[gentoo-science] Mathematica issue

[gentoo-science] Mathematica issue

[christoph irrenfried]

Hello,
my name is Chris and I have a huge Mathemaica problem. I need it for
work on a daily bases but after I did my regular update a couple mounths
ago it just wont start again. I filed this issue on the gentoo forum but
I cant find a way to get rid of this problem. The error message is as a
pic included.
Than a couple weeks ago I bought a laptop and installed gentoo with
gnome on it but there aswell and I got the same error. I have
no idea what's the problem. I wanted to ask you if someone has a
gentoo/gnome system with mathematica running and if so if it is possible
to get the USE flags from your make.conf or I don't know if there is a
way to compare my settings with your system and see if maybe (or
definitely) I messed something up e.g. fonts or something ....

Thanks in advance,
chi

Re: [gentoo-science] Mathematica issue

[christoph irrenfried]

On 02/05/2015 03:52 PM, Markus Oehme wrote:
> (a) Does the license server still operate (maybe your university changed the
>     setup)?
All the others on in my department have no problem with the connection
to the server. The only difference is that they are using Debian.
>
> (b) Are you able to reach the license server on the network (issue a ping or
>     similar)?
yes if i ping the server i get a positive response!
>
> (c) If both have positive answers I would contact the operator of the
>     license server for further trouble shooting.
I contacted than with the answer: That you get from using gentoo (not
the same words but it was what they meant ;( ).

The wired thing is that when I installed it after I had a fresh
installation of gentoo it worked but than I installed other stuff and it
stopped working. I installed gentoo a couple of times but it looks like
it always stops after different installations. It definitely is caused
by gentoo.

Therefore I wanted to know if someone else works with mathematica on
gentoo and if that person can give me his configuration to check it
against mine.

Thanks!

Re: [gentoo-science] Mathematica issue

[François Bissey]

Hi,

Can you point to the post in forum so I can have the details please?

François

> On 6/02/2015, at 02:40, christoph irrenfried <c.irrenfried@gmx.at> wrote:
> 
> Hello,
> my name is Chris and I have a huge Mathemaica problem. I need it for
> work on a daily bases but after I did my regular update a couple mounths
> ago it just wont start again. I filed this issue on the gentoo forum but
> I cant find a way to get rid of this problem. The error message is as a
> pic included.
> Than a couple weeks ago I bought a laptop and installed gentoo with
> gnome on it but there aswell and I got the same error. I have
> no idea what's the problem. I wanted to ask you if someone has a
> gentoo/gnome system with mathematica running and if so if it is possible
> to get the USE flags from your make.conf or I don't know if there is a
> way to compare my settings with your system and see if maybe (or
> definitely) I messed something up e.g. fonts or something ....
> 
> Thanks in advance,
> chi
> 

Re: [gentoo-science] Mathematica issue

[christoph irrenfried]

Sure: https://forums.gentoo.org/viewtopic-t-997008.html


On 02/05/2015 08:46 PM, François Bissey wrote:
> Hi,
>
> Can you point to the post in forum so I can have the details please?
>
> François
>
>> On 6/02/2015, at 02:40, christoph irrenfried <c.irrenfried@gmx.at> wrote:
>>
>> Hello,
>> my name is Chris and I have a huge Mathemaica problem. I need it for
>> work on a daily bases but after I did my regular update a couple mounths
>> ago it just wont start again. I filed this issue on the gentoo forum but
>> I cant find a way to get rid of this problem. The error message is as a
>> pic included.
>> Than a couple weeks ago I bought a laptop and installed gentoo with
>> gnome on it but there aswell and I got the same error. I have
>> no idea what's the problem. I wanted to ask you if someone has a
>> gentoo/gnome system with mathematica running and if so if it is possible
>> to get the USE flags from your make.conf or I don't know if there is a
>> way to compare my settings with your system and see if maybe (or
>> definitely) I messed something up e.g. fonts or something ....
>>
>> Thanks in advance,
>> chi
>>
>

Re: [gentoo-science] Mathematica issue

[Andrew Savchenko]

[-- Attachment #1: Type: text/plain, Size: 3327 bytes --]

Hi,

On Thu, 05 Feb 2015 14:40:10 +0100 christoph irrenfried wrote:
> Hello,
> my name is Chris and I have a huge Mathemaica problem. I need it for
> work on a daily bases but after I did my regular update a couple mounths
> ago it just wont start again. I filed this issue on the gentoo forum but
> I cant find a way to get rid of this problem. The error message is as a
> pic included.
> Than a couple weeks ago I bought a laptop and installed gentoo with
> gnome on it but there aswell and I got the same error. I have
> no idea what's the problem. I wanted to ask you if someone has a
> gentoo/gnome system with mathematica running and if so if it is possible
> to get the USE flags from your make.conf or I don't know if there is a
> way to compare my settings with your system and see if maybe (or
> definitely) I messed something up e.g. fonts or something ....

I don't use Mathematica now (maxima and R are sufficient for my
needs), but I can give you a few ideas.

0) I'm not sure USE flags will help you, problem may lie in other
area: missing library, toolkit, util, version mismatch,
misconfigured system and so on.

1) Run Mathematica from console. It may drop to the terminal some
useful messages (errors, warngings, etc) which may give a clue.

2) If any security mechanisms are engaged (e.g. SELinux, PaX,
GrSecurity, Yama framework and so on), please disable all of them
for testing. They may interfere with complex proprietary software
badly.

3) Please be sure, that iptables on your host are not a hinder
(e.g. disable iptables for testing, if in use).

4) Run tcpdump (net-analyzer/tcpdump) to see what Mathematica tries
to do on the net, in particular what ip and port it tries to
access. Verify that they are accessible using netcat or telnet.

5) Try to run Mathematica in terminal in debug mode. Since I don't
have this app right now I can't tell you exactly how to enable it,
but looking for --help output (or in Mathematica docs) may help.
Maybe it supports something like --verbose or --debug, which will
help you to get more information, then in step 1).

6) Inspect Mathematica requirements from its docs, maybe some
package or tool is missed, maybe version is wrong.

7) If anything above fails, there is one more step, an ultimate one.
Run (assuming Mathematica command runs it):

$ strace -ff Mathematica -o mathematica
(install dev-util/strace if you don't have it)

And look through all mathematica.$pid files. They will contain
every system call. You will be able to see what files application
tries to access, what network actions it tries to perform and so
on. The answer will be there, but it will not be very easy to find.

A good start will be to grep trough all file and socket access (or
use -e filter of strace itself). Maybe it tries to load some
library, or to run some tool and fails. Maybe required component is
present on your system, but version is different from what is
needed.

8) As a complementary to strace you may use dev-util/ltrace. ltrace
will print you *each* library call made buy an application. Output
data will be enormous, but will contain a very detail fingerprint
of what application does. With some effort this should help to
understand what is wrong.

Best regards,
Andrew Savchenko

[-- Attachment #2: Type: application/pgp-signature, Size: 819 bytes --]

Re: [gentoo-science] Mathematica issue

[christoph irrenfried]

[-- Attachment #1: Type: text/plain, Size: 3572 bytes --]

First of all thanks for your great help!

On 02/06/2015 12:24 AM, Andrew Savchenko wrote:

> 1) Run Mathematica from console. It may drop to the terminal some
> useful messages (errors, warngings, etc) which may give a clue.
There is no output at all.

> 2) If any security mechanisms are engaged (e.g. SELinux, PaX,
> GrSecurity, Yama framework and so on), please disable all of them
> for testing. They may interfere with complex proprietary software
> badly.
>
> 3) Please be sure, that iptables on your host are not a hinder
> (e.g. disable iptables for testing, if in use).
I can work with Mathematica on my VirtualBox where I have Win7
installed, therefore I assume that cant be the problem.
> 4) Run tcpdump (net-analyzer/tcpdump) to see what Mathematica tries
> to do on the net, in particular what ip and port it tries to
> access. Verify that they are accessible using netcat or telnet.
I looked through the output and tried to telnet all addresses and it
worked. There is a lot of communication going on between the host and
the server. On Monday when my colleague is back I will compare my output
against his.
> 5) Try to run Mathematica in terminal in debug mode. Since I don't
> have this app right now I can't tell you exactly how to enable it,
> but looking for --help output (or in Mathematica docs) may help.
> Maybe it supports something like --verbose or --debug, which will
> help you to get more information, then in step 1).
/chi@fslgdpc46 ~ $ math -lmverbose//
//Mathematica 10.0 for Linux x86 (64-bit)//
//Copyright 1988-2014 Wolfram Research, Inc.//
//Talk to server 'lizenzxxxxxx.at' on port 16286 ...//
//Error reading: Success.//
//Talk to server '///lizenzxxxxxx/.at' on port 16286 ...//
//Error reading: Success.//
//
//Mathematica cannot find a valid password.//
//
//For automatic Web Activation enter your activation key//
//(enter return to skip Web Activation): /
It looks like the connection is fine but can't find a password which is
wired because with the VirtualBox it worked and that is the same ip
which requests a password.
> 6) Inspect Mathematica requirements from its docs, maybe some
> package or tool is missed, maybe version is wrong.
Thats the wired thing because everything worked fine and it stopped
working after I did a world update. But I will try to find out what
packages Mathematica needs and look if i have them all.

>
> 7) If anything above fails, there is one more step, an ultimate one.
> Run (assuming Mathematica command runs it):
>
> $ strace -ff Mathematica -o mathematica
> (install dev-util/strace if you don't have it)
>
> And look through all mathematica.$pid files. They will contain
> every system call. You will be able to see what files application
> tries to access, what network actions it tries to perform and so
> on. The answer will be there, but it will not be very easy to find.
>
> A good start will be to grep trough all file and socket access (or
> use -e filter of strace itself). Maybe it tries to load some
> library, or to run some tool and fails. Maybe required component is
> present on your system, but version is different from what is
> needed.
>
> 8) As a complementary to strace you may use dev-util/ltrace. ltrace
> will print you *each* library call made buy an application. Output
> data will be enormous, but will contain a very detail fingerprint
> of what application does. With some effort this should help to
> understand what is wrong.
>
> Best regards,
> Andrew Savchenko

These steps I try on the weekend.

Thanks for you afford it is greatly appreciated!!!!

chi


[-- Attachment #2: Type: text/html, Size: 4980 bytes --]

Re: [gentoo-science] Mathematica issue

[christoph irrenfried]

[-- Attachment #1: Type: text/plain, Size: 9339 bytes --]

On 02/06/2015 12:24 AM, Andrew Savchenko wrote:
> 7) If anything above fails, there is one more step, an ultimate one.
> Run (assuming Mathematica command runs it):
>
> $ strace -ff Mathematica -o mathematica
> (install dev-util/strace if you don't have it)
>
> And look through all mathematica.$pid files. They will contain
> every system call. You will be able to see what files application
> tries to access, what network actions it tries to perform and so
> on. The answer will be there, but it will not be very easy to find.
>
> A good start will be to grep trough all file and socket access (or
> use -e filter of strace itself). Maybe it tries to load some
> library, or to run some tool and fails. Maybe required component is
> present on your system, but version is different from what is
> needed.
>
> 8) As a complementary to strace you may use dev-util/ltrace. ltrace
> will print you *each* library call made buy an application. Output
> data will be enormous, but will contain a very detail fingerprint
> of what application does. With some effort this should help to
> understand what is wrong.
Hi again. I had now some time and did the tracing with Strace:

I got heaps of output but I cant really figure out what it means:


Before i put in the license server:
/[pid 18439] futex(0xc6cc80, FUTEX_WAKE_PRIVATE, 1) = 0
[pid 18439] futex(0xc6bf74,
FUTEX_WAIT_BITSET_PRIVATE|FUTEX_CLOCK_REALTIME, 5221, {1423637720,
118416000}, ffffffff) = -1 ETIMEDOUT (Connection timed out)
[pid 18439] futex(0xc6cc80, FUTEX_WAKE_PRIVATE, 1) = 0
[pid 18439] futex(0xc6bf74,
FUTEX_WAIT_BITSET_PRIVATE|FUTEX_CLOCK_REALTIME, 5223, {1423637720,
138564000}, ffffffff <unfinished ...>
[pid 18578] <... poll resumed> )        = 0 (Timeout)
[pid 18573] <... poll resumed> )        = 0 (Timeout)
[pid 18578] poll([{fd=38, events=POLLIN}], 1, 0 <unfinished ...>
[pid 18573] poll([{fd=4, events=POLLIN}], 1, 0 <unfinished ...>
[pid 18578] <... poll resumed> )        = 0 (Timeout)
[pid 18573] <... poll resumed> )        = 0 (Timeout)
[pid 18578] flock(45, LOCK_EX <unfinished ...>
[pid 18573] flock(11, LOCK_EX <unfinished ...>
[pid 18578] <... flock resumed> )       = 0
[pid 18573] <... flock resumed> )       = 0
[pid 18578] flock(45, LOCK_UN <unfinished ...>
[pid 18573] flock(11, LOCK_UN <unfinished ...>
[pid 18578] <... flock resumed> )       = 0
[pid 18573] <... flock resumed> )       = 0
[pid 18578] poll([{fd=38, events=POLLIN}], 1, 50 <unfinished ...>
[pid 18573] poll([{fd=4, events=POLLIN}], 1, 50 <unfinished ...>/



When I put in the license server:
/[pid 18573] <... poll resumed> )        = 0 (Timeout)
[pid 18573] poll([{fd=4, events=POLLIN}], 1, 0) = 0 (Timeout)
[pid 18573] flock(11, LOCK_EX)          = 0
[pid 18573] flock(11, LOCK_UN)          = 0
[pid 18573] poll([{fd=4, events=POLLIN}], 1, 50 <unfinished ...>
[pid 18439] <... futex resumed> )       = -1 ETIMEDOUT (Connection timed
out)
[pid 18439] futex(0xc6cc80, FUTEX_WAKE_PRIVATE, 1) = 0
[pid 18439] futex(0xc6bf74,
FUTEX_WAIT_BITSET_PRIVATE|FUTEX_CLOCK_REALTIME, 13547, {1423637805,
828488000}, ffffffff <unfinished ...>
/

When I quit the program:
/[pid 18428] poll([{fd=3, events=POLLIN|POLLOUT}], 1, 4294967295) = 1
([{fd=3, revents=POLLOUT}])
[pid 18428] writev(3, [{"<\0\2\0\0\0@\2.\0\2\0E\21@\2+\0\1\0", 20},
{NULL, 0}, {"", 0}], 3) = 20
[pid 18428] poll([{fd=3, events=POLLIN}], 1, 4294967295) = 1 ([{fd=3,
revents=POLLIN}])
[pid 18428] recvmsg(3, {msg_name(0)=NULL,
msg_iov(1)=[{"\1\2!F\0\0\0\0\7\0\300\1\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0",
4096}], msg_controllen=0, msg_flags=0}, 0) = 32
[pid 18428] recvmsg(3, 0x7fffb66d1d70, 0) = -1 EAGAIN (Resource
temporarily unavailable)
[pid 18428] recvmsg(3, 0x7fffb66d1d70, 0) = -1 EAGAIN (Resource
temporarily unavailable)
[pid 18428] poll([{fd=3, events=POLLIN|POLLOUT}], 1, 4294967295) = 1
([{fd=3, revents=POLLOUT}])
[pid 18428] writev(3,
[{"<\0\2\0\22\5@\2<\0\2\0\23\5@\2<\0\2\0\24\5@\2b\0\4\0\7\0\0\0", 32},
{"MIT-SHM", 7}, {"\0", 1}], 3) = 40
[pid 18428] poll([{fd=3, events=POLLIN}], 1, 4294967295) = 1 ([{fd=3,
revents=POLLIN}])
[pid 18428] recvmsg(3, {msg_name(0)=NULL,
msg_iov(1)=[{"\1\0%F\0\0\0\0\1\202A\200\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0",
4096}], msg_controllen=0, msg_flags=0}, 0) = 32
[pid 18428] recvmsg(3, 0x7fffb66d1c10, 0) = -1 EAGAIN (Resource
temporarily unavailable)
[pid 18428] recvmsg(3, 0x7fffb66d1c10, 0) = -1 EAGAIN (Resource
temporarily unavailable)
[pid 18428] poll([{fd=3, events=POLLIN|POLLOUT}], 1, 4294967295) = 1
([{fd=3, revents=POLLOUT}])
[pid 18428] writev(3, [{"b\0\10\0\27\0@\2", 8}, {"Generic Event
Extension", 23}, {"\0", 1}], 3) = 32
[pid 18428] poll([{fd=3, events=POLLIN}], 1, 4294967295) = 1 ([{fd=3,
revents=POLLIN}])
[pid 18428] recvmsg(3, {msg_name(0)=NULL,
msg_iov(1)=[{"\1\0&F\0\0\0\0\1\200\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0",
4096}], msg_controllen=0, msg_flags=0}, 0) = 32
[pid 18428] recvmsg(3, 0x7fffb66d1b70, 0) = -1 EAGAIN (Resource
temporarily unavailable)
[pid 18428] recvmsg(3, 0x7fffb66d1b70, 0) = -1 EAGAIN (Resource
temporarily unavailable)
[pid 18428] poll([{fd=3, events=POLLIN|POLLOUT}], 1, 4294967295) = 1
([{fd=3, revents=POLLOUT}])
[pid 18428] writev(3, [{"\200\0\2\0\1\0\0\0", 8}, {NULL, 0}, {"", 0}],
3) = 8
[pid 18428] poll([{fd=3, events=POLLIN}], 1, 4294967295) = 1 ([{fd=3,
revents=POLLIN}])
[pid 18428] recvmsg(3, {msg_name(0)=NULL,
msg_iov(1)=[{"\1\0'F\0\0\0\0\1\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0",
4096}], msg_controllen=0, msg_flags=0}, 0) = 32
[pid 18428] recvmsg(3, 0x7fffb66d1c20, 0) = -1 EAGAIN (Resource
temporarily unavailable)
[pid 18428] recvmsg(3, 0x7fffb66d1c20, 0) = -1 EAGAIN (Resource
temporarily unavailable)
[pid 18428] shmdt(0x7f889804c000)       = 0
[pid 18428] poll([{fd=3, events=POLLIN|POLLOUT}], 1, 4294967295) = 1
([{fd=3, revents=POLLOUT}])
[pid 18428] writev(3, [{"\202\2\2\0\25\5@\2+\0\1\0", 12}, {NULL, 0},
{"", 0}], 3) = 12
[pid 18428] poll([{fd=3, events=POLLIN}], 1, 4294967295) = 1 ([{fd=3,
revents=POLLIN}])
[pid 18428] recvmsg(3, {msg_name(0)=NULL,
msg_iov(1)=[{"\1\2)F\0\0\0\0\7\0\300\1\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0",
4096}], msg_controllen=0, msg_flags=0}, 0) = 32
[pid 18428] recvmsg(3, 0x7fffb66d1d70, 0) = -1 EAGAIN (Resource
temporarily unavailable)
[pid 18428] recvmsg(3, 0x7fffb66d1d70, 0) = -1 EAGAIN (Resource
temporarily unavailable)
[pid 18428] shutdown(3, SHUT_RDWR)      = 0
[pid 18428] close(3)                    = 0
[pid 18428] write(7, "@", 1)            = 1
[pid 18428] close(7)                    = 0
[pid 18428] close(6)                    = 0
[pid 18428] rt_sigaction(SIGCHLD, {0x7f88bb633fe0, [],
SA_RESTORER|SA_NOCLDSTOP, 0x7f88b76d50f0}, {0x7f88bb633fe0, [],
SA_RESTORER|SA_SIGINFO, 0x7f88b76d50f0}, 8) = 0
[pid 18428] rt_sigaction(SIGCHLD, {0x7f88bb633fe0, [],
SA_RESTORER|SA_SIGINFO, 0x7f88b76d50f0}, NULL, 8) = 0
[pid 18428] futex(0x41d143c, FUTEX_CMP_REQUEUE_PRIVATE, 1, 2147483647,
0x41d1410, 24 <unfinished ...>
[pid 18436] <... futex resumed> )       = 0
[pid 18428] <... futex resumed> )       = 1
[pid 18436] futex(0x41d1410, FUTEX_WAIT_PRIVATE, 2, NULL <unfinished ...>
[pid 18428] futex(0x41d1410, FUTEX_WAKE_PRIVATE, 1 <unfinished ...>
[pid 18436] <... futex resumed> )       = -1 EAGAIN (Resource
temporarily unavailable)
[pid 18428] <... futex resumed> )       = 0
[pid 18436] futex(0x41d1410, FUTEX_WAKE_PRIVATE, 1) = 0
[pid 18436] --- SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_ACCERR,
si_addr=0x2381320} ---
[pid 18436] write(2, "Mathematica has received the sig"...,
61Mathematica has received the signal: SIGSEGV and has exited.
) = 61
[pid 18436] write(2, "If possible, please report this "..., 63If
possible, please report this problem to support@wolfram.com
) = 63
[pid 18436] write(2, "describing in as much detail as "..., 61describing
in as much detail as possible what you were doing
) = 61
[pid 18436] write(2, "when the problem occurred.\n", 27when the problem
occurred.
) = 27
[pid 18436] rt_sigaction(SIGSEGV, {SIG_DFL, [], SA_RESTORER,
0x7f88b76d50f0}, {0xae00d0, [], SA_RESTORER, 0x7f88b76d50f0}, 8) = 0
[pid 18436] tgkill(18428, 18436, SIGSEGV) = 0
[pid 18436] rt_sigreturn( <unfinished ...>
[pid 18428] exit_group(0)               = ?
[pid 18436] <... rt_sigreturn resumed> ) = 37229344
[pid 18436] +++ exited with 0 +++
[pid 18428] +++ exited with 0 +++
<... wait4 resumed> [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0, NULL) =
18428
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
--- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=18428,
si_status=0, si_utime=174, si_stime=73} ---
wait4(-1, 0x7fff49e8e898, WNOHANG, NULL) = -1 ECHILD (No child processes)
rt_sigreturn()                          = 0
rt_sigaction(SIGINT, {SIG_DFL, [], SA_RESTORER, 0x7f338d84b560},
{0x43b330, [], SA_RESTORER, 0x7f338d84b560}, 8) = 0
rt_sigprocmask(SIG_BLOCK, NULL, [], 8)  = 0
read(255, "\n\n", 6035)                 = 2
rt_sigprocmask(SIG_BLOCK, NULL, [], 8)  = 0
rt_sigprocmask(SIG_BLOCK, NULL, [], 8)  = 0
read(255, "", 6035)                     = 0
exit_group(0)                           = ?
+++ exited with 0 +++/

Anyone an idea whats going wrong?

I also tried ltrace but it wont work:
/chi@chi: ltrace -f mathematica
"/usr/local/bin/mathematica" is not an ELF file/

A guy from the gentoo forum meant that I should put this in my make.conf
and recompile:
ABI_X86="64 32"
It also did not help at all. ;(


[-- Attachment #2: Type: text/html, Size: 12202 bytes --]