Bug 104363 - acroread abends on Find with no hit
acroread abends on Find with no hit
Status: CLOSED NOTABUG
Product: Red Hat Linux
Classification: Retired
Component: mozilla (Show other bugs)
9
i586 Linux
medium Severity medium
: ---
: ---
Assigned To: Christopher Blizzard
:
Depends On:
Blocks:
  Show dependency treegraph
 
Reported: 2003-09-13 10:30 EDT by Mike Vanecek
Modified: 2007-04-18 12:57 EDT (History)
1 user (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Environment:
Last Closed: 2003-09-19 15:09:53 EDT
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---


Attachments (Terms of Use)

  None (edit)
Description Mike Vanecek 2003-09-13 10:30:31 EDT
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030703

Description of problem:
Open a pdf file either through Mozilla, Nautilus, or directly using Adobe
acroread 5.0.6/7/8 on Red Hat 9. Issue a Find for text not in the document.
Answer OK to no find message. Acroread either exits/abends or hangs.

Same test done on version RH 7.3 without error. Several other RH 9 users have
reported the same problem.


Version-Release number of selected component (if applicable):
acroread 5.0.6 or 5.0.7 or 5.0.8

How reproducible:
Always

Steps to Reproduce:
1.Open pdf using acroread on RH 9
2.Issue FIND for text not in the document
3.Answer OK to the not found message
    

Actual Results:  Acroread either exits/abends or hangs. If it hangs, one must
kill the process with root authority.


Expected Results:  Acroread should have returned to its main screen.

Additional info:

Running RH 9 with all up2date changes. Using Gnome as the desktop. Process works
on RH 7.3.
Comment 1 Christopher Blizzard 2003-09-19 15:09:53 EDT
We can't fix bugs in acrobat since it's not open source.
Comment 2 Mike Vanecek 2003-09-19 18:44:34 EDT
I am not asking for you to fix acroread. It works on windows, it works on red
hat 7.3, and may work on other distros. I am asking you to fix whatever is
broken in red hat 9 so that an application that runs on serveral other systems
also runs on red hat 9.
Comment 3 Christopher Blizzard 2003-09-29 16:56:32 EDT
Have you tried setting LANG=en_US before starting acroread?
Comment 4 Mike Vanecek 2003-10-04 10:33:20 EDT
Yes, LANG is set correctly.

[admin@www admin]$ echo $LANG
en_US

Except for the Find no-find problem, acroread works just fine (Searches that
find something work, printing works, etc.). It is just the Find no-find abend
that is a problem. This has been reported on other RH 8/9 systems on one of the
mailing lists as well.

Could this problem have anything to do with the migration from GCC 2.95 to 3.x?
What is different on RH 8/9 from RH 7.3. 

BTW, this has also been reported to Adobe. No resolution from them either.
Comment 5 Rex Dieter 2003-10-04 12:03:47 EDT
FYI, for me, it crashes just the same on rh73 too.
Comment 6 Bevan Bennett 2003-10-16 14:29:48 EDT
Just to add, I'm running around 40 desktops in various flavors of RedHat, and so
can offer some potentially useful information.

Adobe Acrobat Reader 5.0.x (tested x=6,7,8), on our systems running RedHat
versions 8 and higher (tested 8,9,taroon), consistantly dumps core after the
execution of a "Find" for text that does not exist in the document. 

The same acroread binaries (run from NFS) do not have this problem on any of our
RedHat 7.1, 7.2 or 7.3 systems. 

Since it was a change in RedHat that caused the problem to appear, not an Adobe
change), we were hoping that this would be an appropriate forum to discuss the
issue, even though acrobat reader is not officially part of the distribution.
Comment 7 Joshua Benjamin 2004-01-12 16:06:24 EST
I confirm this as well. Same acroread works fine in RH 7.1/7.3 but
crashes in RH9 if the searched text is not found. Just attempt to
reproduce it yourselves and see.

strace ends with
ioctl(3, FIONREAD, [0])                 = 0
write(3, ";\3\7\0\32\1 \2\0\0\0\0:\1\371\0\t\0\332\0W\2\371\0\10"...,
144) = 144
gettimeofday({1073941705, 442533}, NULL) = 0
--- SIGSEGV (Segmentation fault) @ 0 (0) ---
rt_sigprocmask(SIG_UNBLOCK, [SEGV], NULL, 8) = 0
mmap2(NULL, 2097152, PROT_NONE,
MAP_PRIVATE|MAP_ANONYMOUS|MAP_NORESERVE, -1, 0) = 0x4115b000
munmap(0x4115b000, 675840)              = 0
munmap(0x41300000, 372736)              = 0
mprotect(0x41200000, 32768, PROT_READ|PROT_WRITE) = 0
rt_sigprocmask(SIG_UNBLOCK, [ABRT], NULL, 8) = 0
gettid()                                = 24890
tkill(24890, SIGABRT)                   = 0
--- SIGABRT (Aborted) @ 0 (0) ---
+++ killed by SIGABRT +++
Comment 8 Roderick Johnstone 2004-05-30 09:07:21 EDT
Someone in one of the Adobe forums suggested:

export MALLOC_CHECK_=0

as a workaround for this problem. It seems to cure the problem for me
on FC1, but I've no idea if there are any bad side effects.
Comment 9 Mike Vanecek 2004-05-30 09:51:02 EDT
I tried export MALLOC_CHECK_=0 on my RH 9 system and the find still
abends at the point where I click OK on the no find message. I wonder
why it works with FC1 and not RH 9? Don't plan on leaving RH 9 for a
couple more years or so and would like to see this work.
Comment 10 Mike Vanecek 2004-05-30 10:24:03 EDT
I found the message to which you refer:

Topic name:   Acroread 5.0.6/7/8 on Linux abends with Find no hit
Forum name: Acrobat UNIX    -    Message #14
Posted By: Ar Nehe    -    Mar 31, 2004 01:50 am
Here is a workaround until Adobe has fixed the problem: Add following
2 lines to /usr/X11R6/bin/acroread. MALLOC_CHECK_=0 export
MALLOC_CHECK_ (see also
http://lists.suse.com/archive/suse-linux/2003-Nov/1455.html) 

I added the lines mentioned above to /usr/bin/acroread and indeed the
error does stop. Not sure why it needs this way to work, but at least
it works.

Note You need to log in before you can comment on or make changes to this bug.