Re: Multiplexing redologs - is there still need for it?

"Howard J. Rogers" <hjr_at_dizwell.com> wrote in message news:<40520c09$0$31901$afc38c87_at_news.optusnet.com.au>...
> "Dusan Bolek" <pagesflames_at_usa.net> wrote in message
> news:1e8276d6.0403120815.155dbab6_at_posting.google.com...
> > Today, I have run into an interesting discussion. The question was if
> > there is a need for multiplexing redologs by database (as suggested in
> > Admin manual) if this data are already mirrored (using RAID, disk
> > arrays mirrors, Data Guard, SRDF etc.).
> > I'm somehow paranoid, so I proposed tu use multiplexing, while other
> > party stated that these bytes are already stored on eight different
> > locations. That's sounds sensible, but I remember the issue a long
> > time ago in a company far far away, where one of two multiplexed
> > control files got corrupted and (of course) copies on both mirrored
> > disc vere identical (means corrupted), second redo log on different
> > volume group was OK. So in this case we would been doomed if these
> > redologs weren't multiplexed.
> > What's the opinion on this topic in the newsgroups? Preferably with
> > some technical explanation.
>
>
> Somewhat fascistically, I suspect, but the topic ought not even to be
> discussed. Hardware RAID protects you from hardware failure. It does not and
> cannot protect you from (oracle) software error or (more common) user error.
> Only multiplexing can do that.
>
> Multiplexing should be compulsory. And discussions on whether that's a good
> idea or not ought to be banned.

Few points:
In the beginning I want to apologize for mentioned "control files" in the post, it was my stupid mistake, everything was about REDOLOGs. Sorry again. And yes, I know what's the difference. :-)

I was not only talking about hardware RAID, but also about Data Guard, which ships redolog transaction to another safe location not using hardware, but using Oracle services.

As I thought about it, I think the point is if there is a real possibility that Oracle (with HW assistance) can write one redolog right and other corrupted. I have seen it with my own eyes, but that was on 8.0.4, years before and also on a very inferior hardware. Sybrand mentioned two physical writes and one of them (probably on HW level) went wrong. Well, that's possible, but If you have DG in place, you should be protected against this (primary fails, but secondary got all transactions right).

And for the last HJR's line, I think that sometimes is good to open even "banned discussion". :-)

--
Dusan Bolek