[3134] in Release_Engineering
NetProb ticket #914, RISC/6000: rlogin -x and CTRL-C
daemon@ATHENA.MIT.EDU (probe@MIT.EDU)
Fri Nov 26 16:07:35 1993
From: probe@MIT.EDU
Date: Fri, 26 Nov 1993 16:07:25 -0500
To: rel-eng@MIT.EDU
Cc:
For the records; it is in chain #211, owned by me (in release-wait)
-Richard
>From probe@MIT.EDU Sat Nov 20 00:45:57 1993
>Received: by gregor.MIT.EDU (5.57/4.7) id AA02539; Sat, 20 Nov 93 00:45:56 -05>00
>Received: from TARDIS.MIT.EDU by MIT.EDU with SMTP
> id AA11652; Sat, 20 Nov 93 00:45:47 EST
>Received: by tardis.MIT.EDU (AIX 3.2/UCB 5.64/4.7) id AA21568; Sat, 20 Nov 199>3 00:45:46 -0500
>Date: Sat, 20 Nov 1993 00:45:46 -0500
>Message-Id: <9311200545.AA21568@tardis.MIT.EDU>
>To: dcns-dev-np@MIT.EDU
>Subject: RISC/6000: rlogin -x and CTRL-C
>From: "Richard Basch" <basch@MIT.EDU>
>
>
>Client: RISC/6000
>Release: *ALL* (through 7.6)
>
>It is sometimes the case that when you hit CTRL-C during an encrypted
>rlogin session while another program is being executed and scrolling
>text that the session is terminated.
>
>This is apparently fixed with the Kerberos p10 stuff that I have been
>working on. The suspected problem is that the out-of-band data is being
>inserted where a length field is expected, creating a size field (four
>bytes) where the high-order byte is non-zero. In turn, it will try to
>malloc *LOTS* of memory and fail.
>
>This is related to the Sun rlogin problems, and ticket chain #211, which
>I own (because vrt was about to discard it even though I had a fix).
>
>This bug report is mainly for the files, and an indication that there is
>a more generic problem than just with the Sun Solaris workstations.
>
>-Richard
>
