[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#961345: cups: daemon crashes with invalid free()



Hello Ronny,


> Incidentally, stopping the cups service (new packages) after a single print job when under valgrind gave this in case it's related:
> 
> Aug 28 10:03:59 samba-prn-01.graysofwestminster.co.uk systemd[1]: Stopping CUPS Scheduler...
> Aug 28 10:04:00 samba-prn-01.graysofwestminster.co.uk valgrind[5238]: ==5238== Invalid free() / delete / delete[] / realloc()
> Aug 28 10:04:00 samba-prn-01.graysofwestminster.co.uk valgrind[5238]: ==5238==    at 0x48369AB: free (vg_replace_malloc.c:538)
> Aug 28 10:04:00 samba-prn-01.graysofwestminster.co.uk valgrind[5238]: ==5238==    by 0x4C73629: check_free (dlerror.c:202)
> Aug 28 10:04:00 samba-prn-01.graysofwestminster.co.uk valgrind[5238]: ==5238==    by 0x4C73629: check_free (dlerror.c:186)
> Aug 28 10:04:00 samba-prn-01.graysofwestminster.co.uk valgrind[5238]: ==5238==    by 0x4C73AB1: free_key_mem (dlerror.c:221)
> Aug 28 10:04:00 samba-prn-01.graysofwestminster.co.uk valgrind[5238]: ==5238==    by 0x4C73AB1: __dlerror_main_freeres (dlerror.c:239)

This might be what is described in here:
  https://sourceware.org/bugzilla/show_bug.cgi?id=24476
And I guess not related to the original issue.

At least could not reproduce this message
with libc6 2.31-3 in a up-to-date testing VM.


> So running with the patched cups packages seems to fix the "invalid free" on a test print. I've restored the systemd service file to remove valgrind so let's see how we go on a day's printing. :-).

Any news in this regard?


Kind regards,
Bernhard


Reply to: