Hi Fred
I am glad that your problem is solved but I am still not sure what it was you saw -
you used to have two peaks, when you had the problem you still had two peaks but the peak that used to come out first came out second and the peak that used to come out second came out first, in other words the peaks swapped retention times - yes ?
Were you recognising the peaks on the basis of their size and/or shape by any chance ? If so, then the problem becomes that the peak that used to be the bigger of the two is now the smaller of the two, and the peak that used to be the smaller is now the bigger. This is a lot easier to explain than two peaks swapping position to each other's retention times - the composition of the sample had changed.
Peter