[4885] in testers
Wedged X server
daemon@ATHENA.MIT.EDU (Greg Hudson)
Mon Jun 25 09:57:55 2001
Date: Mon, 25 Jun 2001 09:57:51 -0400
Message-Id: <200106251357.JAA05999@egyptian-gods.MIT.EDU>
From: Greg Hudson <ghudson@MIT.EDU>
To: testers@mit.edu
tess-turbo, a sunblade, came up as far as "The system is ready" but
didn't visibly start X. Exploration with a root shell revealed:
* dm and Xsun were running, but not console or xlogin.
* A ton of reactivate process were blocking writing to the console.
(Presumably dm had grabbed console input but, because console
wasn't running, wasn't eating it.)
* The X server process was looping with (from truss):
poll(0xFFBED708, 0, 750) = 0
ioctl(-1, (('k'<<8)|10), 0xFFBEF784) Err#9 EBADF
ioctl(-1, (('k'<<8)|0), 0xFFBEF784) Err#9 EBADF
ioctl(-1, FIONREAD, 0xFFBEF780) Err#9 EBADF
sigprocmask(SIG_SETMASK, 0xFFBEF798, 0x00000000) = 0
ioctl(-1, (('k'<<8)|8), 0xFFBEF818) Err#9 EBADF
close(-1) Err#9 EBADF
fcntl(2, F_GETFL, 0x00000000) = 10
fstat64(2, 0xFFBEF6D8) = 0
fstat64(2, 0xFFBEF738) = 0
fcntl(2, F_SETFL, 0x0000000A) = 0
write(2, " F a i l e d t o s e".., 22) = 22
setcontext(0xFFBEF9E0)
Incurred fault #1, FLTILL %pc = 0x01000000
siginfo: SIGILL ILL_ILLOPC addr=0x01000000
Received signal #4, SIGILL [caught]
siginfo: SIGILL ILL_ILLOPC addr=0x01000000
sigaction(SIGPOLL, 0xFFBEF820, 0x00000000) = 0
sigprocmask(SIG_BLOCK, 0xFFBEF7A8, 0xFFBEF798) = 0
poll(0xFFBED708, 0, 745) = 0
[...]
A stack trace of the process (from pstack):
368: /usr/openwin/bin/Xsun -fp /usr/openwin/lib/X11/fonts/75dpi,/usr/openwi
fed190cc poll (ffbed708, 0, 2ee)
feccc638 select (0, 0, fed3b1bc, fed3b1bc, fed3b1bc, ffbed708) + 2cc
feb42e28 sunChangeKbdTranslation (feb5d0f0, 0, 0, feb5c000, 227bc, feb42110) + 100
feb4213c ddxSUNWkbdProc (220440, feb5d14c, 36050, 0, feb5c000, 5c52c) + 498
0005c558 AbortDDX (0, 1, 0, 0, 0, 0) + 4c
0005e350 sunSignalHandler (4, 0, ffbef9e0, fed38000, 0, 0) + 4
fed1a6dc sigacthandler (4, 0, ffbef9e0, fed38000, 227bc, fd81b890) + 28
--- called from signal handler with signal 4 (SIGILL) ---
01000000 ???????? (197fa0, fd81b874, fe9d8b54, 297b08, 227bc, 136044)
fe94ac84 __glXResetScreens (fe9dbfd8, 0, 0, 0, 25dc78, 14) + 50
0005bd38 CloseDownExtensions (1a, 17b2b0, 68, 3ffffc, 194718, 0) + 3c
000515f0 main (189068, 18bfb4, ffbefeac, 1, 17b374, 0) + 7e8
0005c6e0 _start (0, 0, 0, 0, 0, 0) + b8
Unfortunately, I had already sent a SIGTERM to the Xsun process before
I took those readings, so it may have been in a different state
before.
Killing the Xsun process (with SIGKILL) unwedged things.
