Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!bloom-beacon!SOL.ENGIN.UMICH.EDU!bhoward
From: bhoward@SOL.ENGIN.UMICH.EDU
Newsgroups: comp.windows.x
Subject: Re:  Xerver bug (Waitforsomething(): errno=22)
Message-ID: <8805231345.AA28351@ATHENA.MIT.EDU>
Date: 23 May 88 13:41:00 GMT
Sender: daemon@bloom-beacon.MIT.EDU
Organization: The Internet
Lines: 50


	From cory.berkeley.edu!dheller Sat May 21 04:05:17 1988
	Sender: xpert-request@athena.mit.edu
	To: xpert@athena.mit.edu
	Date: 21 May 88 03:26:41 GMT
	Organization: University of California, Berkeley
	Message-Id: <3577@pasteur.Berkeley.Edu>
	Subject: Xerver bug (Waitforsomething(): errno=22)
	
	 
	/* I posted this once, but it didn't seem to make it */
	 
	Ever since I installed X11 on my sun3-110, I've had this symptom:
	The screen "blinks" (goes black and then returns immediately) and
	an error message appears in my console window:
	 
	waitforsomething(), errno=22
	 
	Normally, I get about 5 or 6 of them and then it goes away.  This
	doesn't bother me, but the last time it happened, the screen remained
	blank and I couldn't do anything.  I logged in from a remote machine
	and did all the appropriate ps's and pstats and so on looking for
	something unusual, but I couldn't find anything.  I tried to run
	some commands (such as xrefresh), but nothing happened.  I had even
	run a couple programs which access the brame buffer directly with no
	avail.  Curious as to how badly the hardware was munged, I did the
	ever-useful L1-A sequence on the console (with the intent to enter
	"c" should it have worked), but that didn't even take.  I didn't
	even lose my connection from the remote host.  Nothing was happening...
	 
	So, I started to play with signals... I sent sigcont, sighup and finally
	sigterm to the server, but nothing happened.  Finally, I sent the KILL
	signal and my system crashed.  Suggestions as to what {sh,c}ould I have done?
	 
	 
	Dan Heller    <island!argv@sun.com>
	
I have seen the same bug.  At first, I thought it was some kind of problem in
the window manager I was using but the problem persisted even after I switched
to twm.  Things became particularly bad this past week, when on three different
occasions, I had to reboot my diskless Sun3/60 to regain control following a
blank-out.

It seems to happen when I start using the node after long periods of inactivity.

				Bruce

				Systems Group
				Computer Aided Engineering Network
				University of Michgian, Ann Arbor, MI.