From: Michael Monnerie <michael.monnerie@is.it-management.at>
To: xfs@oss.sgi.com
Subject: Re: XFS hangs and freezes with LSI 9265-8i controller on high i/o
Date: Thu, 14 Jun 2012 09:31:20 +0200 [thread overview]
Message-ID: <9309089.RkWiC1Z3da@saturn> (raw)
In-Reply-To: <4FD8BBDF.8060503@hardwarefreak.com>
[-- Attachment #1.1: Type: text/plain, Size: 1451 bytes --]
Am Mittwoch, 13. Juni 2012, 11:12:15 schrieb Stan Hoeppner:
> This is the LSI/3ware firmware default policy Michael, so every user
> is safe out-of-the-box. If the BBU/FBU is not present (never
> installed), or the control logic determines it is not functioning
> properly, the firmware disables writeback caching.
>
> In absence of paying attention to logs/alerts, one will know pretty
> quickly when the BBU has failed, as write performance with many/most
> workloads will fall off a cliff.
That's good, and I guess every serious raid with bbu behaves the same. I
just wanted to make a statement because many people read "maximum
performance" and want this then, without understanding the downsides aka
"all data can be gone if a crash destroys the right metadata".
I get calls from people then whining, and they expect me not yell at
them but be friendly, which I don't like ;-)
I prefer to not speak about max perf if it means "probably eats your
data". Maybe we should always write "set this for max performance" only
with secure values, and then extra "and set this for turbo boost, but it
eats your data on a crash". Hopefully this helps keeping people from
just reading "max perf" and forget about the "eats your data" part.
--
mit freundlichen Grüssen,
Michael Monnerie, Ing. BSc
it-management Internet Services: Protéger
http://proteger.at [gesprochen: Prot-e-schee]
Tel: +43 660 / 415 6531
[-- Attachment #1.2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]
[-- Attachment #2: Type: text/plain, Size: 121 bytes --]
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
next prev parent reply other threads:[~2012-06-14 7:31 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-11 21:37 XFS hangs and freezes with LSI 9265-8i controller on high i/o Matthew Whittaker-Williams
2012-06-12 1:18 ` Dave Chinner
2012-06-12 15:56 ` Matthew Whittaker-Williams
2012-06-12 17:40 ` Matthew Whittaker-Williams
2012-06-13 0:12 ` Stan Hoeppner
2012-06-13 1:19 ` Dave Chinner
2012-06-13 3:56 ` Stan Hoeppner
2012-06-13 8:54 ` Matthew Whittaker-Williams
2012-06-13 11:59 ` Andre Noll
2012-06-13 12:13 ` Michael Monnerie
2012-06-13 16:12 ` Stan Hoeppner
2012-06-14 7:31 ` Michael Monnerie [this message]
2012-06-14 0:04 ` Dave Chinner
2012-06-14 14:31 ` Matthew Whittaker-Williams
2012-06-15 0:16 ` Dave Chinner
2012-06-15 9:52 ` Michael Monnerie
2012-06-15 12:29 ` Dave Chinner
2012-06-15 11:25 ` Bernd Schubert
2012-06-15 12:30 ` Dave Chinner
2012-06-15 14:22 ` Bernd Schubert
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=9309089.RkWiC1Z3da@saturn \
--to=michael.monnerie@is.it-management.at \
--cc=xfs@oss.sgi.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox