The 'odd' areas are, like most people are pointing out, due to the fixer not reaching the areas affected. This is a little bit odd as the developer has clearly managed to reach the area of the emulsion that is 'odd' because the text and frame numbers are developed. The printing is done with a light based system at the time of manufacture, so perhaps the film simply takes a lot more 'fixing' than it does 'developing'.
Usually 135 film is loaded on a spiral with the emulsion facing the centre of the reel. The natural springiness of the film-base keeps it resting on the outer 'edge' of the spiral track. Is it possible that you loaded the spiral with the emulsion facing the outside? This would then cause the emulsion to be pushed against the edge of the spiral track and could easily give the result you have seen. If this was not the case, then damp patches on the spiral could have caused the film to 'stick' in those places - with plastic reels that would have made it very difficult to load but with steel reels it would even be possible to have loaded the film, left the reel in the tank to (partially) dry and 'glue' itself to the spiral, and then processed it later!
The way to remove the unfixed emulsion is to fix the roll again, with fixer that is still at a usable level of activity (not some ancient and dirty leftover fixer) and for twice the clearing time in the same way that you would usually. The film should, of course, then be washed in the usual way.