Good thought, but I did check my MTU - it appears to be solid at 1500
all the way to several test sites.

LAN to DMZ gets 55-60Mbps (Would expect ~100Mbps) 
DMZ to DMZ is wire speed (100Mbps)
DMZ to Internet is 45-60Mbps

The DMZ is a basically the switch connecting the router and firewall.
Everything off WAN interface is running 100MBps FDX, connected to the 1G
Intel card which appears to be happily running at 100Mbps.

WAN
-----
em5: Adapter hardware address = 0xc4ffe948
em5: CTRL = 0x8140248 RCTL = 0x8002
em5: Packet buffer = Tx=20k Rx=12k
em5: Flow control watermarks high = 10240 low = 8740
em5: tx_int_delay = 66, tx_abs_int_delay = 66
em5: rx_int_delay = 0, rx_abs_int_delay = 66
em5: fifo workaround = 0, fifo_reset_count = 0
em5: hw tdh = 174, hw tdt = 174
em5: Num Tx descriptors avail = 256
em5: Tx Descriptors not avail1 = 0
em5: Tx Descriptors not avail2 = 0
em5: Std mbuf failed = 0
em5: Std mbuf cluster failed = 0
em5: Driver dropped packets = 0
em5: Driver tx dma failure in encap = 0
em5: Excessive collisions = 0
em5: Sequence errors = 0
em5: Defer count = 0
em5: Missed Packets = 0
em5: Receive No Buffers = 0
em5: Receive Length Errors = 0
em5: Receive errors = 0
em5: Crc errors = 0
em5: Alignment errors = 0
em5: Carrier extension errors = 0
em5: RX overruns = 0
em5: watchdog timeouts = 0
em5: XON Rcvd = 0
em5: XON Xmtd = 0
em5: XOFF Rcvd = 0
em5: XOFF Xmtd = 0
em5: Good Packets Rcvd = 3240309
em5: Good Packets Xmtd = 5577784

LAN
-----
em4: Adapter hardware address = 0xc4ffa148
em4: CTRL = 0x8140248 RCTL = 0x801a
em4: Packet buffer = Tx=20k Rx=12k
em4: Flow control watermarks high = 10240 low = 8740
em4: tx_int_delay = 66, tx_abs_int_delay = 66
em4: rx_int_delay = 0, rx_abs_int_delay = 66
em4: fifo workaround = 0, fifo_reset_count = 0
em4: hw tdh = 158, hw tdt = 158
em4: Num Tx descriptors avail = 256
em4: Tx Descriptors not avail1 = 0
em4: Tx Descriptors not avail2 = 0
em4: Std mbuf failed = 0
em4: Std mbuf cluster failed = 0
em4: Driver dropped packets = 0
em4: Driver tx dma failure in encap = 0
em4: Excessive collisions = 0
em4: Sequence errors = 0
em4: Defer count = 0
em4: Missed Packets = 0
em4: Receive No Buffers = 0
em4: Receive Length Errors = 0
em4: Receive errors = 0
em4: Crc errors = 0
em4: Alignment errors = 0
em4: Carrier extension errors = 0
em4: RX overruns = 0
em4: watchdog timeouts = 0
em4: XON Rcvd = 0
em4: XON Xmtd = 0
em4: XOFF Rcvd = 0
em4: XOFF Xmtd = 0
em4: Good Packets Rcvd = 4071915
em4: Good Packets Xmtd = 3425928


Ted Crow
Information Technology Manager
Tuttle Services, Inc.

-----Original Message-----
From: Bill Marquette [mailto:[EMAIL PROTECTED] 
Sent: Thursday, July 31, 2008 10:00 AM
To: [email protected]
Subject: Re: [pfSense Support] pfSense 1.2-RELEASE: Performance Issue?

Here's a suggestion somewhat out of left field.  What about MTU?  Any
chance the provider changed it on you?  A machine right on the edge
would handle fragmentation somewhat more gracefully than a firewall
that might decide to drop certain inappropriately fragmented frames.
This would also cause potential slowdown in general.

One thing I didn't see (although I'm likely just missing it), is what
your transfer speeds between DMZ and LAN are.  Also, any chance for a
test, you can remove the router?  And again test LAN to DMZ and LAN to
Internet.  Based on your equipment specs I'm highly skeptical of this
being a hardware capacity issue (a number of us have outperformed your
numbers on _much_ lower end hardware - consider that a Soekris 4801
@266Mhz can easily hit 16Mbit of "normal" traffic, and iperf tests can
get it upwards of 35Mbit).  It might however be a hardware issue.
Also, there are some sysctl's available for troubleshooting the Intel
driver.

Substitute '0' for whichever interface you are trying to debug
sysctl -w dev.em.0.debug=1
sysctl -w dev.em.0.stats=1
The Intel driver will reset these sysctl to their default value on
it's own, it's a one time use type thing.  The results will be
available in dmesg and look like:

<< SNIP >>

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to