[Bioclipse-devel] bugzilla etc

[Egon W. <ego...@gm...>]

Hi all,

while not optimal, this might be the best suboptimal solution we can
find... after some more digging and trying by Samuel (mostly) and me,
I think we have a better idea of why the schema went wrong... one
source of problems is custom fields, which we seemed to have many...
but with the excellent investigative work by Samuel, we at least have
a read-only copy of the bug reports... not all annotation seems
visible, but at least we can see the reports... this is running
Bugzilla 3.2, using the bugzilla3 database in MySQL.

One custom field that seems to have been lost in practice is the
milestone... (it's these custom fields which, seemingly including
those suggested by BZ itself) which breaks a proper upgrading... :(

I am still hoping we can get it to work a slightly bit better, but set
up a second Bugzilla, version 3.6, using the bugs database in MySQL.
Both versions are using the downloaded tar.gz, and not the Ubuntu
versions.

The latter version is a clean version, so everyone would have to get a
new account (inconvenient), but has some upsides too, like a more
graphically pleasing GUI, such as on this page:

http://pele.farmbio.uu.se/bugzilla36/showdependencygraph.cgi?id=1

So, my suggestion is to accept that we will not be able to get the old
Bugzilla3.2 fully operational without considerable more resource
spending, and keep that one around for RO, and use the Bugzilla3.6 for
new bug reports, which would also mean that bugs.bioclipse.net will
have to point to the new URL http://pele.farmbio.uu.se/bugzilla36/...

Comments on having Pelezilla + Pelezilla 2.0 ?

Egon

-- 
Dr E.L. Willighagen
Post-doc @ Uppsala University (only until 2010-09-30)
Proteochemometrics / Bioclipse Group of Prof. Jarl Wikberg
Homepage: http://egonw.github.com/
LinkedIn: http://se.linkedin.com/in/egonw
Blog: http://chem-bla-ics.blogspot.com/
PubList: http://www.citeulike.org/user/egonw/tag/papers