Beyond Compare 3 beta on SUSE 10.1?

[takabanana]

Has anyone tried BC3 beta on SUSE 10.1?
I get a segmentation fault when I run it from the command line.
(I see the BC3 screen for a split second, then it disappears)

This is the first time i've installed it and run it.

[Jacques]

Are you using the enterprise version? We currently don't support that distribution.

If not, is that all that BC3 outputs? Did you remove the output redirection from the shell script that launches BC?

[takabanana]

I am running SUSE 10.1... not OpenSUSE... (was that the Q?)
And yes, it just says "Segmentation fault" - I did remove the redirection in the main script and get the same results.

SUSE is not supported at all? :-(

[chrroe]

I'm using BC3 under openSuSE 10.3 (with KDE 3.5.7 and KDE 4) and it works fine.
Is it possible for you to upgrade your Linux host from 10.1 to 10.3?


Bye
Christoph

[Jacques]

We hope to support you in the next release (456). Check for updates will NOT work however. If you try to use it, beyond compare will crash.

Would you mind listing the contents of "/etc/SuSE-release" for me? Thanks.

[Jacques]

We hope to support you in the next release (456). Check for updates will NOT work however. If you try to use it, beyond compare will crash.

Would you mind listing the contents of "/etc/SuSE-release" for me? Thanks.
Nevermind this post. SuSE Enterprise and other versions that were seg faulting immediately will be fully supported in the next release (457).

[takabanana]

I just want to thank you guys. I tried build 459 on SUSE 10.1.
You guys ROCK!

Thanks for giving us an awesome diffing tool that is also very affordable!

[mrfirmware]

I've got the same problem as takabanana. I'm using BC3 (July 28, 2008) on Ubuntu 8.04 - the Hardy Heron (32-bit). I've added some echos to /usr/bin/bcompare and here's what I see:

LD_LIBRARY_PATH is /usr/lib/beyondcompare
dollar0 is /usr/bin/bcompare, EXEC is /usr/lib/beyondcompare/BCompare, args are 'foo.cpp' 'foo_old.cpp', redirect to /dev/null with /usr/bin/bcompare
hit enter to launch BeyondCompare...
/usr/bin/bcompare: line 38: 13717 Segmentation fault /bin/bash -c "exec -a $0 $EXEC $ARGS >& /dev/null" $0

I realize that Ubuntu isn't supported but I'm not sure how BC could be so specific to a particular disty. I mean 'meld' works fine on Fedora and on Ubuntu.

Thanks.

[Jacques]

Ubuntu has always behaved well for me. As a first step could you remove the redirection to /dev/null and see if BC prints anything useful?

All the scripting variables look normal to me.

[mrfirmware]

I've done that but no new info, e.g.

/usr/bin/bcompare: line 39: 27468 Segmentation fault /bin/bash -c "exec -a $0 $EXEC $ARGS" $0

strace bcompare shows many, many EAGAIN errors being returned on a read() call, e.g.

select(7, [6], [6], NULL, NULL) = 1 (out [6])
writev(6, [{"\22\0\17\0\256\35`\4#\0\0\0#\0\0\0 \0\0\0\t\0\0\0g\0\0"..., 68}], 1) = 68
read(6, "\34\257vj\256\35`\4#\0\0\0\365\206\'\5\0\31\t\10\2 0c \010"..., 4096) = 32
read(6, 0x87bb28c, 4096) = -1 EAGAIN (Resource temporarily unavailable)
read(6, 0x87bb28c, 4096) = -1 EAGAIN (Resource temporarily unavailable)
read(6, 0x87bb28c, 4096) = -1 EAGAIN (Resource temporarily unavailable)
select(7, [6], [6], NULL, NULL) = 1 (out [6])
writev(6, [{"+\0\1\0", 4}], 1) = 4
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\34:wj\256\35`\4R\1\0\0\367\206\'\5\0\257\35\0108\ 310O"..., 4096) = 32
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\34\257wj\256\35`\4R\1\0\0\367\206\'\5\0\31\t\10\2 0c \010"..., 4096) = 32
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\25\0wj\256\35`\4\256\35`\4,\2348\1\0\0\0\0\0&\t\10\224"..., 4096) = 96
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\34:wj\256\35`\4f\1\0\0\370\206\'\5\0\257\35\0100S f\tf"..., 4096) = 224
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\23\0wj\256\35`\4\256\35`\4\0\3769\10x:\315\277\32&\t\10"..., 4096) = 1888
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\34\257wj;\1\0\0?\1\0\0\372\206\'\5\0\31\t\10\20c \010"..., 4096) = 96
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\34\257wj;\1\0\0F\1\0\0\372\206\'\5\0\31\t\10\20c \010"..., 4096) = 32
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\34\257wj;\1\0\0?\1\0\0\372\206\'\5\0\31\t\10\20c \010"..., 4096) = 96
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\34\257wj;\1\0\0F\1\0\0\372\206\'\5\0\31\t\10\20c \010"..., 4096) = 64
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\34\257wj;\1\0\0\"\1\0\0\3\207\'\5\0\31\t\10\20c \0100"..., 4096) = 32
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\34:wj\256\35`\4C\1\0\0\23\207\'\5\0\257\35\0108\3 00Q\10"..., 4096) = 64
read(6, 0x87bb28c, 4096) = -1 EAGAIN (Resource temporarily unavailable)
read(6, 0x87bb28c, 4096) = -1 EAGAIN (Resource temporarily unavailable)
read(6, 0x87bb28c, 4096) = -1 EAGAIN (Resource temporarily unavailable)
select(7, [6], [6], NULL, NULL) = 1 (out [6])
writev(6, [{"\17\0\2\0\256\35`\4", 8}], 1) = 8
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\1:yj\6\0\0\0;\1\0\0,\2348\1\6\0\0\0\310:\315\277\ 233\4"..., 4096) = 56
read(6, 0x87bb28c, 4096) = -1 EAGAIN (Resource temporarily unavailable)
select(7, [6], [6], NULL, NULL) = 1 (out [6])
writev(6, [{"\24\0\6\0,\2348\1\21\1\0\0\6\0\0\0\0\0\0\0\1\0\0\0", 24}], 1) = 24
select(7, [6], [], NULL, NULL) = 1 (in [6])
read(6, "\1\0zj\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\233\4\25\10<c \10"..., 4096) = 32
read(6, 0x87bb28c, 4096) = -1 EAGAIN (Resource temporarily unavailable)
select(7, [6], [6], NULL, NULL) = 1 (out [6])
writev(6, [{"\17\0\2\0,\2348\1", 8}], 1) = 8
select(7, [6], [], NULL, NULL) = ? ERESTARTNOHAND (To be restarted)
+++ killed by SIGSEGV +++
Process 27621 detached

- Mark

[Jacques]

Do you know what command is opening FD 6?

Also have you tried running BC as root, either directly or using gksu(do)? Have any previous versions ran for you or is this the first version you have tried?

[mrfirmware]

I don't know what command is openiong FD 6. Running sudo bcompare has same result. I ran it with gdb and noticed that you are doing some http lookups. I wonder if bcompare is barfing on a auth rejection by our wonderful firewall proxy, BlueCoat a.k.a bcauth?

(gdb) s

Program received signal SIGSEGV, Segmentation fault.
0x081250ad in CheckException (AResponseCode=142243384, AIgnoreReplies=0xb4ae1264, AUnexpectedContentTimeout=-1) at IdHTTP.pas:1821
1821 in IdHTTP.pas
(gdb) bt
#0 0x081250ad in CheckException (AResponseCode=142243384, AIgnoreReplies=0xb4ae1264, AUnexpectedContentTimeout=-1) at IdHTTP.pas:1821
#1 0x081253e2 in @Idhttp@TIdHTTPProtocol@ProcessResponse (Self=0x89d0ad0, AIgnoreReplies=0xb4ae128c) at IdHTTP.pas:1902
#2 0x08125afa in @Idhttp@TIdCustomHTTP@DoRequest (Self=0x89da8d0, AMethod=0xb7960a0c #129'���',
AURL=0x89b6770 'http://www.scootersoftware.com/checkupdates.php?product=bc3&minor=0&maint=0&auth=9352.0397&build=464&edition=prodebug&platform=linux&lang=silent', ASource=0x0,
AResponseContent=0x87a7638, AIgnoreReplies=0x0) at IdHTTP.pas:2095
#3 0x081258ad in @Idhttp@TIdCustomHTTP@Get (Self=0x87a7638,
AURL=0x89b6770 'http://www.scootersoftware.com/checkupdates.php?product=bc3&minor=0&maint=0&auth=9352.0397&build=464&edition=prodebug&platform=linux&lang=silent', AResponseContent=0xb7960a0c,
AIgnoreReplies=0xb4ae12f8) at IdHTTP.pas:2052
#4 0x081257b6 in @Idhttp@TIdCustomHTTP@Get (Self=0x89da8d0,
AURL=0x89b6770 'http://www.scootersoftware.com/checkupdates.php?product=bc3&minor=0&maint=0&auth=9352.0397&build=464&edition=prodebug&platform=linux&lang=silent', AIgnoreReplies=0xb4ae1318, Result=0x0)
at IdHTTP.pas:2038
#5 0x08122bf3 in @Idhttp@TIdCustomHTTP@Get (Self=0x89da8d0,
AURL=0x89b6770 'http://www.scootersoftware.com/checkupdates.php?product=bc3&minor=0&maint=0&auth=9352.0397&build=464&edition=prodebug&platform=linux&lang=silent', Result=0xb4ae135c '') at IdHTTP.pas:853
#6 0x081262f6 in @Httpget@THTTPGet@GetStringThreaded (Self=0x89da848, AThread=0x0) at HTTPGet.pas:182
#7 0x08125cee in TSsRunnerThread.Execute (Self=0x894bfa8) at SsThreads.pas:120
#8 0x0808bf82 in @Classes@ThreadProc () at SsCodePage.pas:116
#9 0xb4ae1b90 in ?? ()
#10 0x0806125a in @SystemThreadWrapper () at SsCodePage.pas:116
#11 0xb4ae13a4 in ?? ()
#12 0x08071a94 in ThreadFuncWrapper (Arg=0x89e0d58) at KylExcept.pas:294
#13 0xb795d4fb in start_thread () from /lib/tls/i686/cmov/libpthread.so.0
#14 0xb78dbe5e in clone () from /lib/tls/i686/cmov/libc.so.6

In general, having a program check for updates at start up doesn't seem to be a good idea. At best the program can connect and continue, otherwise we wait for some timeout, or worse, the program sigsegv's when it gets a request for authentication from BlueCoat.

I'd like the option to not have auto-update checking for my diff program. Is that possible?

- Mark

[Dave_L]

I'd like the option to not have auto-update checking for my diff program. Is that possible?

Tools >> Options >> Startup

Uncheck "Check for updates every:".

[Craig]

Tools >> Options >> Startup

Uncheck "Check for updates every:".

Mark's problem is that he can't start the program because of the check for updates, so that might be difficult. ;)

Mark: Try saving the attached file as "BCPreferences.xml" in your ~/.beyondcompare folder. It will disable checking for updates.

[Dave_L]

Oh, you wanted a practical solution. :o

[mrfirmware]

Craig,

Thanks very much! That did the trick. I just compared some folders and files and it seems to be working fine.

Thanks,

- Mark

[Jacques]

BC should be able to handle this in 3.0.1.