To be able to correctly recognize the genome and build of your BAM files GenomeBrowse requires that the corresponding assembly file for the build contain the recognized "names" of the chromosomes so the correct reference sequence can be used to compute the index and coverage files necessary for plotting.
Fixing Gabriel's issue required that we edit the provided assembly file for the sheep (OAR3.1) genome so that it included OAR1, OAR2, etc. as acceptable names for the chromosomes, we then shipped that version of the assembly file with the next release of GenomeBrowse. So the fix was specific to his issue and not universal to all genomes.
If you could provide the species and build for your data we can provide you with an adjusted assembly file to see if that fixes your issue. I can also provide instructions on how you can make these edits yourself if you would prefer.