I am trying to re-count a bunch of bam files that I have previously opened without problems in easyRNAseq and GenomicAlignments. But now I get. The bam files are seemingly valid and can be accessed by samtools, but the header contains few duplicated entries.

reads = readGAlignments('test.fastq.gz.clipped_M.bam.sorted.bam')

Error in GAlignments(seqnames = bamcols$rname, pos = bamcols$pos, cigar = bamcols$cigar,  :
  'seqlengths' must be an integer vector with unique names
In addition: Warning message:
In `levels<-`(`*tmp*`, value = if (nl == nL) as.character(labels) else paste0(labels,  :
  duplicated levels in factors are deprecated

This is a related thread:

Error when reading bam file with duplicated header

Sorry for the delay, but I believe this is now fixed in GenomicAlignments / Rsamtools 1.4.2 / 1.20.5 (release) and 1.5.18 / 1.21.19 (devel).