Re: SRP timeout

"M Lowe" <[email protected]> Tue, 22 Jun 2010 23:44:50 -0600
Newsgroups gmane.network.etherboot.user
Message-ID <5DA43AA4A04C014F8833467372C303FC04412D@spireserv.ads.spiretechit.com>
On Tue, Jun 22, 2010 at 4:41AM, Stefan Hajnoczi <[email protected]>
wrote:

> gPXE doesn't notice that the session has been started.  It retransmits
> the login request.

> Building with DEBUG=ib_mi:3 would show the management interface debug
> messages from gPXE.  I wonder if whether gPXE sees the reply but
> discards it or doesn't see the reply at all.

I built with DEBUG=ib_mi:3 and attempted another boot. I couldn't quite
catch all
of the debug output, because it scrolls quite fast. From what I could
catch, 
there was not any packets received, only transmits appeared in the
output I could
read. My motherboard doesn't have a serial port, so that's not an
option. Unless
gPXE supports USB-Serial converters? 

I also tried a build with DEBUG=arbel,srp,ib_srp,ib_mi. The arbel driver
reported
some odd errors that I'm not sure are related. 

Arbel 0x1f7c4 firmware version 1.2.0
Arbel 0x1f7c4 requires 8192 kB for firmware
Arbel 0x1f7c4 command failed with status 22:
000404c8: 00 00 00 00 00 00 00 00-00 00 00 00 00 (rest of line cut off)
000404d8: cf ec 00 00 00 00 0 (rest of line cut off)
Arbel 0x1f7c4 requires 1080kB ICM and 12kB AUX ICM
Arbel 0x1f7c4 EQN 0x0 ring at [0x1f940,0x1f9c0]
Arbel 0x1f7c4 command failed with status 0a:
0004019c: 00 00 00 00 cf eb f0 00-00 00 00 02 00 00 00 00 :
...............
000401ac: cf ec 00 00 00 00 00 00-0a 00 30 24             : .........0$
Arbel 0x1f7c4 could not issue MAD IFC: Input/output error (0x1d714039)

After I ran autoboot, I got the following output after each attempt to
retry the
connection:

Arbel 0x1f7c4 CQN 0x83 ring at [0x21ae0, 0x21be0)
Arbel 0x1f7cr4 QPN 0x550404 send ring at [0x22b60,0x22d60)
arbel 0x1f7c4 qpn 0x550404 receive ring at [0x22d60,0x22de0)
SRP 0x20f34 socket closed: Connection timed out (0x4c206035)

To try and step through the whole process I tried to run gdb (following
the
instructions in the wiki). It seems that running gdbstub halts whatever
thread is
handling the network IO, making it impossible to connect to gdbstub over
udp. 
After exiting gdbstub, gPXE starts responding to pings and arp requests
again. 

To ensure the card itself was not damaged I tried a different card that
is exactly
the same model. The results were identical. I also tried booting over
iSCSI which
worked fine.

Any ideas?

------------------------------------------------------------------------------
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo