All of lore.kernel.org
 help / color / mirror / Atom feed
From: Reindl Harald <h.reindl@thelounge.net>
To: d tbsky <tbskyd@gmail.com>, Phillip Susi <phill@thesusis.net>
Cc: Xiao Ni <xni@redhat.com>, list Linux RAID <linux-raid@vger.kernel.org>
Subject: Re: raid10 redundancy
Date: Fri, 7 May 2021 05:26:29 +0200	[thread overview]
Message-ID: <9b57e1bf-a573-e6a1-2db8-080ab05205b1@thelounge.net> (raw)
In-Reply-To: <CAC6SzHKKPCk4fOx7b2CxMWorPghRPMH3GD2v7vcC_YLKbDn7KA@mail.gmail.com>



Am 07.05.21 um 03:47 schrieb d tbsky:
> Phillip Susi <phill@thesusis.net>
>> No, it only depends on the number of copies.  They layout just effects
>> the performance.
> 
> I thought someone test the performance of two ssd, raid-1 outperforms
> all the layout. so maybe under ssd it's not important?
> 
>> No; 2 copies means you can lose one disk and still have the other copy.
>> Where those copies are stored doesn't matter for redundancy, only performance.
> 
>   I am trying to figure out how to lose two disks without losing two
> copies. in history there are time bombs in ssd (like hp ssd made by
> samsung) caused by buggy firmware.
> if I put the orders correct, it will be safe even the same twin ssd
> died together.

forget what you try to figure out, you can't control which two disks 
fail at the same time, period

you can minimize the risk by using different disk types and/or 
distribute their age (i am doing that currently by replacing all disks 
on a backup server with two weeks between)

in case of a frimware timebomb you have no chance in case of identical 
disks and when you seek a soluton whuch survives two failures then for 
the sake of god use one which is called RAID6 and not RAID10

that you can lose two disk in a RAID10 with luck is exactly that: LUCK - 
period

  parent reply	other threads:[~2021-05-07  3:26 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-06  5:09 raid10 redundancy d tbsky
2021-05-06  7:48 ` Xiao Ni
2021-05-06  9:57   ` d tbsky
2021-05-06 12:38     ` Reindl Harald
2021-05-07  1:28       ` d tbsky
2021-05-08 13:47         ` keld
2021-05-09  0:52           ` Phillip Susi
2021-05-12 17:22             ` David T-G
2021-05-12 17:26               ` Reindl Harald
2021-05-12 17:39                 ` David T-G
2021-05-13 15:38               ` Andy Smith
2021-05-13 15:46               ` Phillip Susi
2021-05-13 15:59                 ` Andy Smith
2021-05-14 14:28                   ` Phillip Susi
2021-05-14 14:37                     ` Andy Smith
2021-05-17  2:07                       ` Brad Campbell
2021-05-06 13:09     ` Phillip Susi
2021-05-07  1:47       ` d tbsky
     [not found]         ` <86A76859-3098-4AB8-9AE7-46FF54736B88@websitemanagers.com.au>
2021-05-07  3:05           ` d tbsky
2021-05-07  3:26         ` Reindl Harald [this message]
2021-05-07 14:53         ` Andy Smith
2021-05-08  1:54           ` d tbsky
2021-05-08  5:55             ` Andy Smith
2021-05-09  1:10         ` Phillip Susi
2021-05-07  1:26     ` d tbsky
2021-05-06 10:39 ` Peter Grandi
2021-05-07  1:37   ` d tbsky
     [not found] ` <AD8C004B-FE83-4ABD-B58A-1F7F8683CD1F@websitemanagers.com.au>
2021-05-07  1:12   ` d tbsky
2021-05-07  9:46     ` Wols Lists
2021-05-11  0:04       ` Phil Turmel
2021-05-12 17:27         ` David T-G
2021-05-12 18:20           ` Phil Turmel
2021-05-12 16:31 ` Roy Sigurd Karlsbakk
2021-05-13 15:38   ` Phillip Susi
2021-05-13 18:21     ` Phil Turmel
2021-05-14 14:30       ` Phillip Susi
2021-05-14 14:48         ` Roy Sigurd Karlsbakk
2021-05-17 20:50           ` Phillip Susi
2021-05-17 22:21             ` Wols Lists
2021-05-18  0:12               ` Phil Turmel
2021-05-18 16:05               ` Phillip Susi
2021-05-18 17:38                 ` Reindl Harald
2021-05-18 18:51                   ` Phillip Susi
2021-05-18 19:02                     ` Reindl Harald
2021-05-18 23:48                 ` antlists
2021-05-19  3:42                   ` Adam Goryachev
2021-05-19 13:02                     ` Phillip Susi
2021-05-19 21:19                       ` Reindl Harald
2021-05-20  1:32                       ` Adam Goryachev
     [not found]                         ` <CAAMCDeeOnraMDNCF6ZZqPAxUrih2gSse1wDYgOfd1LqY-Ffqxw@mail.gmail.com>
2021-05-20 15:08                           ` antlists

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=9b57e1bf-a573-e6a1-2db8-080ab05205b1@thelounge.net \
    --to=h.reindl@thelounge.net \
    --cc=linux-raid@vger.kernel.org \
    --cc=phill@thesusis.net \
    --cc=tbskyd@gmail.com \
    --cc=xni@redhat.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.