Andrew Clayton

HI @gourav-kandoria So in trying a variation of your patches to try and handle the client closing the connection, I have (hopefully it's all there as I tried to leave...

Summary of issues. With the following python ASGI application ```python async def application(scope, receive, send): while True: m = await receive() if m['type'] == 'http.disconnect': print("Client Disconnect") break await send(...

On Tue, 07 Oct 2025 23:54:45 -0700 leiserfg ***@***.***> wrote: > leiserfg left a comment (nginx/unit#1667) > > Btw, why the images at https://hub.docker.com/_/unit are not up to date? (They...

HI, thanks for the report. > when we make call to a flask app we get a core dump around 3 mins after, > this is present im versions 1.32.2,...

Another quick check would be to see if the problem happens without threads, i.e. comment out `"threads": 4,` (or change it to `1`) in the config.

Thanks for the backtrace and the log output. It actually looks like it's the router process that's crashing and not the application process. Unfortunately as feared you are missing most...

Thanks. ``` #0 nxt_h1p_complete_buffers (task=task@entry=0x7a1904001620, h1p=h1p@entry=0x636b874f99f0, all=all@entry=1) at src/nxt_h1proto.c:1522 size = b = 0x0 in = next = c = 0x0 ``` `c` is `NULL` here which we're not expecting....

Just to confirm again. If you hit the problematic page, then after about 3 minutes, Unit crashes? It's only that one page that causes the crash? If you leave the...

Weird. Are you absolutely sure you tested the patched version? With that patch, in this case, `b` & `in` should both then be `NULL`. We should then just skip past...

Actually, would it be possible for you test current master? There is a _remote_ possibility it may make a difference... (there is some h1proto related changes...)