bedtools intersect with -g option: chromosomes without regions cause intersections being missed

[ChrissiKalk97]

I have used bedtools intersect for genomic files. As these were very large I had to go with the -sorted and the -g option.
The -g file I obtained in the following way:

cut -f1,2 /path/Homo_sapiens.GRCh38.dna.primary_assembly_110.fa.fai >
$out_path/genome_chrom_ordering.txt

For the two bedfiles I used the same chromsome ordering for sorting and intersected them in the following way:
bedtools intersect
-s
-wao
-a $bedfile1
-b $bedfile2
-sorted
-g $out_path/genome_chrom_ordering.txt
| sort -nr -k13,13
> $interesectBedfile

This seemed to work, however I noticed after a while that I could not find any intersections for the X and Y chromosome which was weird. It turned out that one of my bedfiles did not have a region for the MT chromosome (being in front of X for the chromosome ordering). This resulted in intersections after the MT chromosome to be missed. I then included dummy regions in the bedfile missing the MT chromosome (null regions) and now the intersection is working also for the X and Y chromosome. I do not think that this behavior is desired, so I report my observations here.