[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[no subject]



Monday night and Tuesday morning I noticed several CHAOS GMSGS
servers with an IOC error.  The requesting host was a LispMachine
(today it was LM20; I can't remember who it was last night).  It
got the IOC at CHDUN+5.  the flow of the program near here is:

		SYSCAL FINISH
	CHDUN+4:Send and EOF packet
		 .LOSE
		SYSCAL FINISH

When it tried to send the EOF, the connection was in the LOS
state.  I read in the LOS packet and the message from the LispM
was "No such index exists".  

Is the LispM closing the connection early, or is there a bug in
GMSGS?  GMSGS was recently assembled (two days ago), so was a bug
put in?  There is a seemingly incompatible change at location
LODOL2.  I don't know if this was intended to be a bug fix, or if
it introduced the recent bug with LispMs.