raid6check missing?

[grooveman]

Hi.

I have installed my first buster OS. It includes a large raid 6 volume (md raid). I have mdadm installed, but I cannot find the raid6check utility anywhere on the system. I've been googling, and I've been reviewing the man pages, and searching the system. I cannot find it anywhere for Buster.

It looks like the mdadm version for Buster (as of now anyway) is 4.1. I went to kernel.org, and took a look at the source package for mdadm 4.1, and I do see it is still in there... so I'm wondering if the exclusion of this utility is an oversight, or if there is some other utility or method we should use to rectify parity issues on raid6 volumes... Or am missing something crucial here?

Thanks.

G

[ruwolf]

6 years ago, it was reported as bug (it was removed, because it was messy):
bugs.Debian.org: #715324 mdadm: many files missing from upstream (docs, mdassemble and raid6check)

[grooveman]

Yes, I saw that was some time ago... what was that then etch? lenny? I figured they would have it back in by now.

Since it isn't there, is there something else that should be used to scrub a RAID6? Even the contemporary docs on mdadm say not to echo into proc for RAID6.... they still say that radi6check is the way to go... what is the debian solultion?

Thanks.

[grooveman]

Hi. Just a quick bump here.

I find it hard to believe that the benevolent maintainers at Debian would remove this utility without having something (better) to replace it... yet I cannot find any evidence of it. Surely they have something...

Anyone?

Thanks.

G

[Head_on_a_Stick]

Even the contemporary docs on mdadm say not to echo into proc for RAID6.... they still say that radi6check is the way to go

[citation needed]

FWIW, both the ArchWiki[1] and the Linux Kernel Archive wiki[2] recommend echoing directly into /sys

Arch Linux also does not currently have a raid6check command:

Code: Select all

[root@E485 ~]# pacman -Fs raid6check
[root@E485 ~]#

Disclaimer: I don't use RAID.

[grooveman]

Actually, it says the opposite:

Code: Select all

echo repair > /sys/block/mdX/md/sync_action

Only valid for parity raids - this will also check the integrity of the data as it reads it, and rewrite a corrupt stripe. It will terminate immediately without doing anything if the array is degraded, as it cannot recalculate the faulty data.

DO NOT run this on raid-6 without making sure that it is the correct thing to do.

With a raid-5 array the only thing that can be done when there is an error is to correct the parity. This is also the most likely error - the scenario where the data has been flushed and the parity not updated is the expected cause of problems like this.

With raid-6, it is possible to detect which block is corrupt - certainly on a single-bit error. However, "repair" will not correct this sort of error. Be very careful using it - it may just rewrite both parities and leave any corruption in place. The reason for this behaviour is that there is no easily detectable cause for data error and the correct repair strategy needs user intervention. There is a utility "raid6check" that you should use if "check" flags data errors on a raid-6.



This was one of the documents that prompted this post. They are telling you, essentially, that it could just as well harm your data as fix your parity. They say "make sure" but how can you be sure? If there is damage to your parity, what choice do you have? The documents for mdadm still say use radi6check. Raid6check is still in the source code for mdadm. If the maintainers at Debian (or Arch or whoever) took it out, I have to believe that there is a better alternative. I just want to know what that alternative is. I cannot find it documented anywhere.