I wonder if the BMC internal Linux kernel is generating a HUP on carrier detect changing during the reboot and killing the SOL process inside the BMC.
The plan was to use a console monitoring/aggregation tool like Conman or Conserver so we'll look into implementing serial keepalive there. Is there a non-data packet you could send (like telnet are-you-there) or a duplicate tcp ack that you could send instead of nul bytes that make their way into the input stream? Thanks, Phil P. -- Philip Pokorny, RHCE Chief Hardware Architect PENGUIN COMPUTING, Inc www.penguincomputing.com On Aug 27, 2011, at 9:29 AM, Al Chu <[email protected]> wrote: > Hi Phil, > > I haven't personally witnessed this, but I have heard of this > happening > before. > > There is no direct workaround, since the problem is on the > motherboard. > My assumption is that internally the BMC-UART connection on the > motherboard is reset in some way during a reboot. From the client > perspective, the IPMI connection is fine and dandy. > > Although not available w/ the ipmiconsole tool itself, there is an > indirect workaround in the libipmiconsole library called > SERIAL_KEEPALIVE. As the name suggests, it implements keepalive > messages via serial data instead of IPMI data (internally, it sends > NUL > chars once in awhile). That way, tools like Conman or Conserver can > detect the lost connection quickly and restart it. > > If you're interested, I can add the workaround as an option to > ipmiconsole too. I only left it in libipmiconsole b/c I considered > this > a problem for Conman/Conserver, not for a user using ipmiconsole on > the > command line. > > Al > > On Fri, 2011-08-26 at 20:27 -0700, Phil Pokorny wrote: >> Has anyone else had problems with the BMC and ipmiconsole app >> loosing "sync" >> during a reboot? >> >> The serial console redirect is working fine, then we reboot the >> server and >> after the reset, there is no more serial console data. >> Disconnecting and >> reconnecting to the BMC restores the flow of serial console data. >> >> Is there a work-around switch in FreeIPMI to deal with this >> behavior? Is >> there a "hangup" signal coming from the host that's being ignored by >> ipmiconsole? >> >> System is a Supermicro motherboard with the Winbond BMC and AMI >> firmware. >> >> Thanks, >> _______________________________________________ >> Freeipmi-users mailing list >> [email protected] >> https://lists.gnu.org/mailman/listinfo/freeipmi-users > -- > Albert Chu > [email protected] > Computer Scientist > High Performance Systems Division > Lawrence Livermore National Laboratory > _______________________________________________ Freeipmi-users mailing list [email protected] https://lists.gnu.org/mailman/listinfo/freeipmi-users
