Hello!
I have been running some tests with multi-link PPP and have some questions...
First, I am running one machine on the 2.6.9 kernel one machine, and 2.4.27 on another. Both are using pppd 2.4.1b2, patched to work with Digium T1 cards: http://www.datacanyon.com/pub/telephony/zaptel/misc/
I am using this script to start pppd on each machine, calling it with arguments 'unit 2 1' for one interface, and 'unit 2 25' for the other. #!/bin/sh ARGS="-detach noauth debug 172.100.0.2:172.100.0.1 mp lcp-echo-interval 1 lcp-echo-failure 5 persist holdoff 1 plugin zaptel.so" /usr/sbin/pppd $ARGS $*
The other machine has identical arguments, except that the IP addresses are flipped.
In general, this is working.
I have been seeing an occasional problem when testing link failover. To do this test, I start up a UDP traffic stream across the ppp2 interface (bonded over both T1s for about 2700Kbps throughput). Then, I yank both cables and wait for the ppp2 interface go go away, and plug the cables back in.
Normally, the ppp2 interface comes back into existence and I can continue sending traffic over the link...
But, sometimes I see this error: sent [LCP EchoRep id=0x2a magic=0xcb4069f5] sent [LCP EchoReq id=0x2b magic=0xcb4069f5] rcvd [LCP ConfReq id=0x28 <magic 0x636b978e> <mrru 1500> <endpoint [MAC:00:90:27:65:37:8c]>] sent [LCP ConfReq id=0x19 <magic 0xf5f0bc78> <mrru 1500> <endpoint [MAC:00:07:e9:1f:97:c8]>] sent [LCP ConfAck id=0x28 <magic 0x636b978e> <mrru 1500> <endpoint [MAC:00:90:27:65:37:8c]>] rcvd [LCP ConfAck id=0x19 <magic 0xf5f0bc78> <mrru 1500> <endpoint [MAC:00:07:e9:1f:97:c8]>] sent [LCP EchoReq id=0x0 magic=0xf5f0bc78] Couldn't attach to interface unit 2: Inappropriate ioctl for device
Any idea what the problem is?
Thanks, Ben
-- Ben Greear <[EMAIL PROTECTED]> Candela Technologies Inc http://www.candelatech.com
- To unsubscribe from this list: send the line "unsubscribe linux-ppp" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
