browse Source
This commit is contained in:
k2safe
2026-09-20 22:18:55 +08:00
parent 351a7757ec
commit 6cf43bafd1
2 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/reports/phase-6-backlog.md
View File

@@ -36,7 +36,7 @@
| 后段 | 98 | 158 | 256 | 32 | | 后段 | 98 | 158 | 256 | 32 |
| 截止前最后点 | 83 | 173 | 256 | 32 | | 截止前最后点 | 83 | 173 | 256 | 32 |
本批原始样本没有开启指标监听，也没有独占 CPU、内存、磁盘或数据库资源；因此已保存曲线只证明持久状态和收尾行为，不证明资源消耗曲线。随后脚本已加入临时指标端口和 `metrics.jsonl` 采样，会在下一次复现中留存服务进程 CPU/RSS、MySQL 容器资源、数据库连接池等待和业务快照计数；新文件仍只是单节点原始采样。目标拓扑容量测试还需要采集锁等待、复制延迟和队列阶段延迟，并进行多轮稳态与恢复对照。 本批原始样本没有开启指标监听，也没有独占 CPU、内存、磁盘或数据库资源；因此已保存曲线只证明持久状态和收尾行为，不证明资源消耗曲线。随后脚本已加入临时指标端口和 `metrics.jsonl` 采样，会在下一次复现中留存服务进程 CPU/RSS、MySQL 容器资源、数据库连接池等待、提交前事务重试/耗尽计数和业务快照计数；新文件仍只是单节点原始采样。目标拓扑容量测试还需要采集锁等待、复制延迟和队列阶段延迟，并进行多轮稳态与恢复对照。
## 环境与复现 ## 环境与复现

2
scripts/test-schedule-backlog.sh
View File

@@ -139,6 +139,8 @@ row = {
"db_connections_idle": values.get("orbit_db_connections_idle", 0), "db_connections_idle": values.get("orbit_db_connections_idle", 0),
"db_wait_count_total": values.get("orbit_db_wait_count_total", 0), "db_wait_count_total": values.get("orbit_db_wait_count_total", 0),
"db_wait_duration_seconds_total": values.get("orbit_db_wait_duration_seconds_total", 0), "db_wait_duration_seconds_total": values.get("orbit_db_wait_duration_seconds_total", 0),
"transaction_retries_total": values.get("orbit_db_transaction_retries_total", 0),
"transaction_retry_exhausted_total": values.get("orbit_db_transaction_retry_exhausted_total", 0),
"waiting_tasks": values.get('orbit_pulse_tasks_state{state="WAITING"}', 0), "waiting_tasks": values.get('orbit_pulse_tasks_state{state="WAITING"}', 0),
"running_tasks": values.get('orbit_pulse_tasks_state{state="RUNNING"}', 0), "running_tasks": values.get('orbit_pulse_tasks_state{state="RUNNING"}', 0),
"business_snapshot_duration_count": values.get("orbit_business_snapshot_duration_seconds_count", 0), "business_snapshot_duration_count": values.get("orbit_business_snapshot_duration_seconds_count", 0),