Re: [Vserver] [Bug] sendfile64 stopped working in host server after upgrading from vanilla

From: Björn Steinbrink <B.Steinbrink_at_gmx.de>
Date: Sat 03 Dec 2005 - 19:25:48 GMT
Message-ID: <20051203192548.GA12851@atjola.homenet>

On 2005.12.03 19:39:21 +0100, Bodo Eggert wrote:
> After I upgraded from vanilla 2.6.11.10 to 2.6.14.2-vs2.1.0-rc7, the
> sendfile function in the host server stopped delivering the whole file.
> After reverting to the old kernel, it works correctly again.
>
> ----------
> $ echo -e 'GET http://be10/images/_______.jpg HTTP/1.0\r\n\r' |
> netcat be10 80 | wc
> 62 247 13032
>
> (The file contains 78835 bytes).
> ----------
> open("/home/_______/public_html/images/_______.jpg",
> O_RDONLY|O_LARGEFILE) = 10
> setsockopt(9, SOL_TCP, TCP_NODELAY, [0], 4) = 0
> setsockopt(9, SOL_TCP, TCP_CORK, [1], 4) = 0
> writev(9, [{"HTTP/1.1 200 OK\r\nDate: Sat, 03 D"..., 284}], 1) = 284
> sendfile64(9, 10, [0], 78835) = -1 EOVERFLOW (Value too large
> for defined data type)
> ----------

Ah! That's what's happening... I just noticed that my local webserver
delivers broken images (i.e. only a part of them), but because I'm busy
with other stuff atm, i didn't care to find out what's happening.

Some details:

Kernel
------
Linux atjola 2.6.14.2-vs2.1.0-rc7 #1 SMP Sun Nov 13 17:58:58 CET 2005 x86_64 AMD Athlon(tm) 64 X2 Dual Core Processor 4400+ AuthenticAMD GNU/Linux

Apache
------
Server version: Apache/2.0.54
Server built: Nov 19 2005 22:05:20

FS

--
All partitions are ext3 and located on a lvm2 volume on software RAID 1.
testme.sh passes without errors.
HTH
Björn
_______________________________________________
Vserver mailing list
Vserver@list.linux-vserver.org
http://list.linux-vserver.org/mailman/listinfo/vserver
Received on Sat Dec 3 19:26:08 2005
[Next/Previous Months] [Main vserver Project Homepage] [Howto Subscribe/Unsubscribe] [Paul Sladen's vserver stuff]
Generated on Sat 03 Dec 2005 - 19:26:12 GMT by hypermail 2.1.8