Rocksolid Light

Welcome to novaBBS (click a section below)

mail  files  register  newsreader  groups  login

Message-ID:  

Back when I was a boy, it was 40 miles to everywhere, uphill both ways and it was always snowing.


aus+uk / uk.comp.sys.mac / Re: RAID constantly corrupting

SubjectAuthor
* RAID constantly corruptingMartin S Taylor
+* Re: RAID constantly corruptingTheo
|+- Re: RAID constantly corruptingnospam
|`- Re: RAID constantly corruptingTheo
`- Re: RAID constantly corruptingBruce Horrocks

1
RAID constantly corrupting

<0001HW.29E8658C0042B11570000A21738F@news.eternal-september.org>

  copy mid

https://www.novabbs.com/aus+uk/article-flat.php?id=15562&group=uk.comp.sys.mac#15562

  copy link   Newsgroups: uk.comp.sys.mac
Path: i2pn2.org!i2pn.org!eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: correspo...@mRaErMtOiVnEsTtHaIySlor.com (Martin S Taylor)
Newsgroups: uk.comp.sys.mac
Subject: RAID constantly corrupting
Date: Thu, 13 Apr 2023 17:30:04 +0100
Organization: A noiseless patient Spider
Lines: 29
Message-ID: <0001HW.29E8658C0042B11570000A21738F@news.eternal-september.org>
Reply-To: correspondence@mRaErMtOiVnEsTtHaIySlor.com
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
Injection-Info: dont-email.me; posting-host="cedf92480358a77476e1f2eeb579dc70";
logging-data="1120868"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX19U8oiLTUCLTZAgI/zJfaByvr0XUFrViIo="
User-Agent: Hogwasher/5.24
Cancel-Lock: sha1:WxaEWI4amTCPCOFUKlvJmHvYfgs=
Mail-Copies-To: nobody
X-Face: |DM;jr,c64.*AL%fk,trjr%G%Rk"k^n7!U)u+3muHSnI]TL#_QlfWB.Llwu-&?
OPs<vKf|bUu,MX%HK4dh\!=?3F0%.a-\bgj9CquM24/9YmW&4EX%e;!4\Ls*
vcbN>T3[oMA%R/e,-pg'{s,7Qf9GT0bbf>czov>^4f~jPgN
 by: Martin S Taylor - Thu, 13 Apr 2023 16:30 UTC

Last year I bought a couple of Icy Box RAID enclosures. These have given me
intermittent problems for a while, mainly bit-flip errors, invalid hashes,
and underallocation. All sorted by Disk Utility, I kept trouble at bay by
replacing the USB cables, connecting direct to the Mac (rather than through a
hub) etc. Problems seemed to be caused when using Carbon Copy Cloner, but I'd
guess this is because CCC a) checks the validity of its copying (so I get to
find out about any problems immediately), and b) CCC copies terabytes of data
when I use it.

Anyhow, last week there were big problems. The culmination came yesterday
when I used Carbon Copy Cloner to back up my Mac's internal drive to a folder
on Icy Box. Immediately afterwards there were overlapped extent allocation
errors, and the whole disk is pretty much unusable.

I've applied all my knowledge to track down the cause, but I'm a bit stuck.
What's likely to be causing this, and what can I rule out?

• Hardware/firmware error in the Icy Box?
• Physical problems with the cable?
• Problems because the long (3m) USB cable snakes its way past lots of
other cables, including mains cables?
• Hardware/software error at the Mac end?
• Bug in Carbon Copy Cloner?

Any suggestions on the cause, or advice for future testing gratefully
received.

Martin S Taylor

Re: RAID constantly corrupting

<Icq*gaHdz@news.chiark.greenend.org.uk>

  copy mid

https://www.novabbs.com/aus+uk/article-flat.php?id=15563&group=uk.comp.sys.mac#15563

  copy link   Newsgroups: uk.comp.sys.mac
Path: i2pn2.org!i2pn.org!paganini.bofh.team!newsfeed.xs3.de!callisto.xs3.de!nntp-feed.chiark.greenend.org.uk!ewrotcd!.POSTED.chiark.greenend.org.uk!not-for-mail
From: theom+n...@chiark.greenend.org.uk (Theo)
Newsgroups: uk.comp.sys.mac
Subject: Re: RAID constantly corrupting
Date: 13 Apr 2023 18:47:06 +0100 (BST)
Organization: University of Cambridge, England
Message-ID: <Icq*gaHdz@news.chiark.greenend.org.uk>
References: <0001HW.29E8658C0042B11570000A21738F@news.eternal-september.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Injection-Info: chiark.greenend.org.uk; posting-host="chiark.greenend.org.uk:212.13.197.229";
logging-data="29412"; mail-complaints-to="abuse@chiark.greenend.org.uk"
User-Agent: tin/1.8.3-20070201 ("Scotasay") (UNIX) (Linux/5.10.0-20-amd64 (x86_64))
Originator: theom@chiark.greenend.org.uk ([212.13.197.229])
 by: Theo - Thu, 13 Apr 2023 17:47 UTC

Martin S Taylor <correspondence@mraermtoivnestthaiyslor.com> wrote:
> Last year I bought a couple of Icy Box RAID enclosures. These have given
> me intermittent problems for a while, mainly bit-flip errors, invalid
> hashes, and underallocation. All sorted by Disk Utility, I kept trouble
> at bay by replacing the USB cables, connecting direct to the Mac (rather
> than through a hub) etc. Problems seemed to be caused when using Carbon
> Copy Cloner, but I'd guess this is because CCC a) checks the validity of
> its copying (so I get to find out about any problems immediately), and b)
> CCC copies terabytes of data when I use it.

What kind of RAID is the Icy Box doing? 0/1/5/6/10? Hardware RAID in the
box, or software RAID (as multiple drives joined together on the Mac, like a
Fusion drive)?

Where are you seeing these errors? In CCC, or from the box?

> Anyhow, last week there were big problems. The culmination came yesterday
> when I used Carbon Copy Cloner to back up my Mac's internal drive to a
> folder on Icy Box. Immediately afterwards there were overlapped extent
> allocation errors, and the whole disk is pretty much unusable.
>
> I've applied all my knowledge to track down the cause, but I'm a bit stuck.
> What's likely to be causing this, and what can I rule out?
>
> • Hardware/firmware error in the Icy Box?

If it's hardware RAID I would suspect the Icy Box...

> • Physical problems with the cable?

If there was a cable error the USB protocol should resend it. If it's
serious I'd expect discs dropping out.

Is the USB powering the Icy Box, or does it have external power?

> • Problems because the long (3m) USB cable snakes its way past lots of
> other cables, including mains cables?

Unlikely, but USB should handle that.

> • Hardware/software error at the Mac end?

Unfortunately HFS and APFS don't have checksums, so it is possible for
corrupt data to end up on the disc if there's a problem with the hardware.

(ZFS checksums all on-disc data, so you can confirm everything is
consistent. This is a good way to confirm hardware is behaving itself.
Unfortunately Apple skipped that for APFS, saying its flash was sufficiently
reliable)

> Any suggestions on the cause, or advice for future testing gratefully
> received.

If it's RAID1 it would be possible to pull one of the drives out of the
array and see if it's better with a single drive, but I wouldn't advise that
unless you have a full backup. Although with a dubious RAID I'd be wanting
a full backup in any case.

Theo

Re: RAID constantly corrupting

<130420231440319807%nospam@nospam.invalid>

  copy mid

https://www.novabbs.com/aus+uk/article-flat.php?id=15567&group=uk.comp.sys.mac#15567

  copy link   Newsgroups: uk.comp.sys.mac
Path: i2pn2.org!i2pn.org!eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: nos...@nospam.invalid (nospam)
Newsgroups: uk.comp.sys.mac
Subject: Re: RAID constantly corrupting
Date: Thu, 13 Apr 2023 14:40:31 -0400
Organization: A noiseless patient Spider
Lines: 11
Message-ID: <130420231440319807%nospam@nospam.invalid>
References: <0001HW.29E8658C0042B11570000A21738F@news.eternal-september.org> <Icq*gaHdz@news.chiark.greenend.org.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
Injection-Info: dont-email.me; posting-host="bdc6e30689cefd5f094b2b7558f25ea0";
logging-data="1159369"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX19B1azYyMseylhW4IDyttjH"
User-Agent: Thoth/1.9.0 (Mac OS X)
Cancel-Lock: sha1:RutMBby1vzOOMrRoaPA1wtE4Jew=
 by: nospam - Thu, 13 Apr 2023 18:40 UTC

In article <Icq*gaHdz@news.chiark.greenend.org.uk>, Theo
<theom+news@chiark.greenend.org.uk> wrote:

> (ZFS checksums all on-disc data, so you can confirm everything is
> consistent. This is a good way to confirm hardware is behaving itself.
> Unfortunately Apple skipped that for APFS, saying its flash was sufficiently
> reliable)

btrfs also does that, used on synology nases (and a couple of others).

synology nases can also snapshot files, much like time machine on a mac.

Re: RAID constantly corrupting

<390da203-4653-87aa-cc39-f2ae568f8a45@scorecrow.com>

  copy mid

https://www.novabbs.com/aus+uk/article-flat.php?id=15573&group=uk.comp.sys.mac#15573

  copy link   Newsgroups: uk.comp.sys.mac
Path: i2pn2.org!i2pn.org!weretis.net!feeder8.news.weretis.net!newsreader4.netcologne.de!news.netcologne.de!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail
From: 07....@scorecrow.com (Bruce Horrocks)
Newsgroups: uk.comp.sys.mac
Subject: Re: RAID constantly corrupting
Date: Thu, 13 Apr 2023 21:43:37 +0100
Lines: 39
Message-ID: <390da203-4653-87aa-cc39-f2ae568f8a45@scorecrow.com>
References: <0001HW.29E8658C0042B11570000A21738F@news.eternal-september.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
X-Trace: individual.net eDVpabmKmjbooBYsI18KQQhgaeLsDcx1fPIKwcEPdH9B5/WSRP
Cancel-Lock: sha1:DUDw6QwHcO+KvQjgfZTcYvY9Cr4=
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:102.0)
Gecko/20100101 Thunderbird/102.9.1
Content-Language: en-GB
In-Reply-To: <0001HW.29E8658C0042B11570000A21738F@news.eternal-september.org>
 by: Bruce Horrocks - Thu, 13 Apr 2023 20:43 UTC

On 13/04/2023 17:30, Martin S Taylor wrote:
> Last year I bought a couple of Icy Box RAID enclosures. These have given me
> intermittent problems for a while, mainly bit-flip errors, invalid hashes,
> and underallocation. All sorted by Disk Utility, I kept trouble at bay by
> replacing the USB cables, connecting direct to the Mac (rather than through a
> hub) etc. Problems seemed to be caused when using Carbon Copy Cloner, but I'd
> guess this is because CCC a) checks the validity of its copying (so I get to
> find out about any problems immediately), and b) CCC copies terabytes of data
> when I use it.
>
> Anyhow, last week there were big problems. The culmination came yesterday
> when I used Carbon Copy Cloner to back up my Mac's internal drive to a folder
> on Icy Box. Immediately afterwards there were overlapped extent allocation
> errors, and the whole disk is pretty much unusable.
>
> I've applied all my knowledge to track down the cause, but I'm a bit stuck.
> What's likely to be causing this, and what can I rule out?
>
> • Hardware/firmware error in the Icy Box?
> • Physical problems with the cable?
> • Problems because the long (3m) USB cable snakes its way past lots of
> other cables, including mains cables?
> • Hardware/software error at the Mac end?
> • Bug in Carbon Copy Cloner?
>
> Any suggestions on the cause, or advice for future testing gratefully
> received.

Do the Icy boxes have Ethernet? If so, using that might be more reliable.

Otherwise can you move the Icy boxes closer to the Mac and use a shorter
cable for a week or two?

Is the firmware on the Icy boxes up to date (if it can be updated)?

--
Bruce Horrocks
Surrey, England

Re: RAID constantly corrupting

<Icq*q3Qdz@news.chiark.greenend.org.uk>

  copy mid

https://www.novabbs.com/aus+uk/article-flat.php?id=15665&group=uk.comp.sys.mac#15665

  copy link   Newsgroups: uk.comp.sys.mac
Path: i2pn2.org!i2pn.org!news.nntp4.net!nntp.terraraq.uk!nntp-feed.chiark.greenend.org.uk!ewrotcd!.POSTED.chiark.greenend.org.uk!not-for-mail
From: theom+n...@chiark.greenend.org.uk (Theo)
Newsgroups: uk.comp.sys.mac
Subject: Re: RAID constantly corrupting
Date: 15 Apr 2023 15:40:02 +0100 (BST)
Organization: University of Cambridge, England
Message-ID: <Icq*q3Qdz@news.chiark.greenend.org.uk>
References: <0001HW.29E8658C0042B11570000A21738F@news.eternal-september.org> <Icq*gaHdz@news.chiark.greenend.org.uk>
Injection-Info: chiark.greenend.org.uk; posting-host="chiark.greenend.org.uk:212.13.197.229";
logging-data="404"; mail-complaints-to="abuse@chiark.greenend.org.uk"
User-Agent: tin/1.8.3-20070201 ("Scotasay") (UNIX) (Linux/5.10.0-20-amd64 (x86_64))
Originator: theom@chiark.greenend.org.uk ([212.13.197.229])
 by: Theo - Sat, 15 Apr 2023 14:40 UTC

Theo <theom+news@chiark.greenend.org.uk> wrote:
> Unfortunately HFS and APFS don't have checksums, so it is possible for
> corrupt data to end up on the disc if there's a problem with the hardware.
>
> (ZFS checksums all on-disc data, so you can confirm everything is
> consistent. This is a good way to confirm hardware is behaving itself.
> Unfortunately Apple skipped that for APFS, saying its flash was sufficiently
> reliable)

Interestingly there *is* ZFS for MacOS, and people seem to use it and find
it stable:
https://www.reddit.com/r/zfs/comments/xrxka9/state_of_openzfs_on_macos/

Were I doing DASy things I would be tempted to try it. There is the usual
risk of Apple suddenly deciding to ban the access it needs, but in the worst
case you should be able to plug the drives into a Linux machine and access
the data (or maybe boot Linux or a Linux VM on the Mac and pass through
the drives) - and you can test this fallback in advance.

The advantage is it should be more robust in terms of data storage for the
use case, can you get proper parity RAID. The disadvantage is it's not very
popular on MacOS.

Theo

1
server_pubkey.txt

rocksolid light 0.9.81
clearnet tor