Re: Fwd: Installing Linux directly onto RAID6 Array...........

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Well the swap thing was an easy fix/decision.....

I agree that putting swap into an array seems a bit of overkill,
however one of the things that can be a problem is mounting swap files
into /etc/fstab and then the drive with the swap fails......

My workaround, found on the web and re hashed, is to instead create a
4GB swap partition on each drive then using the following script
automount the swaps at boot time using a systemd script.

#!/bin/bash

#  Script for service that autodetects and starts swap partitions

for f in $(fdisk -l | grep "Linux swap" | sort | cut -d' ' -f1 | tr
'\n' ' '); do swapon $f; done

as it only 'finds' swaps on active partitions it prevents boot
problems in the case of a dead drive.

Due to time constraints I've had to build this using bios_boot and
RAID1 /boot/efi and RAID1 /boot partitions for now and the RAID6
partition is currently syncing with a projected finish around 15 hours
from now.  However given what I've learnt I'm convinced that using
initramfs on pre-existing created partitions is the way to
go.....RAID6 for all the arrays, including /boot and /boot/efi.  Once
I've got this migration out of the way and another test box to use I
intend to take this a stage further and make it work.

Thanks for everyones help and ideas, much appreciated.

Tony



On 24 May 2015 at 15:57, Mikael Abrahamsson <swmike@xxxxxxxxx> wrote:
> On Sun, 24 May 2015, Wols Lists wrote:
>
>> On 24/05/15 15:06, Mikael Abrahamsson wrote:
>>>
>>> On Sun, 24 May 2015, Wols Lists wrote:
>>>
>>>> And if you get read errors, well, aiui, raid won't help here either -
>>>> especially with mirrored raid, you just get a read failure. Raid does
>>>> NOT give you error recovery unless the drive physically fails, and if
>>>> it's a bad block it gets fixed at the disk or disk driver level - well
>>>> below the raid driver.
>>>
>>>
>>> You're wrong. In case of a read error from the physical drive on RAID1,
>>> RAID5 or RAID6 then the information will be re-created from another
>>> drive, and written to the drive that threw a read error. This is the
>>> whole point of RAID with parity information.
>>>
>> Except raid 1 isn't parity ... :-)
>
>
> RAID1 means every drive will have the same information, it's mirrored
> between the member disks. What do you think RAID1 is?
>
>> Personally, I still don't think "raid"ing swap is worth it, though.
>> Horses for courses, ram is cheap, and in my circumstances I don't think
>> I'd gain anything.
>
>
> You're welcome to believe anything you want, but if you're publically
> telling people things that are just not true then you should expect to be
> told so.
>
> You're welcome to tell people to not use SWAP at all, but telling people
> RAID1 has no benefit for SWAP because it won't protect you from read erorrs
> is just wrong.
>
>
> --
> Mikael Abrahamsson    email: swmike@xxxxxxxxx
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html




[Index of Archives]     [Linux RAID Wiki]     [ATA RAID]     [Linux SCSI Target Infrastructure]     [Linux Block]     [Linux IDE]     [Linux SCSI]     [Linux Hams]     [Device Mapper]     [Device Mapper Cryptographics]     [Kernel]     [Linux Admin]     [Linux Net]     [GFS]     [RPM]     [git]     [Yosemite Forum]


  Powered by Linux