Re: ZFS pool balance and performance

From: Frank Leonhardt <freebsd-doc_at_fjl.co.uk>
Date: Sun, 24 Aug 2025 17:36:39 UTC
On 24/08/2025 15:41, Chris Ross wrote:
>> On Aug 24, 2025, at 08:11, Frank Leonhardt<freebsd-doc@fjl.co.uk>  wrote:
>>
>> It looks like one of the drives has been replaced. I've bought replacement drives of the same model only to discover they've changed to SMR - the array ran very badly until removed.
> Well, all of the disk have been replaced.  You mention one, but that may just
> be because one of them is partitioned rather than whole?
>
> And, da1-da7 are all exactly the same part number
> da1: <WDC WUH721414AL4200 A07G> Fixed Direct Access SPC-4 SCSI device

I think that's an WD Ultrastar and AFAIK all Ultrastars are CMR. In one 
instance I got caught out by Seagate Barracuda, which at one time was 
their performance drive.... but definitely not any more! I'm suspicious 
of all large drives, especially those from a few years ago when the 
manufacturers thought that if they didn't tell then no one would notice.





>> You also mentioned you were using NFS, but not what for other than reads. You might want to take a look at this if there are any synchronous writes going on:
>
> I’m almost not using NFS for anything other than reads.  Small writes
> maybe, inode (atime) updates, small files.  <snip> I should find a way to trace that
> activity?

The post gives some hints. Definitely turn off atime with ZFS unless you 
really need it. I really can't remember if this is something that causes 
a sync write over NFS or not, but it ain't good for fragmentation to CoW 
a directory block each time you read a file.

> Yeah, google search suggested there isn’t any way to rebalance.  Is there
> a way I can identify which vdev files are on?  If so, I could delete some
> files that are principally on raidz1-1, which would help some.

My understanding is that files may be stored across all vdevs. I don't 
remember there being any restriction that says a file should be on 
exactly one vdev; it kind of defeats the point.

I'm rather hoping someone working on current ZFS internals will come 
along and give a few definitive answers here!

Regards, Frank.