On Fri, 02 Dec 2005 12:39:42 +0200, Daniel Colicov
<[EMAIL PROTECTED]> wrote:
> salut
>
> am un slack 10.1, kernel 2.6.12.5 si 5 eth-uri. la un
trafic mai mare
> aprox. 25MB/s(si aprox 35000pkts/s) in top imi apare
ksoftirqd/0,1,2,3
> depinde pe ce procesor e alocata intreruperea unei placi
de retea.
> procesorul e ocupat cam 35-40%. flood nu se face sigur ca
de la asta mai
> aparea uneori in top ksoftirqd-ul. masina este un dual
xeon 2.8ghz, 1gb
> ram ecc, 2 din placile de retea sunt giga (intel dual
port) restul de
> 100mbps. cu toate astea am ping in server intre 1ms si
15ms. cand
> traficul scade am <1 constant. intrebarea este: chiar asa
incapabil este
> sistemul sa nu duca traficul asta?este vreo smecherie la
kernel sau la
> driverii de placa de retea care trebuie facuta?daca dau
enable la NAPI
> mai rau face. placa de retea duala este pe slot PCI-X de
64 biti.
>
> orice sfat e binevenit.
>
> multam
Poate-i vorba de bug-ul asta (ce a fost reparat in 2.6.13).
Extras din
changelog:
---
commit 7e1f49da6881bbf938e502d99335ad5488eb93b4
Author: Jeff Dike <[EMAIL PROTECTED]>
Date: Thu Jul 28 21:16:09 2005 -0700
[PATCH] uml: Fix load average >=1
update_process_times was missing its
irq_enter/irq_exit wrapper. This
caused
ksoftirqd to be scheduled on every clock tick.
Signed-off-by: Jeff Dike <[EMAIL PROTECTED]>
Cc: Paolo 'Blaisorblade' Giarrusso
<[EMAIL PROTECTED]>
Signed-off-by: Andrew Morton <[EMAIL PROTECTED]>
Signed-off-by: Linus Torvalds <[EMAIL PROTECTED]>
---
Deci incearca sa vezi daca face la fel pe un 2.6.13+
---
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
__________________________________
Start your day with Yahoo! - Make it your home page!
http://www.yahoo.com/r/hs
_______________________________________________
RLUG mailing list
[email protected]
http://lists.lug.ro/mailman/listinfo/rlug
