[...] > This appears to be an issue with imap_sort() and a workedaround has > been implemented in HEAD and the forthcoming IMP 4.0.4. > Thanks, now it seems to be OK. regards, Richie