[etherlab-users] Timeout while waiting for Sdo upload response

[Erwin Burgstaller]

Funny news!

If I run the bus through a Alied Telesyn FS709FC it works:

| Oct  1 11:41:21 UCL00097 kernel: [70829.035263] EtherCAT: Master driver 1.4.0-rc2 stable r1434
| Oct  1 11:41:21 UCL00097 kernel: [70829.035414] EtherCAT: 1 master waiting for devices.
| Oct  1 11:41:22 UCL00097 NetworkManager: <debug info>^I[1222854082.004362] nm_hal_device_removed (): Device removed (hal udi is '/org/freedesktop/Hal/devices/net_00_1b_21_10_f6_c3'
| Oct  1 11:41:22 UCL00097 kernel: [70829.532274] ACPI: PCI interrupt for device 0000:02:03.0 disabled
| Oct  1 11:41:22 UCL00097 kernel: [70829.540084] EtherCAT Intel(R) PRO/1000 Network Driver - version 7.3.20-k2-NAPI
| Oct  1 11:41:22 UCL00097 kernel: [70829.540090] Copyright (c) 1999-2006 Intel Corporation.
| Oct  1 11:41:22 UCL00097 kernel: [70829.540144] ACPI: PCI Interrupt 0000:02:03.0[A] -> Link [LNKA] -> GSI 11 (level, low) -> IRQ 11
| Oct  1 11:41:22 UCL00097 kernel: [70829.803954] ec_e1000: 0000:02:03.0: e1000_probe: (PCI:33MHz:32-bit) 00:1b:21:10:f6:c3
| Oct  1 11:41:22 UCL00097 kernel: [70829.975477] EtherCAT: Accepting device 00:1B:21:10:F6:C3 for master 0.
| Oct  1 11:41:22 UCL00097 kernel: [70829.975923] EtherCAT: Starting master thread.
| Oct  1 11:41:22 UCL00097 kernel: [70829.975939] ec_e1000: ec0: e1000_probe: Intel(R) PRO/1000 Network Connection
| Oct  1 11:41:24 UCL00097 kernel: [70831.971173] ec_e1000: ec0: e1000_watchdog: NIC Link is Up 100 Mbps Full Duplex, Flow Control: RX/TX
| Oct  1 11:41:24 UCL00097 kernel: [70831.971180] EtherCAT: Link state changed to UP.
| Oct  1 11:41:25 UCL00097 kernel: [70831.974776] EtherCAT: 2 slave(s) responding.
| Oct  1 11:41:25 UCL00097 kernel: [70831.974780] EtherCAT: Slave states: PREOP.
| Oct  1 11:41:25 UCL00097 kernel: [70831.974826] EtherCAT: Scanning bus.
| Oct  1 11:41:25 UCL00097 kernel: [70832.254614] EtherCAT: Bus scanning completed in 280 ms.
| Oct  1 11:41:38 UCL00097 kernel: [70845.447966] EtherCAT WARNING: 1 frame CORRUPTED!
| Oct  1 11:41:39 UCL00097 kernel: [70847.001798] EtherCAT WARNING: 1 frame CORRUPTED!


So it seems, there's something to fast, when starting without the
switch. Unfortunately I cannot watch with wireshark without the switch. Hm, can I?

Found a solution: It's a little bit complicated but running wireshark on
the same machine goes like this:

1) Unplug the ethernet cable, so there's no link on the master's side.

2) Run this commands: 

   /etc/init.d/ethercat stop; /etc/init.d/ethercat start; /sbin/ifconfig ecdbgm0 up;

3) Start wireshark and select ecdbgm0 as capture interface, start
   captureing!

4) Plug in the cable, so the master will see the link and will start with
   bringing up the bus.

So I've got two dumps to attach, one with and one without
switch. 

Starting with the switch in between, it worked again. Without switch
there was the sdo timeout.

-- 
Erwin Burgstaller
-------------- next part --------------
A non-text attachment was scrubbed...
Name: with-switch.dump.bz2
Type: application/octet-stream
Size: 159504 bytes
Desc: not available
URL: <http://lists.etherlab.org/pipermail/etherlab-users/attachments/20081001/1544aec6/attachment-0006.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: without-switch.dump.bz2
Type: application/octet-stream
Size: 478473 bytes
Desc: not available
URL: <http://lists.etherlab.org/pipermail/etherlab-users/attachments/20081001/1544aec6/attachment-0007.obj>