BBR lvchange HFS

[sevi]

I want to do an lvchange on non-root HFS fs. Want to turn Bad Block Reallocation to None. It is currently been used by application like oracle. I'm pretty confident doing it on JFS, but has anyone seen any problems with changing this attribute on a HFS type. Concern about performance, apps don't work etc.

thks all!

[A. Clay Stephenson · Accepted solution]

Because this happens below the level of the filesystem, hfs should be as stable as vxfs. The one exception is that vxfs has the intentlog to add a bit more integrity. I question why you would want to disable the default behavior when it actually adds a bit of protection for only a very small decrease in total available disk space. If this is an array then you should have no bad blocks (at least visible to the host) and it makes sense. One thing that I note is that you mention hfs and performance. Vxfs filesystems have outperformed hfs filesystems for at least the last 5 or 6 years, so if performance is a concern you shouldn't be using hfs anyway.

[sevi]

thks Clay, these disk are part of array and vendor recommandation to change it to None. We've been having it as default for a while, and don't really have a prob, but vendor recommandation, so looking into it.

I agree regarding the HFS vs vxfs thing, which is something down the road.

thks again.

[Shyjith P K]

Hi,

You can try lvchange -r n

But upon a media failure, LVM will mark the failed block as bad in the Bad Block directory, but will NOT attempt to relocate the bad block to a new location on disk. Future I/O requests which contain the bad
block will return with an I/O error. No attempt will be made to access the bad block.


cheers
Shyjith

[SureshKumar_2]

Hi

if u use lvchange, you will not face any problem as of now.

*) umount the FS (umount )
*)lvchange -r n

But once if that particular disk goes bad, then I/O request will show an error encounted..... It will become very complex to recover data.

suresh