On 2013-08-05 10:08, Sebastian Hölzer wrote: > Hi > > I'm trying to configure a Realtime Master-Slave Network. > I am using linux 2.6.32.11, rtai 3.9.1, rtnet 0.9.13 on two similar > machines. > > If I type in "rtnet start" i get the following output > "rt_e1000e: Failed to initialize MSI-X interrupts. Falling back to MSI > interrupts." > on both the Master and the Slave. > > If i try to Ping the Slave i get the folling output > "ioctl: No route to host" > I can however ping the localhost without a problem. > > If i try to ping the master or the localhost from the slave the machine > prints > "ioctl: No buffer space available" > > If i type dmesg on the master it says > "TDMA: Failed to transmit sync frame!" multiple times while the output on > the slave seems to be ok. > > Please provide suggestions on how i can further narrow down the Problem.
You are likely facing interrupt problems. The NIC is not getting any IRQ through, thus will quickly run out of TX buffers. Does the e1000e driver of Linux fall back to MSI as well? What does RTAI report about registered RT IRQs, is the NIC listed? You may try to force the NIC to use legacy INTx (module parameter IntMode=0), but that may also create IRQ conflicts with Linux devices. Jan -- Siemens AG, Corporate Technology, CT RTC ITP SES-DE Corporate Competence Center Embedded Linux ------------------------------------------------------------------------------ Get your SQL database under version control now! Version control is standard for application code, but databases havent caught up. So what steps can you take to put your SQL databases under version control? Why should you start doing it? Read more to find out. http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk _______________________________________________ RTnet-users mailing list RTnet-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rtnet-users
