Re: [casper] Fix - Simulink R2021a crashes on startup on Ubuntu 18.04

2022-06-14 Thread Yaddehi Udara De Silva 
Hi all,

I have also faced this issue. As initially mentioned in this thread,
Mathworks has provided a workaround which can also be found here:
https://www.mathworks.com/matlabcentral/answers/1454674-why-does-matlab-crash-on-linux-with-inconsistency-detected-by-ld-so-elf-dl-tls-c-597-_dl_allo

If you are like myself who works in a server environment and cannot update
glibc, the good news is that this crash does not always happens. The crash
is related to the order libraries are loaded to the memory. So if you
managed to load the libraries before opening Simulink you are safe.

I've found opening a live script and opening the .slx without opening the
.prj helps avoid the crash. Also, once you open Simulink never close it :D
Hopefully this will allow you to continue work until a proper fix is
available. Thank You.

Best Regards,
Udara De Silva

On Tue, Jun 14, 2022, 2:16 PM 'Greg Lindahl' via casper@lists.berkeley.edu <
casper@lists.berkeley.edu> wrote:

> I just corresponded with Mathworks, and it turns out that RHEL very
> recently backported this libc fix into RHEL 8.
>
> It's likely that Debian/Ubuntu will follow.
>
> On Thu, Jun 9, 2022 at 7:45 AM Adam Isaacson 
> wrote:
>
>> Well done, Morag and team! That is a great catch and very helpful to me
>> and others!
>>
>> Kind regards,
>>
>> Adam
>>
>> On Thu, Jun 9, 2022 at 2:43 PM Morag Brown  wrote:
>>
>>> Hey all,
>>>
>>> I had a quick look in the mail archive and didn't see anything on this,
>>> but it's not impossible others will come across/have come across the same
>>> issue I've encountered, so figured I'd potentially save some people future
>>> frustrations and heart attacks.
>>>
>>> There's an issue with later versions of Matlab (R2021a and up) on older
>>> versions of Ubuntu (18.04 and down) where opening Simulink intermittently
>>> causes Matlab to crash with some variation of the following error:
>>>
>>> *Inconsistency detected by ld.so: ../elf/dl-tls.c: 517:
>>> _dl_allocate_tls_init: Assertion `listp != NULL' failed!*
>>>
>>> This is due to a bug in the glibc library present on older versions of
>>> Linux distros that doesn't play nicely with the new way Matlab does some
>>> things. Fortunately, Mathworks doth provide an answer. Not so fortunately,
>>> their official answer is "upgrade to a newer operating system with a
>>> version of glibc that doesn't have this bug". Lol.
>>>
>>> But they also kindly provide a solution
>>>  for those of
>>> us who can't just upgrade; an unofficial patch of the glibc libraries that
>>> fixes this bug for various OS versions. This of course comes with the major
>>> caveat that modifying system files is done at your own risk - all processes
>>> in the OS share glibc, so the patch will affect your whole system, not just
>>> Matlab, and this could cause weirdness.
>>>
>>> So if you find yourself in the same position I was yesterday, wondering
>>> if you should risk it all just for Simulink - I'm happy to report that I
>>> applied this patch, the universe did not end and nothing on my machine
>>> *seems* to be broken. Great success.
>>>
>>> That is all,
>>> Morag
>>>
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "casper@lists.berkeley.edu" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to casper+unsubscr...@lists.berkeley.edu.
>>> To view this discussion on the web visit
>>> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAGH-0Tc-dJ0AUnC21S5oTP-pRXyhMC_Ld%2B8KyQ31NuSFZ7HfkQ%40mail.gmail.com
>>> 
>>> .
>>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "casper@lists.berkeley.edu" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to casper+unsubscr...@lists.berkeley.edu.
>> To view this discussion on the web visit
>> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CADTJ%3DnGPD6cV7Vv0XuMVAezU%3D755iXVkMNzw%3DtGgc6Mc%3D_aTZQ%40mail.gmail.com
>> 
>> .
>>
>
>
> --
> Greg Lindahl
> Software Architect, Event Horizon Telescope
> Smithsonian Astrophysical Observatory
> 60 Garden Street | MS 66 | Cambridge, MA 02138
>
> --
> You received this message because you are subscribed to the Google Groups "
> casper@lists.berkeley.edu" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to casper+unsubscr...@lists.berkeley.edu.
> To view this discussion on the web visit
> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAJT0TuNT%3Da2EsjgLXBB%2BcVjOJ_eQkS0CQ_ovbxq7Oq9irih9XA%40mail.gmail.com
>

Re: [casper] Fix - Simulink R2021a crashes on startup on Ubuntu 18.04

2022-06-14 Thread 'Greg Lindahl' via casper@lists.berkeley.edu 
I just corresponded with Mathworks, and it turns out that RHEL very
recently backported this libc fix into RHEL 8.

It's likely that Debian/Ubuntu will follow.

On Thu, Jun 9, 2022 at 7:45 AM Adam Isaacson  wrote:

> Well done, Morag and team! That is a great catch and very helpful to me
> and others!
>
> Kind regards,
>
> Adam
>
> On Thu, Jun 9, 2022 at 2:43 PM Morag Brown  wrote:
>
>> Hey all,
>>
>> I had a quick look in the mail archive and didn't see anything on this,
>> but it's not impossible others will come across/have come across the same
>> issue I've encountered, so figured I'd potentially save some people future
>> frustrations and heart attacks.
>>
>> There's an issue with later versions of Matlab (R2021a and up) on older
>> versions of Ubuntu (18.04 and down) where opening Simulink intermittently
>> causes Matlab to crash with some variation of the following error:
>>
>> *Inconsistency detected by ld.so: ../elf/dl-tls.c: 517:
>> _dl_allocate_tls_init: Assertion `listp != NULL' failed!*
>>
>> This is due to a bug in the glibc library present on older versions of
>> Linux distros that doesn't play nicely with the new way Matlab does some
>> things. Fortunately, Mathworks doth provide an answer. Not so fortunately,
>> their official answer is "upgrade to a newer operating system with a
>> version of glibc that doesn't have this bug". Lol.
>>
>> But they also kindly provide a solution
>>  for those of
>> us who can't just upgrade; an unofficial patch of the glibc libraries that
>> fixes this bug for various OS versions. This of course comes with the major
>> caveat that modifying system files is done at your own risk - all processes
>> in the OS share glibc, so the patch will affect your whole system, not just
>> Matlab, and this could cause weirdness.
>>
>> So if you find yourself in the same position I was yesterday, wondering
>> if you should risk it all just for Simulink - I'm happy to report that I
>> applied this patch, the universe did not end and nothing on my machine
>> *seems* to be broken. Great success.
>>
>> That is all,
>> Morag
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "casper@lists.berkeley.edu" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to casper+unsubscr...@lists.berkeley.edu.
>> To view this discussion on the web visit
>> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAGH-0Tc-dJ0AUnC21S5oTP-pRXyhMC_Ld%2B8KyQ31NuSFZ7HfkQ%40mail.gmail.com
>> 
>> .
>>
> --
> You received this message because you are subscribed to the Google Groups "
> casper@lists.berkeley.edu" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to casper+unsubscr...@lists.berkeley.edu.
> To view this discussion on the web visit
> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CADTJ%3DnGPD6cV7Vv0XuMVAezU%3D755iXVkMNzw%3DtGgc6Mc%3D_aTZQ%40mail.gmail.com
> 
> .
>


-- 
Greg Lindahl
Software Architect, Event Horizon Telescope
Smithsonian Astrophysical Observatory
60 Garden Street | MS 66 | Cambridge, MA 02138

-- 
You received this message because you are subscribed to the Google Groups 
"casper@lists.berkeley.edu" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to casper+unsubscr...@lists.berkeley.edu.
To view this discussion on the web visit 
https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAJT0TuNT%3Da2EsjgLXBB%2BcVjOJ_eQkS0CQ_ovbxq7Oq9irih9XA%40mail.gmail.com.

Re: [casper] Fix - Simulink R2021a crashes on startup on Ubuntu 18.04

2022-06-09 Thread Adam Isaacson 
Well done, Morag and team! That is a great catch and very helpful to me and
others!

Kind regards,

Adam

On Thu, Jun 9, 2022 at 2:43 PM Morag Brown  wrote:

> Hey all,
>
> I had a quick look in the mail archive and didn't see anything on this,
> but it's not impossible others will come across/have come across the same
> issue I've encountered, so figured I'd potentially save some people future
> frustrations and heart attacks.
>
> There's an issue with later versions of Matlab (R2021a and up) on older
> versions of Ubuntu (18.04 and down) where opening Simulink intermittently
> causes Matlab to crash with some variation of the following error:
>
> *Inconsistency detected by ld.so: ../elf/dl-tls.c: 517:
> _dl_allocate_tls_init: Assertion `listp != NULL' failed!*
>
> This is due to a bug in the glibc library present on older versions of
> Linux distros that doesn't play nicely with the new way Matlab does some
> things. Fortunately, Mathworks doth provide an answer. Not so fortunately,
> their official answer is "upgrade to a newer operating system with a
> version of glibc that doesn't have this bug". Lol.
>
> But they also kindly provide a solution
>  for those of us
> who can't just upgrade; an unofficial patch of the glibc libraries that
> fixes this bug for various OS versions. This of course comes with the major
> caveat that modifying system files is done at your own risk - all processes
> in the OS share glibc, so the patch will affect your whole system, not just
> Matlab, and this could cause weirdness.
>
> So if you find yourself in the same position I was yesterday, wondering if
> you should risk it all just for Simulink - I'm happy to report that I
> applied this patch, the universe did not end and nothing on my machine
> *seems* to be broken. Great success.
>
> That is all,
> Morag
>
> --
> You received this message because you are subscribed to the Google Groups "
> casper@lists.berkeley.edu" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to casper+unsubscr...@lists.berkeley.edu.
> To view this discussion on the web visit
> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAGH-0Tc-dJ0AUnC21S5oTP-pRXyhMC_Ld%2B8KyQ31NuSFZ7HfkQ%40mail.gmail.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"casper@lists.berkeley.edu" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to casper+unsubscr...@lists.berkeley.edu.
To view this discussion on the web visit 
https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CADTJ%3DnGPD6cV7Vv0XuMVAezU%3D755iXVkMNzw%3DtGgc6Mc%3D_aTZQ%40mail.gmail.com.