Multiplex Redo Logs with Mirrored Disks?

  • From: Paul Drake <bdbafh@xxxxxxxxx>
  • To: "oracle-l@xxxxxxxxxxxxx" <oracle-l@xxxxxxxxxxxxx>
  • Date: Tue, 23 Nov 2004 12:19:47 -0500

> Anyone disagree?
yes.

I don't multiplex online redo logs on my laptop, nor on destructo boxes.
On dev, qa, prod, I would not use less than 2 members per redo log group.

How much do you trust your RAID controllers, drivers, hard drives? (I don't)
How much do you trust your FCHBAs, SAN switches, management software,
SAN admins?
Have you ever seen an entire controller channel go offline at a client
site? (yes, I have)
Have you ever seen an external storage unit become unavailable? (yes, I have)

3 copies of the control file, 2 redo log members per group and
duplexed archived redo log destinations - minimum for a real
production system.

Even where we have a mount point on a SAN in place, I keep a copy of
such files on local storage. External storage over fibrechannel can
and will go offline. External storage mounted over SCSI/RAID can and
will go offline. Its nice when that happens to have last night's hot
backup set, the archived redo logs, online redo log member and a
control file on local (internal) storage ... and say ... I can just
open things up local until you guys get the SAN sorted back out ... a
mirrored set of 146 GB drives will take care of such things quite
nicely.

Paul




On Tue, 23 Nov 2004 10:46:51 -0500, David Wagoner
<dwagoner@xxxxxxxxxxxxxxxxxx> wrote:
> Is there any need to multiplex redo logs with mirrored disks (e.g., RAID-1
> or RAID-1+0)?
> Example- Oracle recommends multiplexing redo logs on separate disks, like
> redo01a.log on Disk1 and redo01b.log on Disk2, etc.
>
> However, now that mirrored disks are in common use with RAID-1+0, RAID-1,
> etc. it seems that sufficient protection is in place to use only a single
> copy of each redo log.  This would also provide the benefit of reduced disk
> I/O to write redo information to disk.
>
> Anyone disagree?
>
> Best regards,
>
> David B. Wagoner
> Database Administrator
>
> --
> //www.freelists.org/webpage/oracle-l
>
--
//www.freelists.org/webpage/oracle-l

Other related posts: