home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!usc!wupost!darwin.sura.net!haven.umd.edu!mimsy!jsw
- From: jsw@umiacs.umd.edu (Jeff Webber)
- Newsgroups: comp.unix.ultrix
- Subject: Re: Ultrix 4.2 rlogin somewhat broken
- Message-ID: <60286@mimsy.umd.edu>
- Date: 14 Sep 92 01:54:38 GMT
- References: <1992Sep12.070314.23900@jarvis.csri.toronto.edu>
- Sender: news@mimsy.umd.edu
- Organization: UMIACS, University of Maryland, College Park, MD 20742
- Lines: 30
-
- In article <1992Sep12.070314.23900@jarvis.csri.toronto.edu> cks@hawkwind.utcs.toronto.edu (Chris Siebenmann) writes:
- >! $ stty all
- >! new tty, speed 9600 baud , 32 rows, 83 columns; tabs
- >[...]
- >! $ /usr/ucb/rlogin `hostname`
- >[..login chatter deleted deleted..]
- >! : hawkwind.utcs ; stty all
- >! new tty, speed 9600 baud , 0 rows, 0 columns; tabs
- >[...]
- >
- > So much for window size information being propagated around.
-
- I have been investigating this problem. Its related to the bug
- that causes rlogin to hang when going from an Ultrix machine
- to a fast sparc. Its due to a kernel bug in the OOB handling
- stuff in the socket code (soreceive()).
-
- The problem of rlogin hanging is caused by Ultrix sending
- rlogin a SIGURG and then not delivering the OOB packet. The
- rlogin program just eats everything coming across the connection
- waiting for a MSG_OOB packet that never arrives.
-
- The problem is intermittant (race condition or some such) and
- is less severe under 4.2a then 4.2. It however still occurs
- even under 4.2a.
-
- --
- Jeff Webber / U of Maryland Institute for Advanced Computer Studies
- path: uunet!mimsy!jsw phone (301)-405-6746 jsw@umiacs.umd.edu DoD #243
- 1965 Ford Mustang Convertible 1988 Honda Hawk GT 650
-