Hello all,
during the migration of old hardware to new one, I've incurred in a unpredictable problem: data size from user grow up a lot ( but I've been assured this is a temporary condition).
So I have a filepool policy to move all data ( and snapshot ) from old pool to new one, and is running from almost 30 days.
The old pool is quite empty, but the new one is almost full.
When the smartpool job would be finished, I will probably be, on new pool, at 98%.
My idea was to cancel the smartpool job, modify the filepool policy to permit write "Anywhere" which should not move the actual data from the actual pool but only permit to use the old pool ( and probably onefs would use this old one to write new data as it is almost free ) and as soon as the unpredicted load will end start it again.
I'm not sure what will happens, though, if I cancel the actual smartpool job.
Any experience on this ?
Or is better to create a new policy to move some old data ( by modify time, for example ) to old pool ?
And if I do this, would OneFS allow to have two smartpool job running or I have to cancel the actual anyway ( or wait for his termination ) ?
Thanks in advance for any hint
Pierluigi