This is the mail archive of the cygwin-xfree@cygwin.com mailing list for the Cygwin XFree86 project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Getting multiple lines "Xlib: unexpected async reply (sequence0xAdress)!" when runnning startx


Alexander Gottwald writes:

> Vincenzo Daniele wrote:
>
>> Xlib: unexpected async reply (sequence 0x1921)!
>> Xlib: unexpected async reply (sequence 0x1923)!
>> XIO:  fatal IO error 0 (No error) on X server ":0.0"
>>       after 111 requests (6436 known processed) with 0 events remaining.
>> XIO:  fatal IO error 0 (No error) on X server ":0.0"
>>       after 7 requests (6 known processed) with 0 events remaining.
>
> This should never happen unless you have some network troubles.

I know the first two of these error messages, but not the XIO sort.
Some additional samples (not necessarily in the correct order):
------------------------------------------------------------
X protocol error: BadIDChoice (invalid resource ID chosen for this
connection) on protocol request 1

Connection lost to X server `127.0.0.1:0.0'

Xlib: sequence lost (0x6000f > 0x60009) in reply type 0x2!
------------------------------------------------------------

In addition, my Xwin.log is clobbered with lines like the
following:

------------------------------------------------------------
winTopLevelWindowProc - BitBlt failed: Das Handle ist ungültig.
(some 1000 times per session)

winMultiWindowXMsgProcErrorHandler - ERROR: BadWindow (invalid Window parameter)

winMultiWindowGetWMHints - pWin or hints was NULL
------------------------------------------------------------

Actually, I don't care too much because everything works just fine.
It is quite surprising how far I can go after the 'sequence lost'
messages, but usually I'm using lunchtime (or another break) to
restart X, just for safety.

But here and then (about once a week, probably less), after having one
or two of these "unexpected async reply" messages, emacs crashes.

I haven't been able to make that error reproducible, but here are the
common characteristics:

 - This is all on localhost, no personal firewall installed.

 - It happened with Windows2000, and happens with WindowsXP.

 - Command line is:
   XWin -clipboard -multiwindow -xkbmodel pc105 -xkblayout de -xkbvariant nodeadkeys

 - I'm using the X selection/clipboard quite frequently to exchange
   data between windows and cygwin applications.

 - it's always emacs which crashes (sometimes with "exit 70",
   sometimes segfault).  I'm usually having half a dozen emacs windows
   (emacs would call them "frames") plus a few xterms (mostly
   localhost, only on rare occasions ssh'ing to other hosts) plus
   another dozen windows from Windows applications.  But most of the
   *work* is done in emacs, so it isn't that surprising.

 - emacs doesn't crash immediately, but only if after one of the
   messages quoted one of the following actions "happens".  All of
   those spawn at least one new process:

    * registering spam with bogofilter (through a bash script)

    * completing a mail address in gnus (via EUDC->ldapsearch)

    * opening another frame

    * etc...

 - This can't be the whole story, though.  Emacs never crashes when
   I'm running xmllint (through psgml's 'validate' command) or make
   (through emacs's 'compile' command), which spawn external processes
   as well.

Re-installing both X and emacs did not help.  Since I seem to be
completely unable to narrow it down and since I don't feel seriously
annoyed, please don't spend too much time on it.  This message is just
to confirm that strange Xlib-messages happen.  If someone has an idea
about how to stop this, it would be appreciated though.
-- 
Cheers,
haj


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]