Problem when simulate a simple graph

[slbai01]

To track the minigraph behaviour, I design a simple graph which contains some INS/DEL and INV.

for x in `seq 1 500000`; do echo $(( ( RANDOM % 4 ) + 1 )); done | sed 's/1/A/g' | sed 's/2/T/g' | sed 's/3/C/g' | sed 's/4/G/g' | tr '\n' ' ' | sed 's/ //g'  > seg1.fa
for x in `seq 1 350000`; do echo $(( ( RANDOM % 4 ) + 1 )); done | sed 's/1/A/g' | sed 's/2/T/g' | sed 's/3/C/g' | sed 's/4/G/g' | tr '\n' ' ' | sed 's/ //g'  > seg2.fa
for x in `seq 1 410000`; do echo $(( ( RANDOM % 4 ) + 1 )); done | sed 's/1/A/g' | sed 's/2/T/g' | sed 's/3/C/g' | sed 's/4/G/g' | tr '\n' ' ' | sed 's/ //g'  > seg3.fa
for x in `seq 1 320000`; do echo $(( ( RANDOM % 4 ) + 1 )); done | sed 's/1/A/g' | sed 's/2/T/g' | sed 's/3/C/g' | sed 's/4/G/g' | tr '\n' ' ' | sed 's/ //g'  > seg4.fa
for x in `seq 1 460000`; do echo $(( ( RANDOM % 4 ) + 1 )); done | sed 's/1/A/g' | sed 's/2/T/g' | sed 's/3/C/g' | sed 's/4/G/g' | tr '\n' ' ' | sed 's/ //g'  > seg5.fa
for x in `seq 1 500000`; do echo $(( ( RANDOM % 4 ) + 1 )); done | sed 's/1/A/g' | sed 's/2/T/g' | sed 's/3/C/g' | sed 's/4/G/g' | tr '\n' ' ' | sed 's/ //g'  > seg6.fa
for x in `seq 1 450000`; do echo $(( ( RANDOM % 4 ) + 1 )); done | sed 's/1/A/g' | sed 's/2/T/g' | sed 's/3/C/g' | sed 's/4/G/g' | tr '\n' ' ' | sed 's/ //g'  > seg7.fa
for x in `seq 1 380000`; do echo $(( ( RANDOM % 4 ) + 1 )); done | sed 's/1/A/g' | sed 's/2/T/g' | sed 's/3/C/g' | sed 's/4/G/g' | tr '\n' ' ' | sed 's/ //g'  > seg8.fa
for x in `seq 1 520000`; do echo $(( ( RANDOM % 4 ) + 1 )); done | sed 's/1/A/g' | sed 's/2/T/g' | sed 's/3/C/g' | sed 's/4/G/g' | tr '\n' ' ' | sed 's/ //g'  > seg9.fa


echo ">seq1" > seq1.fa; cat seg1.fa seg2.fa seg3.fa seg4.fa seg5.fa seg6.fa >> seq1.fa 
echo ">seq2" > seq2.fa; cat seg1.fa seg7.fa seg8.fa seg2.fa seg4.fa seg6.fa >> seq2.fa 
echo ">seq3" > seq3.fa; cat seg1.fa seg9.fa seg3.fa seg5.fa seg4.fa seg6.fa >> seq3.fa

But when I run the command, only one sequence generates, this makes me confused.
I install the software through conda, and I used two version (0.19-r551 and 0.15-r426)

minigraph -cxggs -l10k seq1.fa seq2.fa seq3.fa   > sample.gfa && cut -f 1,2,4- sample.gfa
# S       s1      LN:i:2540000    SN:Z:seq1       SO:i:0  SR:i:0