Rolloff - fatalx: too few snps

It shows too few snps; numsnps: 0 first and then shows numsnps as 753k. Any idea what might be wrong?

## rolloff.  Version 1100

parameter file: ped-eigen.par
genotypename: ./files/chr21_eur_natam_bgi_hli.geno
snpname: ./files/chr21_eur_natam_bgi_hli.snp
indivname: ./files/chr21_eur_natam_bgi_hli.ind
poplistname: ./files/pop.ref
admixlist: ./files/pop.target
binsize: 0.001
maxdis: 0.8
seed: 7
runmode: 1
zdipcorrmode: YES
checkmap: YES/
seed: 7
chithresh:     6.000
./files/chr21_eur_natam_bgi_hli.snp: genetic distance set from physical distance
snp order check fail; snp list not ordered: ./files/chr21_eur_natam_bgi_hli.snp (processing continues)    1 10113 1011278 
zzz 0 394075
snp order check fail; snp list not ordered: ./files/chr21_eur_natam_bgi_hli.snp (processing continues)    1 10204 1020428 
zzz 1 394076
snp order check fail; snp list not ordered: ./files/chr21_eur_natam_bgi_hli.snp (processing continues)    1 10214 1021403 
zzz 2 394077
snp order check fail; snp list not ordered: ./files/chr21_eur_natam_bgi_hli.snp (processing continues)    1 10388 1038818 
zzz 3 394078
snp order check fail; snp list not ordered: ./files/chr21_eur_natam_bgi_hli.snp (processing continues)    1 10398 1039813 
zzz 4 394079
snp order check fail; snp list not ordered: ./files/chr21_eur_natam_bgi_hli.snp (processing continues)    1 10548 1054842 
zzz 5 394080
snp order check fail; snp list not ordered: ./files/chr21_eur_natam_bgi_hli.snp (processing continues)    1 10559 1055892 
zzz 6 394081
snp order check fail; snp list not ordered: ./files/chr21_eur_natam_bgi_hli.snp (processing continues)    1 10775 1077546 
zzz 7 394082
snp order check fail; snp list not ordered: ./files/chr21_eur_natam_bgi_hli.snp (processing continues)    1 11097 1109721 
zzz 8 394083
genotypename:  ./files/chr21_eur_natam_bgi_hli.geno
fatalx:
too few snps; numsnps: 0
valid snps: 753136
group 0 EUR
group 1 AMR
number admixed: 120 number of references: 2
numsnps: 753136  numindivs: 319

Reproduce the report by running rolloff with the shown ped-eigen.par settings and input files. Start by reviewing the SNP-order warnings and the fatalx path that reports numsnps 0 alongside 753136 valid SNPs. Done means identifying the cause of the discrepancy and verifying the behavior with a regression check.