TCP stalls in 2.2.0-final and earlier

bryard@usa.net
Thu, 21 Jan 1999 20:34:59 +0100 (MET)


There are strange TCP stalls in recent development kernel, at least from
2.1.132 and up that does not occur on 2.0.x. After a period of some
seconds to several minutes of network activity over a PPP dialup connection
(ISDN "modem" over serial port, 16550A, 115k2) nntp, imap and http
connections begin to stall, while ping continues to work. For example, I see

# netstat -na
Active Internet connections (including servers)
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 42 212.242.4.81:1032 212.242.40.196:119 ESTABLISHED

It just sits there forever. It seems almost certain that the problem is
triggered by disk activity.

# hdparm /dev/hda
/dev/hda:
multcount = 8 (on)
I/O support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 1 (on)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 1229/255/63, sectors = 19746720, start = 0

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/