On Mon, Jul 20, 2009 at 03:38:06PM +0100, Steve Cotton wrote:
package apt
severity 535218 important
merge 295051 535218
thanks
That is no solution. libapt should never ever ever ever
segfault.
[..]
That is likely to make this segfault unreproducible. It's not a
fix to this bug; but
Hans-J. Ullrich hans.ullr...@loop.de writes:
Workaround
Just enter this into /etc/apt/apt.conf.d/20archive:
APT::Cache-Limit 1;
Worked well for me.
Otherwise: You are right, there is NO apt.conf any more!
Good luck!
Hans
That is no solution. libapt should never ever ever
package apt
severity 535218 important
merge 295051 535218
thanks
That is no solution. libapt should never ever ever ever
segfault.
JackYF wrote a patch that completely replaces the Dynamic MMap
code (it's discussed in #474947).
That is likely to make this segfault unreproducible. It's not a
Workaround
Just enter this into /etc/apt/apt.conf.d/20archive:
APT::Cache-Limit 1;
Worked well for me.
Otherwise: You are right, there is NO apt.conf any more!
Good luck!
Hans
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe.
Same problem here. I've just increased APT::Cache-Limit and the
problem's gone, but I think too that the problem must be fixed in repo
(just changing the default) or provide users better messages in case of
too small amount of cache (not just SEGV).
This problem is caused in amd64 by i386
But apt should increases its default size so most people can stick
with that even with ia32-apt-get installed.
Aye, I think that's the important thing. The common user will be
terrified by the warnings or segfaults, and at the same time they are
well likely to be wanting ia32-libs-tools
6 matches
Mail list logo
