On Tue, 15 Feb 2000, Torsten Landschoff wrote: > Hi James, > > On Mon, Feb 14, 2000 at 09:41:25PM -0500, James R. Van Zandt wrote: > > > Judging from the mailing list archives, there has not been much > > Quality Assurance activity in support of Debian :-) I would like to > > Right. It seems nobody currently has enough free time. I would have but > I am organizing the CeBIT fair now :(
<aol>, but I'm trying to organise a DPL campaign :/ > > suggest a type of test that is comparably easy to perform: checking > > for bad memory references using electric-fence. Here is the relevant > > part of /usr/share/doc/electric-fence/README.Debian: > > Sorry, but this is not what debian-qa is going to do (I hope): There are > enough known bugs. We will have to fix them first before going to hunt > even more bugs. <fx: puts on electric-fence maintainer hat>IMO, you should be using efence for occasions when you think the bug to be memory-related. Even then, it's not infallible, particularly if you happen to be relying on another library which is buggy (I've had this problem before). Sure, people should use debugging tools for debugging, but I think it's better to look at packages which we know to be buggy in certain regards, rather than just trying to find victims... > Whoever has the time - go ahead. Let's see how much the BTS can take. > Still I would like to concentrate on the known bugs first. I agree with this sentiment entirely. Matthew -- "At least you know where you are with Microsoft." "True. I just wish I'd brought a paddle." http://www.debian.org
