You were most likely doing a 'mkfs -b 4096' after creating an array and
while the array was still resyncing. Right?! The default block size of a
raid device is 512 bytes. A mkfs with '-b 4096' option changes the block
size to 4096 bytes. The resync operation starts using the default 512, but
then notices the block size change and adjusts to the change, but gives some
confusing messages as you have seen. Here is what happens...
1. You create the array
2. The array starts resyncing (using 512 byte blocks.)
3. You do a mkfs -b 4096
4. The block size for the md device is modified to 4096 by mkfs
5. Queued requests from the resync process generate the majority of the
error messages.
6. The resync process detects the block size change between doing groups of
buffers.
7. The resync process re-issues the requests using the new block size.
NO HARM DONE!!!
<>< Lance.
[EMAIL PROTECTED] wrote:
> I got lots of theese messages:
> set_blocksize: b_count 1, dev md(9,4), block 124792 ,
> from c013cd36
> md4: blocksize changed during write
> nr_blocks changed to 32 (blocksize 4096, j 31168,
> max_blocks 3840064)
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to [EMAIL PROTECTED]
