On Sat, 30 Mar 2002, Neil Brown wrote: > The current raid5 requires all requests that is recieves to be the > same size. If it gets requests of different sizes, it does cope but > it needs to flush its stripe cache and start again. This effectively > serialises all requests around a request-size change. > So it is probably fair to say that what you are trying to do is > currently not a supported option. It may be in 2.6 if we ever get > raid5 working again in 2.5 :-) i know the example i gave was kind of silly (growing a fs while also making a snapshot of it and mounting the snapshot). but fwiw, i also just experienced the problem while i had a live filesystem, and i was creating a second logical volume and doing mke2fs on the second logical volume. this is an itsy bit more common usage scenario i think, but still not terribly common i suppose. it just means that some "online" operations are really not a good idea because of the raid cache size issues. > You could simply remove the printk and recompile and it might go a bit > faster, but you would still be loosing lots of performance due to lost > opportunities for concurrency. i could probably also try to go bang e2fstools over the head and make them agree with the ext3 kernel code access sizes :) -dean - To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
