Re: When will ZFS become stable?

From: Ivan Voras <ivoras_at_freebsd.org>
Date: Sun, 06 Jan 2008 17:45:01 +0100
Kris Kennaway wrote:

> No, clearly it is not enough 

This looks like we're constantly chasing the "right amount". Does it 
depend so much on CPU and IO speed that there's never a generally 
sufficient "right amount"? So when CPU and drive speed increase, the new 
amount will always be some bigger value?

>(and you claimed previously to have done 
> more tuning than this). 

Where? What else is there except kmem tuning (including KVA_PAGES)? IIRC 
Pawel said all other suggested tunings don't do much.

> I have it set to 600MB on the i386 system with 
> a 1.5GB KVA.  Both were necessary.

My point is that the fact that such things are necessary (1.5 GB KVA os 
a lot on i386) mean that there are serious problems which aren't getting 
fixed since ZFS was imported (that's over 6 months ago).

I see you've added to http://wiki.freebsd.org/ZFSTuningGuide; can you 
please add the values that work for you to it (especially for KVA_PAGES 
since the exact kernel configuration line is never spelled out in the 
document; and say for which hardware are the values known to work)?

> ZFS already tells you up front that it's experimental code and likely to 
> have problems.  

I know it's experimental, but requiring users to perform so much tuning 
just to get it work without crashing will mean it will get a bad 
reputation early on. Do you (or anyone) know what are the reasons for 
not having vm.kmem_size to 512 MB by default? Better yet, why not 
increase both vm.kmem_size and KVA_PAGES to (the equivalent of) 640 MB 
or 768 MB by default for 7.0?

 >Users of 7.0-RELEASE should not have unrealistic
 > expectations.

As I've said at the first post of this thread: I'm interested in if it's 
ever going to be stable for 7.x.



Received on Sun Jan 06 2008 - 15:45:13 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:25 UTC