I upgraded my RAID-Z to new drives and the resilvering process went well, but now I have a "One or more devices are configured to use a non-native block size. Expect reduced performance." warning. It appears to be configured to 512B blocks, but the native size is 4096B. I saw some discussion of this on the forums, but was unclear: is there a way I could correct this on a drive level basis one drive at a time to maintain the existing pool? Or do I need to start over with a new pool to correct this? I suspect the latter is the simplest course of action, but after replacing a failed drive and enlarging the pool successfully at the same time, I'm sort of seeing it as a challenge I'd like to accomplish to get everything configured without simply wiping and starting over.
Incidentally, is there a way to guesstimate how much of a performance hit I'm taking with the non-native size? Maybe this is a lot of worry over nothing? I'm getting about 84MB/sec both read and write over GigE at the moment. It's been a long time since I tested the performance, but I thought I was closer to the upper threshold of GigE's real world performance the last time I tested it, years ago.
Incidentally, is there a way to guesstimate how much of a performance hit I'm taking with the non-native size? Maybe this is a lot of worry over nothing? I'm getting about 84MB/sec both read and write over GigE at the moment. It's been a long time since I tested the performance, but I thought I was closer to the upper threshold of GigE's real world performance the last time I tested it, years ago.
