Trying to get a handle on what's going on with salvage on a pool/volume....

The PoolHighWaterMark is 10% and PoolLowWaterMark 5%.
Pool is 850GB, freespace is approx 9% (83GB), but purgeable space (freeablesize) is 217GB - i.e. actual space 'available' is 300GB

So, no salvage at the moment - presumably because actual free space is below the PoolHighWaterMark? But what about all the existing 'freeable' space? There appears to be a dead zone here, with old, purgeable data preventing newer deleted data from entering the salvage system...
What happened to Last In First Out, etc.?!?
Also, there appear to be contradictions as to when 'autopurge' kicks in... doc states lowwatermark, and then until reaches high...TID 3934260 states 'It will not start to automatically purge old files until the volume is 100% full, and it has to purge to be able to save new files.'... Any clarification on this?

Also, this pool is currently on an oes2sp2+ box, but started life out under NW65 (SAN LUN was re-assigned, etc.)

I wonder if this is a hangover from the 'early' NW65 issues with salvage, free space, etc. and whether a
'nss /poolrebuild=<pool_name> /purge'
is required to kick this into shape?

Any thoughts or experiences?


(Apologies for apparent duplicate - posted in OES-NW first, by mistake!)