koomen@CS.ROCHESTER.EDU (10/21/87)
System: Lyric on 1108 File guess: TCPLLICMP of 8-Jan-87 16:44:25 Folks, every now and then I get an unbound var break in the \10MBWATCHER process. BT shows that \HANDLE.RAW.IP is calling the function \ICMP.TIME.EXCEEDED (params: PACKET and CODE) with localvar IP and freevar \ICMP.TRANSIT.TIME.EXCEEDED as arguments. The latter turns out to be unbound. Does anyone know why it's unbound, and what the value ought to be? -- Hans ps. I did (APROPOS "EXCEEDED" T) in case \ICMP.TRANSIT.TIME.EXCEEDED was a misspelling. This turned up \ICMP.FRAGMENT.TIME.EXCEEDED, but that one is unbound as well.
darrelj@sm.unisys.COM (10/21/87)
by SUMEX-AIM.STANFORD.EDU with TCP; Tue, 20 Oct 87 18:51:28 PDT Received: by cayuga.cs.rochester.edu (5.52/e) id AA24932; Tue, 20 Oct 87 21:50:30 EDT AA03087; Tue, 20 Oct 87 21:50:25 EDT Message-Id: <8710210150.AA03087@cursa.cs.rochester.edu> To: info-1100@sumex-aim.stanford.edu Cc: koomen@cs.rochester.edu Subject: TCP/IP: unbound var Date: Tue, 20 Oct 87 21:50:20 -0400 From: koomen@cs.rochester.edu Status: RO System: Lyric on 1108 File guess: TCPLLICMP of 8-Jan-87 16:44:25 Folks, every now and then I get an unbound var break in the \10MBWATCHER process. BT shows that \HANDLE.RAW.IP is calling the function \ICMP.TIME.EXCEEDED (params: PACKET and CODE) with localvar IP and freevar \ICMP.TRANSIT.TIME.EXCEEDED as arguments. The latter turns out to be unbound. Does anyone know why it's unbound, and what the value ought to be? -- Hans I reported this one to Xerox last month. In Koto, the corresponding code had 0 compiled inline, so presumably a CONSTANT declaration was missing at last compilation. Workaround is (SETQ \ICMP.TIME.EXCEEDED 0). DARREL
AISupport.pasa@XEROX.COM (10/23/87)
AR #8897: "TCPLLIP'S \HANDLE.RAW.IP freely references unbound \ICMP.TRANSIT.TIME.EXCEEDED and \ICMP.FRAGMENT.TIME.EXCEEDED." was previously submitted on this problem. The workaround is to bind these variables to 0 and 1 respectively at top level. Judy Dering AISupport