Repository navigation
SQLite unified storage degraded to legacy-only (bili v0.1.209, node v22.22.0, win32-x64) #5324
Description
=== billion-context diagnostics ===
generated: 2026-10-31T07:51:12.522Z
bili version: 0.1.209 (commit 39a0ada)
node: v22.22.0
platform: win32 x64
os: Windows_NT 10.0.26300
uptime: 30m 52s
--- storage ---
requested mode: full
effective mode: legacy
degrade reason: index database failed to open (Error: database is locked)
engine: none
engine bootstrap: not-needed
BILI_STORAGE_UNIFIED: (unset)
storage.unified (config file): (unset)
BILI_STORAGE_JSON_WRITER: (unset)
storage.jsonWriter (config file): (unset)
BILI_SQLITE_AUTO_DOWNLOAD: (unset)
storage.sqliteAutoDownload (config file): (unset)
BILI_STORAGE_WAL_JOURNAL_SIZE_LIMIT_BYTES: (unset)
storage.walJournalSizeLimitBytes (config file): (unset)
BILI_ENCRYPTION_KEY set: no
sessions on disk: 5
last [storage] log lines:
2026-10-31T14:27:00.940 [info] [sess=session-25bcfe84-503c-4ea2-a95d-c19dcf394892] [v=0.1.209] [storage] unified import on first open: session-25bcfe84-503c-4ea2-a95d-c19dcf394892
2026-10-31T14:27:04.610 [error] [sess=session-25bcfe84-503c-4ea2-a95d-c19dcf394892] [v=0.1.209] [storage] unified shadow DIGEST DRIFT for session-25bcfe84-503c-4ea2-a95d-c19dcf394892: unified replay != legacy load view [diverged: state] — …
2026-10-31T14:41:47.770 [info] [sess=session-715ae9aa-9358-47af-b7a3-0b8fc3514b51] [v=0.1.209] [storage] unified import on first open: session-715ae9aa-9358-47af-b7a3-0b8fc3514b51
2026-10-31T15:02:34.971 [info] [sess=session-eeb4387a-33b8-4848-b52f-f618e1045012] [v=0.1.209] [storage] unified import on first open: session-eeb4387a-33b8-4848-b52f-f618e1045012
2026-10-31T15:03:01.957 [error] [sess=session-715ae9aa-9358-47af-b7a3-0b8fc3514b51] [v=0.1.209] [storage] unified shadow DIGEST DRIFT for session-715ae9aa-9358-47af-b7a3-0b8fc3514b51: unified replay != legacy load view [diverged: state] — …
2026-10-31T15:03:37.101 [error] [sess=session-eeb4387a-33b8-4848-b52f-f618e1045012] [v=0.1.209] [storage] unified shadow DIGEST DRIFT for session-eeb4387a-33b8-4848-b52f-f618e1045012: unified replay != legacy load view [diverged: state] — …
2026-10-31T15:17:04.654 [debug] [sess=session-6f904a7f-187e-4b86-ba9e-55b955f3df31] [v=0.1.209] [storage] periodic WAL checkpoint at ingest 20: busy=0 log=553 checkpointed=553
2026-10-31T15:20:20.489 [error] [v=0.1.209] [storage] another process (pid 114692) is already writing ~.local\share\billion-context\sessions/index.db — two processes on one index.db corrupt it (#3923/#3959). One session dir must have exac…
2026-10-31T15:20:26.106 [info] [v=0.1.209] [storage] background quick_check started for ~.local\share\billion-context\sessions/index.db (2387.3 MiB over the 512 MiB sync threshold) — serving continues; the verdict lands in /__bili/status …
2026-10-31T15:20:31.642 [warn] [v=0.1.209] [storage] unified sidecar failed to open ~.local\share\billion-context\sessions/index.db (Error: database is locked) — staying legacy-only; delete the file to reset the index
2026-10-31T15:39:16.724 [info] [v=0.1.209] [storage] background quick_check started for ~.local\share\billion-context\sessions/index.db (2387.3 MiB over the 512 MiB sync threshold) — serving continues; the verdict lands in /__bili/status …
2026-10-31T15:39:22.265 [warn] [v=0.1.209] [storage] unified sidecar failed to open ~.local\share\billion-context\sessions/index.db (Error: database is locked) — staying legacy-only; delete the file to reset the index
--- host client versions ---
dsh profile acp: dsh ?, dsh-compaction-basic ?, dsh-web-app ?
dsh profile backups: dsh ?, dsh-compaction-basic ?, dsh-web-app ?
dsh profile desktop: dsh ?, dsh-compaction-basic ?, dsh-web-app ?
dsh profile headless: dsh ?, dsh-compaction-basic ?, dsh-web-app ?
dsh profile web: dsh ?, dsh-compaction-basic ?, dsh-web-app ?
--- bili config (secrets masked) ---
{"providers":{},"compress":{"nudgeGrowthTokens":20000,"maxContextLimit":"25%","emergencyThresholdPercent":"80%","preserveRecentMessages":5,"promptPack":"lean","reasoning":{"drop":false,"threshold":1024},"ccr":{"enabled":true,"minToolTokens":1500},"absorb":{"enabled":true,"mediaOnly":false,"minToolTokens":1500},"outputSteering":{"enabled":true,"verbosityLevel":1}},"sessions":{"residentMaxBytes":2147483648}}
Activity
ranxianglei commentedon Oct 31, 2026
旧版本（v0.1.209）形态的已知问题，新版行为已根本改变，关闭。
这条的上报语义是「静默降级成纯 JSON 模式（legacy-only），事后才发现」。自 2026-10-30 起该行为已被移除（#5213，已并入 master）：索引库打不开时不再悄悄降级——
BILI_STORAGE_UNIFIED=legacy选择纯 JSON 模式（这时是用户的明示选择，不再是意外）。日志里的 DIGEST DRIFT 是双视图对账的报警面（丢水位重灌尾部让其收敛），不是新的数据丢失路径。升级到包含该变更的版本后，本条这类问题要么消失、要么在启动那一刻就把原因摆到你面前。
ranxianglei commentedon Oct 31, 2026
(ai: vllm-qwen/qwen3.8-27b)
收到，同意该判定，已按 master 核实：#5213（merge 7c9e65b）已在 origin/master，后续提交 5159ce9 进一步把「unified-sidecar open 失败」统一改为拒绝启动——v0.1.209 里
shadow.ts的 "staying legacy-only" 静默降级分支在 master 上已不存在。本条上报的语义（locked 环境故障 → 静默落回纯 JSON、事后才发现）在新版下不再成立：占锁方释放后重启即恢复，或用户显式设BILI_STORAGE_UNIFIED=legacy主动选择纯 JSON 模式。agent 侧无动作：不改代码、不另立项（DIGEST DRIFT 按你的定性是双视图对账报警面，非数据丢失路径；index.db 体积问题由既有的
bili storage gc/ opt-in GC 覆盖，不单独立项）。关闭动作留给你执行（agent 无关闭权）。收工判定：本条随 #5213 的行为变更结案，等待包含该版本的升级后如仍复现再 reopen。