Sebastian Stach schreef:
Am 09.06.2012 um 10:51 schrieb Miroslav Lachman:
Hi,
I changed the switch to 1Gbps and run the test again.
No problems with the NICs. The iperf is running for 10 hours now. 2TB of data
was transmitted in both directions.
I am running an endless loop on a client side
while 1
iperf -c xx.xx.xx.xx --format k -m -p 999 -t 1800 -d
sleep 5
end
------------------------------------------------------------
Server listening on TCP port 999
TCP window size: 64.0 KByte (default)
------------------------------------------------------------
------------------------------------------------------------
Client connecting to yy.yy.yy.yy, TCP port 999
TCP window size: 137 KByte (default)
------------------------------------------------------------
[ 5] local xx.xx.xx.xx port 18834 connected with yy.yy.yy.yy port 999
[ 4] local xx.xx.xx.xx port 999 connected with yy.yy.yy.yy port 59754
[ ID] Interval Transfer Bandwidth
[ 4] 0.0-1800.0 sec 82823213 KBytes 376938 Kbits/sec
[ 4] MSS size 1448 bytes (MTU 1500 bytes, ethernet)
[ 5] 0.0-1800.0 sec 73954944 KBytes 336575 Kbits/sec
[ 5] MSS size 1448 bytes (MTU 1500 bytes, ethernet)
And another endless loop on server side
while 1
iperf -s -p 999
end
------------------------------------------------------------
Server listening on TCP port 999
TCP window size: 64.0 KByte (default)
------------------------------------------------------------
[ 4] local yy.yy.yy.yy port 999 connected with xx.xx.xx.xx port 18834
------------------------------------------------------------
Client connecting to xx.xx.xx.xx, TCP port 999
TCP window size: 65.0 KByte (default)
------------------------------------------------------------
[ 6] local yy.yy.yy.yy port 59754 connected with xx.xx.xx.xx port 999
Waiting for server threads to complete. Interrupt again to force quit.
[ ID] Interval Transfer Bandwidth
[ 6] 0.0-1800.0 sec 79.0 GBytes 377 Mbits/sec
[ 4] 0.0-1800.0 sec 70.5 GBytes 337 Mbits/sec
Client is on the Supermicro X9SCA-F
em0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=4219b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_HWCSUM,TSO4,WOL_MAGIC,VLAN_HWTSO>
ether 00:25:90:73:d1:76
inet xx.xx.xx.xx netmask 0xffffff80 broadcast xx.xx.xx.xx
media: Ethernet autoselect (1000baseT <full-duplex>)
status: active
Server is running on the Cisco UCS C200 M2
igb0: flags=8943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=401bb<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,TSO4,VLAN_HWTSO>
ether 50:57:a8:af:eb:0a
inet yy.yy.yy.yy netmask 0xffffff80 broadcast yy.yy.yy.yy
media: Ethernet autoselect (1000baseT <full-duplex>)
status: active
Both sides are running FreeBSD 8.3-RELEASE amd64 GENERIC
So the only difference is that I am using NIC em0 in shared mode for remote
management. Can you try your test with shared mode?
Miroslav Lachman
I did a test on my system again. My second NIC is working without problems
but the first one is still hanging after a while even when configured in shared
mode.
I will contact the support again and if the issue cannot be resolved shortly i
will
RMA the board.
Thanks again for doing the test on your system.
Sebastian Stach
_______________________________________________
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
The support team from supermicro send me a new NIC EEPROM.
After installing, both nics are seen by FreeBSD.
1. This EEPROM package is for X9SCM(-F) and X9SCL(-F).
2. Changed SVID to 15D9 in EEPROM files.
3. For X9SCL-F and X9SCM-F, please use X9SCLF batch file.
4. For X9SCL and X9SCM, please use X9SCL batch file.
So for me the problems are solved.
regards
Johan
_______________________________________________
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"