different output from bowtie2 and samtools flagstat

[anu18]

Hello, I am new to bioinformatics and trying to analyse paired end RAD seq data. I have aligned the sequences using bowtie2 and have my output as sorted bam after piping the results straight into samtools. the output that I get for this std. output is different than samtools flagstat. Here is the output from aligner
2170017 reads; of these:
2170017 (100.00%) were paired; of these:
608209 (28.03%) aligned concordantly 0 times
994001 (45.81%) aligned concordantly exactly 1 time
567807 (26.17%) aligned concordantly >1 times
----
608209 pairs aligned concordantly 0 times; of these:
49881 (8.20%) aligned discordantly 1 time
----
558328 pairs aligned 0 times concordantly or discordantly; of these:
1116656 mates make up the pairs; of these:
488912 (43.78%) aligned 0 times
232584 (20.83%) aligned exactly 1 time
395160 (35.39%) aligned >1 times
88.73% overall alignment rate

and here is the output from flagstat

2322547 + 0 in total (QC-passed reads + QC-failed reads)
0 + 0 secondary
0 + 0 supplementary
0 + 0 duplicates
2322547 + 0 mapped (100.00% : N/A)
2322547 + 0 paired in sequencing
1132936 + 0 read1
1189611 + 0 read2
2073092 + 0 properly paired (89.26% : N/A)
2225207 + 0 with itself and mate mapped
97340 + 0 singletons (4.19% : N/A)
110386 + 0 with mate mapped to a different chr
110386 + 0 with mate mapped to a different chr (mapQ>=5)

these two are very different. Can someone help me in reconciling them. Also when i try to find the uniquely mapped reads using samtools
samtools view -F 4 412_sorted.bam | grep -v "XS" | wc -l
2139690
I get this result which again is different from the bowtie output of aligned exactly 1 time. Can someone help me.