slow data movement accross lan
Joseph M Bironas
fedora at josebiro.com
Fri Nov 14 19:13:43 UTC 2003
On the Yarrow box, disable kudzu at all run levels. There is a conflict
with the kudzu probe and the 3c59x driver that caused the card to
underperform and report the wrong speed/duplex using mii-tool.
# chkconfig --level 345 kudzu off
worked for me.
Regards,
Joseph
On Fri, 2003-11-14 at 09:37, Harry Putnam wrote:
> Wondering if I've run into something yarrow specific here.
>
> I'm noticing what seem like very slow speeds moving data from a
> running rh 9 box to a yarrow box accross a local network. With ncftp
> running from yarrow to rh9 on a get action I'm seeing a high of 5.65
> KBs (mostly less) being moved. Isn't that very slow?
>
> I have nsf setup between the two as well but a cp -a seems slow too.
> Its way faster than ftp of course but still a timed cp -a of some
> 380868986k across nfs directory looks like:
> (over 1/2 hour for some 380+MB)
> real 32m40.217s
> user 0m0.060s
> sys 0m1.490s
>
> (That seems quite slow too isn't it?)
> So maybe there is more than ethernet speed involved?
>
> Concerning the ethernet speed:
> I've sort of lost track of how to check on my ethernet settings.
> Seems like we once used something like mii-tool.
>
> That returns:
> # mii-tool
> eth0: 10 Mbit, full duplex, link ok
>
> And dmesg shows the device on yarrow:
>
> 3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
> See Documentation/networking/vortex.txt
> 00:0d.0: 3Com PCI 3c905B Cyclone 100baseTx at 0x9000. Vers LK1.1.18-ac
> 00:01:02:ed:56:8a, IRQ 11
> product code 4347 rev 00.12 date 12-06-00
> Internal config register is 1800000, transceivers 0xa.
> 8K byte-wide RAM 5:3 Rx:Tx split, autoselect/Autonegotiate interface.
> MII transceiver found at address 24, status 786d.
> Enabling bus-master transmits and whole-frame receives.
> 00:0d.0: scatter/gather enabled. h/w checksums enabled
> divert: allocating divert_blk for eth0
> divert: freeing divert_blk for eth0
> ip_tables: (C) 2000-2002 Netfilter core team
> PCI: Found IRQ 11 for device 00:0d.0
> PCI: Sharing IRQ 11 with 01:00.0
> 3c59x: Donald Becker and others. www.scyld.com/network/vortex.html
> See Documentation/networking/vortex.txt
> 00:0d.0: 3Com PCI 3c905B Cyclone 100baseTx at 0x9000. Vers LK1.1.18-ac
> 00:01:02:ed:56:8a, IRQ 11
> product code 4347 rev 00.12 date 12-06-00
> Internal config register is 1000000, transceivers 0xa.
> 8K byte-wide RAM 5:3 Rx:Tx split, autoselect/10baseT interface.
> Enabling bus-master transmits and whole-frame receives.
> 00:0d.0: scatter/gather enabled. h/w checksums enabled
> divert: allocating divert_blk for eth0
>
>
> mii-tool on On rh9
>
> eth0: negotiated 100baseTx-HD, link ok
> eth1: no link
>
>
>
> --
> fedora-list mailing list
> fedora-list at redhat.com
> http://www.redhat.com/mailman/listinfo/fedora-list
More information about the fedora-list
mailing list