On Sat, 2007-02-10 at 10:49 +1100, Helen Borrie wrote:

>
> I haven't heard of any stability issues with running 32-bit Fb 1.5.3
> on AMD64 Linux,

Further information and research has shown this is stemming from a
previous post and thread to this list, as well as the dev one.

http://tech.groups.yahoo.com/group/firebird-support/message/75536

Where a reply from Ann Harrison kinda hints that there are some known
issues with Firebird 1.5 on amd64. Then it's also stated that it's a bad
idea to compile Firebird from source on amd64. I am not sure if this is
due to later mention of Fb not making heavy use of large memory systems.

I am just trying to get some sort of official confirmation from upstream
(you all). Since we as downstream must make decisions based on upstreams
recommendations.

Is the official stance that the Firebird team prefers people to only run
official binaries, and not run a version of Firebird compiled from
official sources?

If that is the case, does it apply to just amd64, or all architectures?

I would also like to state and make it clear that the original thread
(link provided above) is asking about the package being masked and
umasking. They are not stating they tried to use the package, or it has
issues, or they ran into problems.

The reasons for the mask on Gentoo in the first place, I mentioned in my
first post. Which seem to be false in the first place. Having run it for
some time. I would like to get some confirmation of any known issues.
Otherwise I think we have clear cases of it being stable on amd64 Gentoo
Linux.

Which if there is anything we as downstream can do to help upstream with
amd64 related issues or other architectures. Please let us know. Part of
our goal with compiling from source with all packages is tighter
integration with upstream. At least being able to provide testing
feedback, patches, or etc to improve an open source application.

--
William L. Thomson Jr.
Gentoo/Java


[Non-text portions of this message have been removed]