Hello,
On 12/20/06 19:36, Mark Price wrote:
On 12/20/06, *Klaus Darilion* < klaus.mailinglists(a)pernau.at
<mailto:klaus.mailinglists@pernau.at>> wrote:
Hi Mark!
Postgres should work well - I use it since ser 0.8. Just make sure
that
the hard disk does not get full, because this breaks the index inside
postgres and postgres is getting real slow (re-create the index if it
happens).
Last time I tested openser_postgres.sh it worked fine. If you find
a bug
please let us know.
Please check to use the latest versions from CVS (for 1.1 use CVS
rel_1_1_0)
The latest version from cvs doesn't work out of the box with openser
build from the release tarball, because the release tarball doesn't
included /usr/sbin/openser_gen_ha1 (although I could get past this by
replacing it with the md5sum equivelant).
However, the release tarball doesn't work because of this bug:
https://sourceforge.net/tracker/?func=detail&atid=743020&aid=151873…
<https://sourceforge.net/tracker/?func=detail&atid=743020&aid=1518732&group_id=139143>
So, 1. is the cvs version of rel_1_1_0 considered safe for production use?
yes, it is the most recommended version to use -- several issues were
found and fixed since 1.1.0 release.
2. if so and there are important fixes such as
this in CVS,
would
openser.org <http://openser.org> consider
releasing a point release containing such changes?
It is an option, and
perhaps should be taken in consideration, as patch
release with not so much packaging.
This may help some which want to maintain platforms updated, it is
exampled with 1.0.x, bust just replace the version with 1.1.0 in the
document and should work (rel_1_0_0 => rel_1_1_0)
http://openser.org/dokuwiki/doku.php/install:openser-from-cvs
Cheers,
Daniel
Thanks,
Mark Price
------------------------------------------------------------------------
_______________________________________________
Users mailing list
Users(a)openser.org
http://openser.org/cgi-bin/mailman/listinfo/users