The positions of the cytobands are available in : http://hgdownload.cse.ucsc.edu/goldenPath/hg18/database/cytoBand.txt.gz

curl -s "http://hgdownload.cse.ucsc.edu/goldenPath/hg18/database/cytoBand.txt.gz" | gunzip -c | grep acen | head
chr1    121100000    124300000    p11.1    acen
chr1    124300000    128000000    q11    acen
chr10    38800000    40300000    p11.1    acen
chr10    40300000    42100000    q11.1    acen
chr11    51400000    52900000    p11.11    acen
chr11    52900000    56400000    q11    acen
chr12    33200000    35400000    p11.1    acen
chr12    35400000    36500000    q11    acen
chr13    13500000    16000000    p11.1    acen
chr13    16000000    18400000    q11    acen

http://genome.ucsc.edu/cgi-bin/hgTables

Just choose All Tables -> table:gap

Centromeres are treated as a gap "type"

#bin    chrom    chromStart    chromEnd    ix    n    size    type    bridge
23    chr1    121535434    124535434    1270    N    3000000    centromere    no

Chromosome length can be accessed from chromInfo

#chrom    size    fileName
chr1    249250621    /gbdb/hg19/hg19.2bit
chr2    243199373    /gbdb/hg19/hg19.2bit
chr3    198022430    /gbdb/hg19/hg19.2bit
chr4    191154276    /gbdb/hg19/hg19.2bit
chr5    180915260    /gbdb/hg19/hg19.2bit
chr6    171115067    /gbdb/hg19/hg19.2bit

http://genome.ucsc.edu/FAQ/FAQtracks.html#tracks20

Question:
"How do I find the positions of the centromeres and telomeres in a particular assembly?"

Response:
This information can be found in the "gap" database table. Use the Table Browser to extract it. To do this, select your assembly and the gap table, then click the "filter Create" button. Set the "type" field to centromere telomere (separated by a space). For help using the Table Browser, visit the User's Guide.

I am also looking for this, and I find all what we need are included in package GWASTools https://www.rdocumentation.org/packages/GWASTools/versions/1.18.0/topics/centromeres

I have tried

curl -s "http://hgdownload.cse.ucsc.edu/goldenPath/dm3/database/cytoBand.txt.gz" | gunzip -c | grep acen | head

, but it didn't show any results.I also tried

http://genome.ucsc.edu/cgi-bin/hgTables

just choose All Tables -> table:gap

Centromeres are treated as a gap "type". The result is different from human genome, the type of the reads are contig not centromere.

#bin    chrom    chromStart    chromEnd    ix    n    size    type    bridge
594    chr4    1221288    1221388    138    N    100    contig    yes
587    chrU    325170    325270    141    N    100    contig    yes
588    chrU    429354    442354    143    N    13000    contig    yes
589    chrU    543815    543915    145    N    100    contig    yes
589    chrU    639309    639409    147    N    100    contig    yes

Lastly I tried the third method, it showed the message as follows:

#filter: (gap.type = 'telomere' OR gap.type = 'centromere')
#bin    chrom    chromStart    chromEnd    ix    n    size    type    bridge
# No results passing filter.

I face to one problem that I can not resolve it. I have a problem for drawing idiogram in sheep chromosomes (I analysis with software: http://visualization.ritchielab.psu.edu/phenograms/plot) because I must have sheep Centromeres Position and chr size in this graph but I can't find information about it.

You can pre-process this file http://hgdownload.cse.ucsc.edu/goldenPath/hg19/database/gap.txt.gz and extract the centromeric regions for hg19/GRCh37.