Repository navigation
Description
Description *
A session on the OpenAI-compatible DashScope lane (https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions) becomes permanently unable to talk once its real input passes the 1,000,000-token input cap. Every turn ends in the same upstream 400 and the retry re-sends the same payload:
- The conversation is a fork that could not inherit compression state (
409 SNAPSHOT_UNAVAILABLE — multimodal or opaque content cannot be compared by text, the multimodal variant of [Bug] Fork of DSH branch clone: inheritance permanently abandoned by 409 SNAPSHOT_UNAVAILABLE — zero-fold clone triggers context_length_exceeded 400 on first turn | [Bug] DSH 分支克隆的 fork 继承被 409 SNAPSHOT_UNAVAILABLE 永久放弃:克隆零折叠,首轮即 context_length_exceeded … #3011), so it restarts from the raw 1,632,489-token parent history instead of the parent's 48 blocks. - preflight starts folding — correct — but the client aborts the stream at ~300s (
summarization aborted: client disconnectedafter 298,929 ms), so only 4 ranges landed: 1,632,489 → 932,640 (estimate). - The next turn hits the billion-context 卡死问题 —— 上报材料 usage获取不到导致上下文估计失误 #2313 probe gate (estimate-only baseline, no current-route evidence) and forwards once to acquire evidence: 2,146 messages / 16.9 MiB.
- DashScope answers 400:
{"error":{"message":"<400> InternalError.Algo.InvalidParameter: Range of input length should be [1, 1000000]","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"},"id":"chatcmpl-...","request_id":"..."}inspectContextOverflow()does not recognize this wording —CONTEXT_OVERFLOW_PATTERNS(src/util.ts:274) has no input-length-range marker andparseOverflowWindow()(src/util.ts:338) has no[1, N]form — so the 400 is not an overflow: nothing arms, nothing shrinks, the 400 is passed through, and the next turn repeats steps 2–4 with the identical payload. The session is a livelock: the user sees a timer and no reply, indefinitely, and each attempt burns the client's stream idle budget.
Expected: the stated range is an overflow with window 1,000,000 → armOverflowShrink arms → the #1195 in-request refold folds below the declared window and the turn recovers. That path already works for the analogous phrasings (prompt is too long: X tokens > Y maximum, maximum context length is N, Gemini maximum number of tokens allowed (N)).
Fix: add the marker /range of input length should be \[/i and the window form /range of input length should be \[\s*\d[\d,]*\s*,\s*(\d[\d,]*)\s*\]/i (the upper bound is the input ceiling). PR follows.
bili version *
0.1.202 (6deb4547)
Usage mode *
Plugin native (dsh) — billion-context 0.1.202 installed in the DSH desktop profile, plus proxy mode for the same upstream.
Client and version *
DSH (deepseek-harness) web UI, session session-283f18fb-e4a6-47d0-b551-7ba655f24641 (parent session-183ec51b-47f5-4d6b-b725-7514d5bf8475).
Operating system *
Windows 11
Upstream and model *
https://dashscope.aliyuncs.com/compatible-mode/v1 · deepseek-v4.1-flash-1017 · declared 1,000,000 tokens (compatible-mode hard input cap). Note the estimate under-read: 932,640 local estimate vs > 1,000,000 real tokens (~12%), which is why preflight believed it was under the cap.
Log excerpt
04:07:03.911 context 1632489 tokens reached preflight target 750000 (usage-baseline=none local-est=1592199)
04:07:33.914 preflight still running after 30000ms grace — committed early SSE headers + keep-alive to hold the client (#568)
04:12:00.575 [preflight] summarization aborted: client disconnected
04:12:02.841 preflight compressed 4 range(s), ~448858 tokens saved (1632489 → 932640) in 298929ms
04:12:05.098 preflight trigger fired on estimate only (~932870 text + ~40290 image vs window 750000); baseline 932640 (estimate) carries no current-route upstream evidence — forwarding once to acquire usage/overflow evidence before compressing (#2313)
04:12:05.104 forward POST → https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions
04:12:18.755 [pck] A/B retry without prompt_cache_key also failed (400) — not the key's fault; key stays, original 400 forwarded (#2645)
04:12:18.757 ← upstream 400 request-id=55bbf071-1040-9261-8cd4-76088d25ee5c: {"error":{"message":"<400> InternalError.Algo.InvalidParameter: Range of input length should be [1, 1000000]","type":"invalid_request_error","param":null,"code":"invalid_parameter_error"} ...}
Related
- [Bug] Fork of DSH branch clone: inheritance permanently abandoned by 409 SNAPSHOT_UNAVAILABLE — zero-fold clone triggers context_length_exceeded 400 on first turn | [Bug] DSH 分支克隆的 fork 继承被 409 SNAPSHOT_UNAVAILABLE 永久放弃:克隆零折叠,首轮即 context_length_exceeded … #3011 — fork inheritance abandoned on a 409 snapshot (this report is the multimodal/opaque variant).
- [Bug] Excessive preflight compression time on ultra-large contexts; client mid-disconnect causes repeated request forwar | [Bug] 超大上下文 preflight 压缩耗时过长,客户端中途断开导致请求反复无法转发 #3978 — very large preflight folding vs the client's stream idle budget (steps 2–3 here).
(ai: deepseek/deepseek-v4.1-flash-1017)
Activity
fix(overflow): recognize the DashScope input-length-range 400 so the …
ranxianglei commentedon Oct 23, 2026
(ai: vllm-qwen/qwen3.8-27b)
查重:不重复(与 #4125 同错误串、不同机制)。 #4125 里也出现了
Range of input length should be [1, 1000000],但那条的根因是插件 bootstrap 多实例端口冲突 → 放弃代理、流量直连上游撞限(bili.log 里根本看不到这些请求);本条 bili 代理链路正常,根因在溢出识别层——同一个 400 到了 bili 手里却不被认成 overflow。失败域不同(实例生命周期 vs 请求管道的错误分类),按「症状+机制都不同才分家」的标准,两者各自成立,互不并入。审核结论(三步已走):
CONTEXT_OVERFLOW_PATTERNS(src/util.ts:270-326)没有任何 input-length-range 标记,parseOverflowWindow()(src/util.ts:338)没有[1, N]形式;DashScope 这个 400 body 逐条过完全部 pattern,无一命中 →inspectContextOverflow返回isOverflow: false→ 不 arm、不 shrink、400 原样透传。日志时间线与此完全吻合(billion-context 卡死问题 —— 上报材料 usage获取不到导致上下文估计失误 #2313 probe forward 一次拿到 400,然后每轮原样重发)。armOverflowShrink(src/server/relay.ts:1044)会按声明窗口 arm 紧急收缩、带 fix: overflow-arm stored above the headroom-reserved judging limit → unbounded CONTEXT_WINDOW_EXCEEDED loop #3738 clamp(arm ≤ 判定上限)、带 billion-context 卡死问题 —— 上报材料 usage获取不到导致上下文估计失误 #2313 origin stamp(让 probe gate 把这次拒绝当作 current-route evidence,不再盲目重发)。缺的只是「这个措辞 = overflow,窗口取上界 N」这一层认识。issue 给的方案打到了核心,不是治标。/range of input length should be \[/i+ 窗口形式/range of input length should be \[\s*\d[\d,]*\s*,\s*(\d[\d,]*)\s*\]/i(上界即输入上限,下界恒为 1 无信息量)。措辞足够特异(短语+方括号锚定),不会误伤限流类 400;marker 命中但窗口解析不出时走既有的无窗 fallback(按被拒 payload 自身大小 arm,context window problem continues #987),与其它无数字措辞行为一致。影响面:DashScope compatible-mode 车道上真实输入越过 1M token 上限的会话,触发后永久 livelock(每轮 preflight→客户端 300s 掐流→probe forward→同一 400→透传),用户看到的是计时器空转、永远没有回复。触发条件特定但一旦触发该会话完全不可用,定 p1。
标签:bug + dsh + scope/wiring + p1(缺陷;DSH 桌面 profile plugin native 车道;溢出识别属请求管道/错误分类;触发后会话完全不可用)。
开始实现:pattern + 窗口解析 + 回归测试(正反向各钉住),随后提 PR 引用本 issue。
Zhishulo commentedon Oct 23, 2026
补充:这条 issue 的最小修法已经实现并在跑了 —— PR #4263(
fix(overflow): recognize the DashScope input-length-range 400 so the probe path can arm,head9a57510fe,rebase 到07888d74)。请优先 review/合并它,避免再开一条同内容的孪生 PR。实现与你的三步结论逐条对应:
src/util.ts→CONTEXT_OVERFLOW_PATTERNS):/range of input length should be \[/i—— 短语 + 方括号锚定,不会误伤限流类或参数越界类 400。parseOverflowWindow):/range of input length should be \[\s*\d[\d,]*\s*,\s*(\d[\d,]*)\s*\]/i→ 取上界 N(下界恒为 1,不构成窗口)。回归(正反向都钉住):
tests/context-overflow.test.ts(+22):真实 body 必须解析出 window = 1000000;参数范围类 400(非输入长度)必须不命中 marker。tests/issue2313-probe-forward.test.ts(+38):e2e#2313 e—— 改前 probe 400 原样透传(会话 livelock),改后 arm 紧急收缩 + 使用pi 压缩后就提示prompt is too long #1195 本轮 refold 恢复。本地验证(Windows / node v24.19.0):
tsc --noEmit -p tsconfig.jsonexit 0;受影响两文件node --test --experimental-test-isolation=none= 41 tests / 41 pass / 0 fail。CI 说明:head 上的 workflow 目前是
completed/cancelled(仓库侧推送风暴取消,非失败;同 head 的e2e-native-dsh已 success),且 master 又前进导致behind。我会在 CI 容量恢复后重新 rebase 到最新 master 并重跑。若你仍要自己实现,请以 #4263 为基线,避免重复劳动。ranxianglei commentedon Oct 23, 2026
(ai: vllm-qwen/qwen3.8-27b)
收到,已确认 PR #4263(head
9a57510fe,单 commit、3 文件 +68 行)存在且内容与你描述一致——没有开孪生 PR:我此前建的本地空分支已删,本 issue 的修复以 #4263 为唯一实现。以下是独立 review 结论(diff 逐行读过,验证是自己在本 head 上重跑的,不是转述你的本地结果)。八问:
Range of input length should be [1, N])不被inspectContextOverflow认成 overflow,紧急收缩从不 arm,billion-context 卡死问题 —— 上报材料 usage获取不到导致上下文估计失误 #2313 probe 每轮原样重发同一超长 payload,会话永久 livelock。[1, N]形式;下游 shrink/arm/probe-origin-stamp 机器本身是好的,缺的只是「这个措辞 = overflow,取上界」这一层认识。根因在识别层而非机制层。/range of input length should be \[/i(短语+方括号锚定)+ 窗口形式取上界 N(下界恒 1 无信息量);marker 命中但解析不出 N 时落既有 context window problem continues #987 按被拒 payload 自身大小 arm 的路径,不新增分支。inspectContextOverflow/parseOverflowWindow的全部既有行为——新标记只精确匹配该短语(我手工把 DashScope body 过了一遍所有既有 pattern,无交叉命中;负向测试钉住temperature should be in range [0, 2]这类参数范围 400 保持暗);窗口形式只在该短语出现时触发,toTokenNumber≥1000 守卫让退化区间落到 fallback。有意行为变更(已披露):此前原样透传的该类 400 现在会 arm 紧急收缩并触发本轮 refold——这正是修复本身,且只发生在真正收到该 400 的会话上。缓存/重试/去重策略不受影响(arm 字段是 per-session stats,origin stamp 沿用 billion-context 卡死问题 —— 上报材料 usage获取不到导致上下文估计失误 #2313/DSH: 切换模型后,preflight 在窗口远未满时提前触发压缩 #1933 F2 语义)。测试有效性(自己复跑的):在 PR head 上两个受影响文件 41/41 全绿、
tsc --noEmit干净;负控制:把src/util.ts单独退回修复前版本,新增的单元用例和e2e #2313 e都红(e2e 正是 livelock 本身:actual 400, expected 200),恢复后全绿——测试钉的是行为不是实现,有牙。判定:代码可合,还差一步。 head
9a57510fe上的 CI 全部 cancelled(推送风暴取消,非失败),当前没有有效的 CI 绿灯;master 已前进到b2ad1b5de(merge base07888d74,GitHub 报 mergeable 无冲突)。等你 rebase 到最新 master 后 CI 跑绿即可合。合并动作按规则只能由人执行:#4263 ——rebase 后 CI 绿了我可以再核一遍 check-runs。fix(overflow): recognize the DashScope input-length-range 400 so the …
fix(overflow): recognize the DashScope input-length-range 400 so the …
ranxianglei commentedon Oct 23, 2026
(ai: vllm-qwen/qwen3.8-27b)
状态更新(这条是同一评论的重复同步,无新指令;我按最新仓库状态核了一遍):
9a57510fe变为0636d3a28(新 merge base8d24bf9d4,master 已前进)。我把新旧两个 commit 的 patch 逐字节比过——完全一致(同样 3 文件 +68 行),纯 rebase、无任何内容改动,所以我上一轮的全部审核结论原样有效。0636d3a28):截至刚才 10 success / 12 queued / 11 skipped / 0 failure——已跑完的全绿,其余还在排队等容量(与之前说的推送风暴后遗症一致)。判定不变:代码可合。 等 head 上 CI 全绿即可由人合并:#4263 (合并动作按规则只能人执行)。CI 全绿后如需我再核一遍 check-runs,说一声即可。
fix(overflow): recognize the DashScope input-length-range 400 so the …
fix(overflow): recognize the DashScope input-length-range 400 so the …
fix(overflow): recognize the DashScope input-length-range 400 so the …
fix(overflow): recognize the DashScope input-length-range 400 so the …
fix(overflow): recognize the DashScope input-length-range 400 so the …
fix(overflow): recognize the DashScope input-length-range 400 so the …
fix(overflow): recognize the DashScope input-length-range 400 so the …
fix(overflow): recognize the DashScope input-length-range 400 so the …
fix(overflow): recognize the DashScope input-length-range 400 so the …
fix(overflow): recognize the DashScope input-length-range 400 so the …