Re: fou4s update checking problems

On Tue, Dec 07, 2004 at 07:06:30PM +1300, Lars Ellenberg wrote:
> / 2004-12-06 23:31:30 +0100
> \ Kai Schaetzl:
> > Lars, it seems you changed download servers? Problem is the new server is 
> > Apache 2 and the same problem I reported back in April now hits any fou4s 
> > update checks (see Message-ID: <X1zmIpBktph/[email protected]>).
> 
> its Markus' Server. I've nothing to do with it...
> but yes, it seems to run Apache2 now.
> 
> > Getting patch descriptions from 
> > http://ftp.gwdg.de/pub/linux/suse/ftp.suse.com/suse
> > 22:34:17 URL:http://fou4s.gaugusch.at/patches/ [62/62] -> "index.html" [1]
> > 
> > Continued download failed on this file, which conflicts with `-c'.
> > Refusing to truncate existing file `fou4s-47'.
> 
> I still don't know whether this is a problem of wget or apache, or both.
> but my (suse 9.2) wget at least can cope with it.
> the relevant piece from the wget code ([wget-1.9.whatnot]/src/http.c,
> part of the 700 (!wow... guess wget needs a cleanup!) lines
> function gethttp):
> 
>    1650   if (statcode == HTTP_STATUS_RANGE_NOT_SATISFIABLE)
>    1651     {
>    1652       /* If `-c' is in use and the file has been fully downloaded (or
>    1653          the remote file has shrunk), Wget effectively requests bytes
>    1654          after the end of file and the server response with 416.  */
>    1655       logputs (LOG_VERBOSE, _("\
>    1656 \n    The file is already fully retrieved; nothing to do.\n\n"));
>    1657       /* In case the caller inspects. */
>    1658       hs->len = contlen;
>    1659       hs->res = 0;
>    1660       /* Mark as successfully retrieved. */
>    1661       *dt |= RETROKF;
>    1662       xfree_null (type);
>    1663       CLOSE_INVALIDATE (sock);  /* would be CLOSE_FINISH, but there
>    1664                                    might be more bytes in the body. */
>    1665       return RETRUNNEEDED;
>    1666     }

Not only wget, but also curl has been fixed meanwhile (between 9.0 and
9.1, version 7.10.8) to cope with a missing Content-Range header.

rfc2616 states (sections 10.4.17 and 14.16) that the server SHOULD send
a Content-Range header in the 416 case. Thus, apache is not wrong, and
the clients should not get confused if it is missing.

Peter

-- 
_______________________________________________
Fou4s-devel mailing list
[email protected]
https://mail.gaugusch.at/mailman/listinfo/fou4s-devel

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)

iD8DBQFBtet69J3FhL+LCI0RAo0kAJ4hMlOND0PBbgQNMuWzz946csR5wQCgnjZp
TYjTRGazPsryUZpMi7eUPA0=
=DRnd
-----END PGP SIGNATURE-----