[Gmod-help] Gbrowse fails with too many tracks

[Grace]

Hi -

I've got two GBrowse-related questions and I'd appreciate it if you
can point me towards people who can help me troubleshoot them.

I've been having a problem with uploaded annotation.  When I upload
several annotation files to my GBrowse installation, the annotation
tracks display correctly at first, but after looking at a few
genes/etc., all but one of the tracks will stop showing.  These errors
show up in my apache logfiles.

[Sat Sep 18 15:32:55 2010] [error] [client ::1] Use of uninitialized
value $nr_tracks_added in numeric eq (==) at
/opt/local/lib/perl5/site_perl/5.10.1/darwin-multi-2level/Bio/Graphics/Browser.pm
line 1397., referer:
http://localhost/cgi-bin/gbrowse/genome/?name=genename
[Sat Sep 18 15:32:55 2010] [error] [client ::1] Use of uninitialized
value $nr_tracks_added in subtraction (-) at
/opt/local/lib/perl5/site_perl/5.10.1/darwin-multi-2level/Bio/Graphics/Browser.pm
line 1404., referer:
http://localhost/cgi-bin/gbrowse/genome/?name=genename

This is while running the original GBrowse on OS X 10.6.4; everything
else currently works without problem.  Is there anything I can do to
address this?

I've also got an unrelated bonus question regarding GBrowse2, which
I'm in the process of setting up (using a MySQL backend.)  The example
yeast and volvox databases work fine (again, using the MySQL backend);
the genome I am studying does not.   Permissions seem to be the same
on conf files, gff3 files, databases, etc.  The GFF3 file validates on
the EColiHub validator, and loads fine with bp_seqfeature_load.pl  The
MySQL database appears to be correctly populated  with the information
imported from the GFF3 file (meaning that with the correct genome
database loaded in MySQL, I can, for example, enter "select * from
sequence;" and get a list of all the sequences/matching IDs from the
sequence table.)

Looking for any valid landmark results in a "The landmark GeneName was
not found."  Only the search panel (containing the "landmark or
region", "examples", and "data source" options") appears; Overview,
Region, and Details all do not appear at all.  (The result is the same
as what you would get if you searched for some random invalid phrase
in a working setup (e.g. my yeast or volvox sample databases), but it
occurs even when I search for contig names and gene names/IDs that I
know exist in the GFF3 file/MySQL database.)  Nothing untoward shows
up in my MySQL error logs, and there's generally nothing odd in my
Apache logs either except for specific actions like trying to get
chromosome size (n.b. the genome in this case is the default source in
GBrowse.conf):

[Mon Sep 20 17:08:06 2010] [error] [client ::1] Can't find chromosome
chromname in default database at
/opt/local/lib/perl5/site_perl/5.10.1/darwin-multi-2level/Bio/Graphics/Browser2/DataLoader.pm
line 173., referer:
http://localhost/cgi-bin/gb2/gbrowse/genome/?name=genename

However, simply trying to load a normal GBrowse page does not produce
any errors, even though no tracks are visible and no landmarks are
found.

Again, any ideas?  I can send logs, gff3 files, conf files, etc. as
necessary.  I'd appreciate any help anyone can offer with regards to
either of these problems.
Thanks!

-Grace