Re: netem: network hang (2.6.12.3/2.6.13-rc4)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Stephen Hemminger schrieb:
On Thu, 04 Aug 2005 22:15:38 +0200
Markus Rehbach <markus.rehbach@xxxxxx> wrote:


Hi all,

used the very basic command 'tc qdisc add dev eth0
root netem delay 10ms' on my box and 'ssh'ed to another
box. For a little traffic I did a 'find /', after a
short time the output stopped. I had to bring the
interface down and up again to solve the problem
(for a short time only;-).

I saw this behaviour in the kernels 2.6.12.3 and 2.6.13-rc4
on 2 different PCs. 2.6.8(.1) was ok and worked as
espected on both machines.

If more information is needed to track it down please give
me a hint what is necessary.

Cheers

Markus



What is the clock source for PSCHED?
  Ie. CONFIG_NET_SCH_CLK_CPU? CONFIG_NET_SCH_CLK_JIFFIES?
What value of HZ?

In 2.6.11-rc3 I used 250HZ, in all other cases
1000HZ (silly question: what´s better for the
scheduling precision? 1000 I assume...).

Part of .config:
....
CONFIG_NET_SCHED=y
# CONFIG_NET_SCH_CLK_JIFFIES is not set
CONFIG_NET_SCH_CLK_GETTIMEOFDAY=y
# CONFIG_NET_SCH_CLK_CPU is not set
....

Part of dmesg:
....
Using tsc for high-res timesource
....

CONFIG_NET_SCH_CLK_CPU=y solves or hides the
problem. In any case a short test using 'iperf'
was successful. I did not revert it and test it again.

Markus




-
: send the line "unsubscribe linux-net" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[Index of Archives]     [Netdev]     [Ethernet Bridging]     [Linux 802.1Q VLAN]     [Linux Wireless]     [Kernel Newbies]     [Security]     [Linux for Hams]     [Netfilter]     [Git]     [Bugtraq]     [Yosemite News and Information]     [MIPS Linux]     [ARM Linux]     [Linux RAID]     [Linux PCI]     [Linux Admin]     [Samba]

  Powered by Linux