On Mon, 2009 Aug 17 21:25+0100, Mark Hindley wrote: > > > > I'll gladly put one together, but what should the behavior be? My > > preliminary patch had it not closing the connection for 404 errors. > > Should that be broadened to all errors (in which case, the whole if- > > conditional can be removed), or are there some errors for which the > > connection should be closed immediately? I've looked through some of > > the HTTP/1.1 RFCs, but I can't find anything addressing this. > > No, I can't either.
I suspect the original intention behind that code was to close a Keep- Alive connection on error, in the context of non-pipelined (singleton or serial) requests. There, the behavior at least makes sense... though I'm still not so sure about RFC compliance. > Could you test the patch below and make sure it fixes what you see. Provided that path_map does not define multiple servers for a given path (thus avoiding the other failure mode), this change makes the difference between seeing "Connection reset by peer" error messages, and a clean run of "apt-get update". It's good. > I have a git repository, but it is on my laptop. I shall investigate > making it public It would be ideal for the apt-cacher repository to live on Alioth or the like. -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
