Hardy's netzwerk lahmt

aX0r

Lt. Junior Grade
Registriert
Juni 2005
Beiträge
313
hallo gemeinde,

seit tagen rumgegoogelt und keine lösung in sicht....
das netzwerk in hardy ist extrem langsam, hab kein plan was ich da einstellen soll.
netzwerk ist manuell mit statischer ip,dns,gateway eingerichtet.
downloads sind noch ok, aber der upload..hmm..na ja.
sowohl über vds50 auch über samba (int. netzwerkverb.) ein upload von max 500 kb/s, wobei der download bei ca 3,5 mb/s auch nicht der hit ist.

mfG

unter xp hatte ich traumwerte vdsl: 5,8 mb/s down und 1,2 mb/s up, internes netzwerk bei ca 8-9 mb/s
 
nun mir geht um den datendurchsatz...der fuchs rockt schon gut ab, aber dank dir
 
Wenn ich dich richtig verstanden hab, geht's dir um den Datendurchsatz bei einer Datenübertragung mit Samba (Nautilus), richtig?
Bei ubuntuusers.de hab ich was gefunden, dass Samba über Nautilus wohl nicht die fixeste Variante sein soll... Hast du zufällig die
möglichkeit mal ftp oder ähnliches auszuprobieren?
Ansonsten dachte ich eigentlich, dass sich IPv6 ja eigentlich auch intern auswirken könne müsste... naja, egal :)

mfg asmoday
 
teste mal bitte bei speedtest.net bei ein paar servern und sag mal was da beim upload resp. download rauskommt

es wär sinnvoll wenn du noch die hardware nennen könntest, die sich hier abquält ;)

Code:
sudo update-pciids && sudo update-usbids

dann

Code:
lspci
<-- output posten
 
ahoy @freak01

ich brech wech hier...
306863828.png


PHP:
sudo update-pciids && sudo update-usbids
 % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  130k  100  130k    0     0  66680      0  0:00:02  0:00:02 --:--:-- 95638
Done.
--11:23:01--  http://linux-usb.sourceforge.net/usb.ids
           => `/var/lib/misc/usb.ids.new'
Auflösen des Hostnamen »linux-usb.sourceforge.net«.... 66.35.250.209
Verbindungsaufbau zu linux-usb.sourceforge.net|66.35.250.209|:80... verbunden.
HTTP Anforderung gesendet, warte auf Antwort... 200 OK
Länge: 352.269 (344K) [text/plain]

100%[====================================>] 352.269      192.65K/s             

11:23:04 (192.16 KB/s) - »/var/lib/misc/usb.ids.new« gespeichert [352269/352269]

Done.

PHP:
lspci
00:00.0 Host bridge: Intel Corporation 440BX/ZX/DX - 82443BX/ZX/DX Host bridge (rev 03)
00:01.0 PCI bridge: Intel Corporation 440BX/ZX/DX - 82443BX/ZX/DX AGP bridge (rev 03)
00:07.0 ISA bridge: Intel Corporation 82371AB/EB/MB PIIX4 ISA (rev 02)
00:07.1 IDE interface: Intel Corporation 82371AB/EB/MB PIIX4 IDE (rev 01)
00:07.2 USB Controller: Intel Corporation 82371AB/EB/MB PIIX4 USB (rev 01)
00:07.3 Bridge: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 02)
00:0a.0 Ethernet controller: 3Com Corporation 3c905B 100BaseTX [Cyclone] (rev 30)
00:0c.0 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 61)
00:0c.1 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 61)
00:0c.2 USB Controller: VIA Technologies, Inc. USB 2.0 (rev 63)
00:10.0 Multimedia audio controller: Ensoniq ES1371 [AudioPCI-97] (rev 08)
01:00.0 VGA compatible controller: Matrox Graphics, Inc. MGA G400/G450 (rev 04)

Ethernet controller: 3Com Corporation 3c905B 100BaseTX [Cyclone] (rev 30) hat bis jetzt immer gute dienste geleistet.


ps: eigentlich rockte alles auf mint4.0, seit hardy spackt hier einiges rum.

mfG und dankend,
ax0r
 
ahoy aX0r :)

sodele, hier die lösung:

The reason why some of the "faulty" cards worked "flawlessly" in operating systems such as FreeBSD or Microsoft Windows is because these operating systems have drivers capable of overriding the on board chip configuration (I gather that Windows drivers systematically do that). As far as I am concerned I believe that it is bad engineering to force hardware capabilities as opposed to write drivers that simply exploit these capabilities.

It is possible to go around this problem using a tool called 3c90xCFG.exe [version 5.0] (DOS mode only). This tool allows accessing the on board chip, from there one can modify the registers to activate "auto-negotiation". Since this modification, we no longer experience problem, in fact we have somewhat better performances .

http://ubuntuforums.org/showthread.php?t=192799

das scheint wohl etwas häufiger vorzukommen,

komisch, dass es mit früheren kernels ging :(

google: 3c905B ubuntu cyclone
 
traraaaaa :D
nach der kunst des krieges mit erstellung einer dosbootdisk unter linux, irgenwo im gentoo-forum, voila...puhhhh...ich habe fertig..
war die olle netzwerkkarte doch tatsächlich nicht im fullduplex..

hier die neuen settings
PHP:
;

; NIC Setup file for a 3C90X NIC.

;

; Created 08/09/08, 20:34:22.

;

;



@Boot PROM size:                   128K

@Media type:                       100BASE-TX

@Network driver optimization:      Network

@Full Duplex:                      enabled

System Information:

-------------------

Processor Type					80486

Processor Mode					16-bit Real Mode

Processor Speed					248MHz

Bus Type					ISA

DMA Services Available				NO

EMS Memory Manager Present			NO

EMS Memory Size					0 KBytes

Conventional Memory Size			639 KBytes

Extended Memory Size				0 KBytes

User Interface Version				v1.0q

Mewel Library Version				v3.4

NIC Diagnostic Tests Version		v3.0j

Operating System Version			v7.10







NIC Configuration and Product Information:

----------------------------------------------

Device number:                           14

Bus number:                              0

I/O port address:                        6800h to 683Fh

Interrupt request level:                 5

Media type:                              100BASE-TX

Boot PROM size:                          128K

Boot PROM base address:                  0H

Network speed:                           100M Bit/s

Receive FIFO size:                       2K

Transmit FIFO size:                      2K

ASIC revision:                           ch

Date of manufacture:                     12/20/99

Division code:                           6

Product code:                            XG

Network driver optimization:             Maximized Network Performance

Full duplex:                             Enabled

Remote Wake-Up Connector:                Yes

Remote Wake-Up method:                   Level


dangööööö @freak01

ps: 5,9 mb/s down auf vdsl :evillol:
 
Zurück
Oben