Error with bowtie2: Saw ASCII character 10 but expected 33-based Phred qual

kneaddata v0.10.0

Issue 1:
I installed kneaddata v0.10.0 through conda. When I used one database as an index for bowtie2, everything ran fine except deleting the intermediate files containing the index name. Following are my scripts.

$ kneaddata -i /home/gfz3/mojf/rawdata/B11_1.fq.gz \
-i /home/gfz3/mojf/rawdata/B11_2.fq.gz \
-o /hdd/mojf/output/kneaddata \
-t 20 -p 20 \
-db /database/kneaddata_db/invertebrate_db/INVERTEBRATE \
--run-fastqc-start --run-fastqc-end \
--trimmomatic /home/gfz3/miniconda3/envs/kneaddata/bin/Trimmomatic-0.39 \
--trimmomatic-options "SLIDINGWINDOW:4:15 MINLEN:36" \
--remove-intermediate-output

Issue 2:
When I used two databases as indexes, bowtie2 ran with an error. The error message I got is as follows.

Saw ASCII character 10 but expected 33-based Phred qual.
terminate called after throwing an instance of 'int' Aborted (core dumped)
(ERR): bowtie2-align exited with value 134

I just added the second -db to the previous scripts and didn’t change anything else. Following are my scripts.

$ kneaddata -i /home/gfz3/mojf/rawdata/B11_1.fq.gz \
-i /home/gfz3/mojf/rawdata/B11_2.fq.gz \
-o /hdd/mojf/output/kneaddata \
-t 20 -p 20 \
-db /database/kneaddata_db/invertebrate_db/INVERTEBRATE \
-db /database/kneaddata_db/vertebrate_db/VERTEBRATE \
--run-fastqc-start --run-fastqc-end \
--trimmomatic /home/gfz3/miniconda3/envs/kneaddata/bin/Trimmomatic-0.39 \
--trimmomatic-options "SLIDINGWINDOW:4:15 MINLEN:36" \
--remove-intermediate-output

The first 8 lines of B11_1.fq.gz are as follows.

@LH00391:13:22F2NMLT3:5:1101:8055:1048 1:N:0:GAGCAAGATC+AAGTTACGGA
CNGACAACAACAGTGAGAAAAAAATAAAATAGCAGCCAAGAAGTACCGGAATACTACATCCATTGGTATCCAAGACGTCTGCGTGCAAGCACTTAGAAGAACGGGGGCAAAGTTTGAAAATATGGCCAAGTTCGATACCCTTGGGGAACT
+
I#IIIII9III-III-I--999999---------9---9-9---9--9999------9-9-9---9-9I9I99999-I-II99999-I9---I9999999--99--9-9-999-99999II9999--9-9-9I9999-99I--9--I9--
@LH00391:13:22F2NMLT3:5:1101:12326:1048 1:N:0:GAGCAAGATC+AAGTTGCGGA
TNAGAATCACGAGCAGCTAGCAAAACACGAAATAATAACAGCCGGAAATCCCAGGACCAAGGATGGACGCCAGCTGAGTCAAAGTCCTTCCATTCAACTACCCGCCGCCGGTCTGGTGACGCATAGCATCCAAAGTTACCAATATGAGCA
+
I#IIIIII9IIIIIIIIIIIIII-I-9--II-II99-9-9--9-9999-9999----99--99-I99-99-III---9--9I9999-9I----999---9-9----99-I-9--9-I-I-9-999----999-999999999-9-99999

The first 2 lines of invertebrate.fasta used for indexing are as follows. The 6 files with the suffix .bt2 as the output of index were complete.

>NC_004354.4 Drosophila melanogaster chromosome X
GAATTCGTCAGAAATGAgctaaacaaatttaaatcattaaatgcGAGCGGCGAATCCGGAAACAGCAACTTCAAACCAGT

I would like to know why the above two issues arise, especially the second one, and how to solve them. Looking forward to the solutions.