Strange LilyPond segfault error after update

creatid · 2010-08-20 20:56:35

Since my last update, I get a segmentation fault when I use LilyPond to compile sheet music.
Now LilyPond wasn't included in the upgrade, so I guessed it might have something to do with guile, python, tetex or ghostscript (that LilyPond depends on).
So in order to find out more about the problem (after visiting the forums of course!) I ran LilyPond with the -V parameter for verbose output. Guess what: it compiled successfully and generated the .pdf ! Remove the -V and it segfaults again.

Again, it is my guess that it has nothing to do with LilyPond.
Anyone experience something similar?
Any solutions?

Thanks!

P.S. I reinstalled elfutils and glibc just to be sure since I read about some issues with them on the forum, but alas: no solution...

Last edited by creatid (2010-08-20 21:10:32)

karol · 2010-08-20 21:00:58

Are you sure it segfaults w/o creating a valid pdf when run w/o the '-V' switch?

creatid · 2010-08-20 21:06:10

Yes, I am sure.
It segfaults when "Preprocessing graphical objects..." so it doesn't get to the layout stages, the .ps creation or the .pdf conversion.

karol · 2010-08-20 21:08:52

Maybe you should open a bug report.
And *please* add 'LilyPond' to this thread's title.

creatid · 2010-08-20 21:13:23

karol wrote:

And *please* add 'LilyPond' to this thread's title.

I left that out intentionally because I think the problem is with one of the dependencies (possibly guile or python) so it might effect more people than just LilyPond users. But I have added it

Now of to figuring out how to file (my first) bug report.

karol · 2010-08-20 21:23:48

creatid wrote:

karol wrote:
And *please* add 'LilyPond' to this thread's title.
I left that out intentionally because I think the problem is with one of the dependencies (possibly guile or python) so it might effect more people than just LilyPond users. But I have added it
Now of to figuring out how to file (my first) bug report.

I understand why you didn't want to associate it with LilyPond, but are you having any other problems with python, guile, tetex etc.? So far we know that LilyPond segfaults w/o the '-V' parameter so we start from here.
'error after update'-type titles are unhelpful as it's not immediately obvious what is this error concerning and it's easier to search for 'searchterm' just in topic titles as opposed to whole threads as it produces far less false positives.

creatid · 2010-08-20 21:30:34

karol wrote:

'error after update'-type titles are unhelpful as it's not immediately obvious what is this error concerning and it's easier to search for 'searchterm' just in topic titles as opposed to whole threads as it produces far less false positives.

I fully agree with you here. I just realised that I often tend to skip those posts myself as well...

karol · 2010-08-20 21:45:39

creatid wrote:

karol wrote:
'error after update'-type titles are unhelpful as it's not immediately obvious what is this error concerning and it's easier to search for 'searchterm' just in topic titles as opposed to whole threads as it produces far less false positives.
I fully agree with you here. I just realised that I often tend to skip those posts myself as well...

https://bbs.archlinux.org/viewtopic.php?id=103268 - the title says " Something i've been wondering." .-)

As for the bug report, you can basically paste your first post, just make it a bit less verbose (hmmm, a nice situational pun), less conversational in tone.

Last edited by karol (2010-08-20 21:47:21)

creatid · 2010-08-20 21:56:24

karol wrote:

As for the bug report, <snip>

I will wait with that one until tomorrow. If I remember correctly, the computer in my studio didn't have the problem and I updated it at the same time.
Thanks for your advice anyway!

Last edited by creatid (2010-08-20 21:56:58)

creatid · 2010-08-20 23:10:49

Update
This is getting stranger by the minute...
In order to make sure that the segfault wasn't caused by my LilyPond code, I commented out everything I had added since the update. And indeed: no segfault. So I uncommented line by line to find out where the offending code might be. In the end, I had everything back as it was when the trouble started, only this time without the segmentation fault!

I am very sure that I didn't change as much as a single character in my source code.
Therefore I am not yet marking this thread as [Solved], seeing that there are some other threads about segmentation faults after recent updates as well. I have a suspicion these might be related.

Arch Linux

#1 2010-08-20 20:56:35

Strange LilyPond segfault error after update

#2 2010-08-20 21:00:58

Re: Strange LilyPond segfault error after update

#3 2010-08-20 21:06:10

Re: Strange LilyPond segfault error after update

#4 2010-08-20 21:08:52

Re: Strange LilyPond segfault error after update

#5 2010-08-20 21:13:23

Re: Strange LilyPond segfault error after update

#6 2010-08-20 21:23:48

Re: Strange LilyPond segfault error after update

#7 2010-08-20 21:30:34

Re: Strange LilyPond segfault error after update

#8 2010-08-20 21:45:39

Re: Strange LilyPond segfault error after update

#9 2010-08-20 21:56:24

Re: Strange LilyPond segfault error after update

#10 2010-08-20 23:10:49

Re: Strange LilyPond segfault error after update

Board footer