Koozali.org: home of the SME Server

SME 9.2.0 64 bits - Activité disque dur et bruits...

Offline STRyk

  • ****
  • 190
  • +0/-0
SME 9.2.0 64 bits - Activité disque dur et bruits...
« on: April 28, 2020, 03:19:07 PM »
Bonjour,
J'ai un problème récurent sur mon server SME ou j'ai installé 4 disques de 8To en RAID1.
Toutes les 10 secondes les disques sont sollicités, ils font du bruit.
J'ai un second server en 4x 3To toujours en RAID1, machine identique, disque de même marque sans aucun problème de ce coté là.
J'ai testé sur une autre machine, pas de problème non plus.
SME 9.2.0 à été installé avec ces options :
Quote
> vmlinuz initrd=initrd.img text selinux=0 raid=1 nolvm nospare
Ensuite : Toutes mises à jour faites.

A la sortie de la commande ci-dessous, je lis ceci :
Quote
# iotop --only -b

TID  PRIO  USER     DISK READ  DISK WRITE  SWAPIN      IO    COMMAND
352 be/3 root        0.00 B/s    3.95 K/s  0.00 %  4.72 % [jbd2/md1-8]
352 be/3 root        0.00 B/s   31.58 K/s  0.00 %  4.38 % [jbd2/md1-8]
352 be/3 root        0.00 B/s    7.89 K/s  0.00 %  4.59 % [jbd2/md1-8]
352 be/3 root        0.00 B/s    0.00 B/s  0.00 %  3.02 % [jbd2/md1-8]
352 be/3 root        0.00 B/s  102.62 K/s  0.00 % 36.93 % [jbd2/md1-8]
352 be/3 root        0.00 B/s   67.10 K/s  0.00 %  4.00 % [jbd2/md1-8]
352 be/3 root        0.00 B/s   33.48 K/s  0.00 %  5.48 % [jbd2/md1-8]
352 be/3 root        0.00 B/s    0.00 B/s  0.00 %  3.94 % [jbd2/md1-8]

J'ai la même chose avec un PID 856 que j'ai killé et qui reviens parfois, ou parfois ne reviens pas mais cette fois ci pour "jbd2/md3-8" :
Quote
856 be/3 root        0.00 B/s    0.00 B/s  0.00 %  5.54 % [jbd2/md3-8]

Savez vous comment je peux chercher plus loin et comprendre le phénomène afin de ne plus avoir ce soici ?
Car j'ai peur pour les disques à terme. :(
Merci par avance.

Offline ReetP

  • *
  • 3,291
  • +5/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #1 on: May 01, 2020, 01:28:12 AM »
4 * 8Tb in one RAID 1?

Errrrrrr. Ok. As a matter of interest why?

I had a read on your issue and from what I can see the most likely culprit is a bad drive.

Jdb is the journalling process. I would not try to kill it.... it is trying to save your data!

I'm not exactly sure of your drive partition layout. Why no LVM?

You may want to ask on the linux raid mailing list. Nice bunch of people there. They are the gurus.
...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline STRyk

  • ****
  • 190
  • +0/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #2 on: May 01, 2020, 01:42:16 AM »
4 * 8Tb in one RAID 1?

Errrrrrr. Ok. As a matter of interest why?
Et pourquoi pas ? :)
Audio et video utilise beaucoup de place...
Avec un grand disque récent il y a moins de consommation electrique que plusiers disques et il était hors de question d'avoir plein de disques : pour quel intérrêt plein de disques ?
Pourquoi RAID 1 parce que le RAID5 est moins fiable.

I had a read on your issue and from what I can see the most likely culprit is a bad drive.
Jdb is the journalling process. I would not try to kill it.... it is trying to save your data!
Dans ce cas, pourquoi un mauvais disque ? Je ne comprends pas : sauver mes données ?

Merci.
Je suis en RAID 1, pourquoi il sauverait mes données et comment ?

I'm not exactly sure of your drive partition layout. Why no LVM?
You may want to ask on the linux raid mailing list. Nice bunch of people there. They are the gurus.
En cas de montage sur une autre machine dans le cas d'un très gros crash c'est moins compliqué pour moi.

Offline ReetP

  • *
  • 3,291
  • +5/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #3 on: May 01, 2020, 02:12:47 AM »
We can discuss the merits of different RAID all day. Each has their merits....

I think the extra capital cost of big disks may compensate the use of electricity..... YMMV.

Agreed RAID 5 on a few big disks is a disaster waiting to happen. MTBF + rebuild time.... Even RAID 6 is not great.

LVM.... someone else can comment but it isn't that difficult.

Journalling..... you really need to understand what it does.....

If you use ext3 or ext4 you use journalling.

There are plenty of pages online you can read.

And lots of journalling may suggest a bad disk. I'd do a thorough test on each disk to check.

Afterwards you can reconsider some of the cost/benefits of small disks :-)
...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline STRyk

  • ****
  • 190
  • +0/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #4 on: May 01, 2020, 04:15:07 AM »
Le débat des différent RAID n'est pas le sujet ici tout comme la taille du disque, à moins que je n'ai pas compris...
Oui j'utilise le formatage ext4.
Ces disques sont neufs.
Que dois-je vérifier ?
Parce que j'ai déjà cherché pendant des heures interminables sans aboutir à quoi que ce soit.
Sinon je ne posterai pas ici. :(
Je comprends que c'est le "jbd2/md1-8" mais après...

Dois-je laisser travailler le programme ?
Combien de temps va t'il tourner ?

Si je lance "cat /proc/mdstat" je lis ceci :

Quote
Personalities : [raid1]
md0 : active raid1 sda1[0] sdb1[1]
      255936 blocks super 1.0 [2/2] [UU]

md3 : active raid1 sdd1[1] sdc1[0]
      7813894144 blocks super 1.2 [2/2] [UU]
      bitmap: 0/59 pages [0KB], 65536KB chunk

md1 : active raid1 sdb3[1] sda3[0]
      7807526912 blocks super 1.1 [2/2] [UU]
      bitmap: 1/59 pages [4KB], 65536KB chunk

md2 : active raid1 sdb2[1] sda2[0]
      6107136 blocks super 1.1 [2/2] [UU]

Parfois sur le md1, le paramètre "bitmap" affiche 4/59, parfois 1/59 ou autre chose.
Je ne sais pas pourquoi...
Faudrait il reconstruire le RAID ? mais quel disque est en cause ? Si c'est l'un des deux qui pose problème ?
« Last Edit: May 01, 2020, 05:17:45 AM by STRyk »

Offline ReetP

  • *
  • 3,291
  • +5/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #5 on: May 01, 2020, 11:39:48 AM »
Please note - I am no guru on this. The gurus on RAID live on the linux raid mailing list....

No one else replied so I thought I would see what we can discover.

Your setup is a little unusual.

Looking at your setup, have you got two Mirrors with two disks??

Probably ought to look at
Code: [Select]
mount and see how you have created this.

What make and model are these drives? Are they desktop or enterprise ones?

Quote
Ces disques sont neufs.

And? It means nothing.

Disks can be dead out of the box. They can fail at ANY time. That is why you use RAID.

I usually run some burn in tests on any new disks before deploying them.


Code: [Select]
Dois-je laisser travailler le programme ?
Combien de temps va t'il tourner ?

A I said, read about journalling, and do NOT kill it - it is trying to save your data..... but I think you have deeper problems.

Quote
Parfois sur le md1, le paramètre "bitmap" affiche 4/59, parfois 1/59 ou autre chose.
Je ne sais pas pourquoi...
Faudrait il reconstruire le RAID ? mais quel disque est en cause ? Si c'est l'un des deux qui pose problème ?

You really need to ask that on the linux raid mailing list. I suggest you start by reading this:

https://raid.wiki.kernel.org/index.php/Linux_Raid

Particularly this:

https://raid.wiki.kernel.org/index.php/Linux_Raid#When_Things_Go_Wrogn

https://raid.wiki.kernel.org/index.php/Choosing_your_hardware,_and_what_is_a_device%3F

Bitmaps:

https://raid.wiki.kernel.org/index.php/Write-intent_bitmap

Ahhh - and this may well be very relevant:

https://raid.wiki.kernel.org/index.php/Timeout_Mismatch

...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline STRyk

  • ****
  • 190
  • +0/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #6 on: May 01, 2020, 01:54:40 PM »
Thank you for sending me to specific documentation on the subject.
Alone we get too lost in things that we think are important.
When I read you, I understand the subject better than trying to understand certain things that may not concern me by browsing different Internet pages.

Yes, I actually have 2 RAID1 volumes.
The first RAID1 is created from the installation of SME and the second RAID1 I created it myself.
It's the first one which is problematic at present.
It had been running 2 months in trial without problem.

# mount :
Quote
/dev/md1 on / type ext4 (rw,usrquota,grpquota)
proc on /proc type proc (rw)
sysfs on /sys type sysfs (rw)
devpts on /dev/pts type devpts (rw,gid=5,mode=620)
tmpfs on /dev/shm type tmpfs (rw)
/dev/md0 on /boot type ext4 (rw)
/dev/md3 on /mnt/Raid-2 type ext4 (rw,usrquota,grpquota)
/home/e-smith/files/users on /mnt/Raid-1 type none (rw,noexec,nosuid,nodev,bind)
none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw)

Information :
Quote
# smartctl -l scterc /dev/sda
device does not support SCT Error Recovery Control command
# smartctl -l scterc /dev/sdb
device does not support SCT Error Recovery Control command

# smartctl -l scterc /dev/sdc
SCT Error Recovery Control:
           Read:     70 (7.0 seconds)
          Write:     70 (7.0 seconds)
# smartctl -l scterc /dev/sdd
SCT Error Recovery Control:
           Read:     70 (7.0 seconds)
          Write:     70 (7.0 seconds)

# cat /proc/mdstat
Quote
Personalities : [raid1]
md3 : active raid1 sdc1[0] sdd1[1]
      7813894144 blocks super 1.2 [2/2] [UU]
      bitmap: 0/59 pages [0KB], 65536KB chunk

md0 : active raid1 sda1[0] sdb1[1]
      255936 blocks super 1.0 [2/2] [UU]

md1 : active raid1 sdb3[1] sda3[0]
      7807526912 blocks super 1.1 [2/2] [UU]
      bitmap: 3/59 pages [12KB], 65536KB chunk

md2 : active raid1 sdb2[1] sda2[0]
      6107136 blocks super 1.1 [2/2] [UU]

unused devices: <none>

/dev/sda : ST8000DM004-2CX188
/dev/sdb : ST8000DM004-2CX188
/dev/sdc : ST8000VN0022-2EL112
/dev/sdd : ST8000DM002-1YW112

Quote
I usually run some burn in tests on any new disks before deploying them.
Is it possible to know how make this test ?

;)

Offline ReetP

  • *
  • 3,291
  • +5/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #7 on: May 01, 2020, 03:05:51 PM »
Quote
It had been running 2 months in trial without problem.

That is what HAS happened. It may have been perfect yesterday, but this is today. The only thing that is important now is what WILL happen.

First, you have got a good backup?

Second:

Quote
smartctl -l scterc /dev/sda
device does not support SCT Error Recovery Control command
# smartctl -l scterc /dev/sdb
device does not support SCT Error Recovery Control command

Oh dear. And here is why:

Quote
/dev/sda : ST8000DM004-2CX188
/dev/sdb : ST8000DM004-2CX188

Seagate Barracuda desktop drives???

We can see from the "Choosing Your Hardware" page that this is not good. Please read it again.

Quote
/dev/sdc : ST8000VN0022-2EL112
/dev/sdd : ST8000DM002-1YW112

Seagate IronWOlf NAS drives.

These should be OK, but you should test them in case.

Try a few of these to thrash your disks - the disk wipers are fairly effective at testing as they have to write to every sector.... but you will lose any data.

https://www.ultimatebootcd.com/

I'd suggest you start with those desktop drives - that is most likely where the issues are as per the wiki page. They should be removed and only be used in a desktop......

Ultimately, you have illustrated the reason I prefer to run a bunch of smaller (and therefore cheaper) enterprise disks..... my home server and work servers have RAID 6, a hot spare, and a cold spare. Rebuild time is quite fast.

I need security over speed, but it is fast enough for us. With a decent hardware RAID Card and some BBWC they are pretty fast with all the normal small stuff. But if I wanted speed I could use other RAID systems.

Like everything RAID, a lot of reading and a good understanding is required before you start. It will save a lot of issues later......
...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline STRyk

  • ****
  • 190
  • +0/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #8 on: May 01, 2020, 08:44:24 PM »
I understand better but I had no choice. I have already waited months for availability on some discs. In vain.
We had to find a solution. Given the size of the disks, no, I only have this archive.
It's a server that only works when I need to go and get items.
The budget is very tight, I can not invest more. I'm DJ, and Covid-19 is now here...
 :-(
I'm searching a solution for my files :(

Does this problem is it specific to SME ? Because I never saw this on my Mac (I used 2x 8TB RAID1 too).
Again, on a Windows XP on RAID1 with 2x 1TB drive : no problem.
(With desktop drives).
Same for my friends : 2x 2TB (desktop) drive on a Ubuntu without this problem.
« Last Edit: May 01, 2020, 10:06:43 PM by STRyk »

Offline ReetP

  • *
  • 3,291
  • +5/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #9 on: May 04, 2020, 05:40:54 PM »
I understand better but I had no choice. I have already waited months for availability on some discs. In vain.
We had to find a solution. Given the size of the disks, no, I only have this archive.

Oooops :-(

Quote
It's a server that only works when I need to go and get items.
The budget is very tight, I can not invest more. I'm DJ, and Covid-19 is now here...
 :-(
I'm searching a solution for my files :(

It's one reason why lots of small disks are better than a few big ones  :-( And why you need a good understanding before trusting all your data with no backup.....

I know you are struggling for money, but saving say 100 euros is now going to cost you a lot more..... :-(

Quote
Does this problem is it specific to SME ? Because I never saw this on my Mac (I used 2x 8TB RAID1 too).

Nope. It is to do with drives doing things they were not designed to do. Desktop drives are NOT designed for 24/7 operation. And the most likely time with failure is when it arrives, or the first few weeks.

Quote
Again, on a Windows XP on RAID1 with 2x 1TB drive : no problem.
(With desktop drives).
Same for my friends : 2x 2TB (desktop) drive on a Ubuntu without this problem.

It is almost totally irrelevant. They could have had the same issues. They just got lucky and didn't. It does NOT mean they made the right decision.

From the RAID pages you will also notice that there are less issues with smaller drives for various reasons.

But ultimately using desktop drives in a server - as per their documentation - is just a disaster waiting to happen. They are NOT suitable for RAID systems. At all, ever. And the bigger they are, the more problems you are going to have if it goes wrong. And backups are there to save you from yourself......

You never actually mentioned if you had lost any data, but I am now presuming you have.

Have you tried to mount each one of the drives in another computer? One may be OK and can be mounted by itself?


...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline ReetP

  • *
  • 3,291
  • +5/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #10 on: May 04, 2020, 05:45:17 PM »
Note this regarding smaller older drives...

https://raid.wiki.kernel.org/index.php/Choosing_your_hardware,_and_what_is_a_device%3F

Quote
When buying disk drives, make sure they support SCT/ERC, otherwise known as TLER. Most drives of 1TB or less are okay, manufacturers seem to have removed this support from their cheaper drives as capacities went over 1TB.
...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline ReetP

  • *
  • 3,291
  • +5/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #11 on: May 04, 2020, 05:48:02 PM »
Also

Quote
Desktop and Enterprise drives

Disk drives now tend to come in two varieties, desktop drives from which most of the features needed for a decent raid have been deleted, and enterprise drives, which have the features but are designed to run 24/7. So if you want to run raid on a desktop system it's rather difficult to find a drive that is suitable.

If you're going to buy desktop drives (raid drives are a bit more expensive, but not much) you MUST read the Timeout Mismatch section.

And

https://raid.wiki.kernel.org/index.php/Timeout_Mismatch

Quote
Most cheap modern desktop drives do not support some form of managed error recovery. This seems to have started when the typical drive size hit 1TB. So for drives over 1TB, you should buy drives that are explicitly suitable for RAID. Most drives of 1TB or less are okay, but you should check first. RAID-rated drives aren't that much more expensive

...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline krisden

  • *
  • 40
  • +0/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #12 on: May 04, 2020, 08:45:08 PM »
Hello french friend  8-)

Before to focalize on a software issue, did you check all and all possible hardware issue ?
I mean :
  • server power cable
  • server outlet
  • PSU
  • sata cable

Did you already swap disks on your another server ? (note each disks emplacement before to do that)
Most common issues are hardware relative

Bon courage en espérant que tu retrouves tes datas (mon sme me sert de pare-feu et héberge mon Dolibarr et pas grand chose d'autre sur deux WD black 500 GB sans problèmes, pour le moment :D  )

Offline ReetP

  • *
  • 3,291
  • +5/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #13 on: May 05, 2020, 11:15:28 AM »
Hello french friend  8-)

LOL

Quote
Before to focalize on a software issue, did you check all and all possible hardware issue ?
I mean :
  • server power cable
  • server outlet
  • PSU
  • sata cable

Indeed. But from the description I don't think this is likely.

Quote
Did you already swap disks on your another server ? (note each disks emplacement before to do that)
Most common issues are hardware relative

In this instance it is likely the result of using desktop drives as I described, and described on the linux raid list.

At this point, messing about with the disks too much is not a great idea (see "When things go wrong")

If possible each drive should be imaged, but that requires....lots of storage!!!!

I would suggest each of the desktop drives (I suspect they are the problem) are plugged individually into another machine and checked. Hopefully it is only one of them that has an issue and data can be recovered.

Quote
Bon courage en espérant que tu retrouves tes datas (mon sme me sert de pare-feu et héberge mon Dolibarr et pas grand chose d'autre sur deux WD black 500 GB sans problèmes, pour le moment :D  )

As I have already said and per the raid wiki, disks UNDER 1Tb seems to be better behaved. Once they go over 1Tb you may find issues.

Years ago when I first started messing about with RAID I could not afford new enterprise drives so played with desktop drives. It soon became apparent that you needed a) good backups and b) spare disks on hand

So I started to run RAID 1 with a hot spare AND another cold spare, all the time knowing that I was playing with fire.

As soon as I could I swapped to enterprise drives, and still ALWAYS have a hot spare and at LEAST one cold spare. And backups including USB and affa. I have never run RAID 5 - far too dangerous for me. And even RAID 6 is going to highlight its limitations when it is used on big drives.

Even now the biggest drives I run are 1Tb WD RE4 drives in my backup server.....

Essentially you have got to put a price on your data.

You have to choose whether you can't afford to lose your data, or you can't afford good drives :-)
...
1. Read the Manual
2. Read the Wiki
3. Don't ask for support on Unsupported versions of software
4. I have a job, wife, and kids and do this in my spare time. If you want something fixed, please help.

Bugs are easier than you think: http://wiki.contribs.org/Bugzilla_Help

If you love SME and don't want to lose it, join in: http://wiki.contribs.org/Koozali_Foundation

Offline STRyk

  • ****
  • 190
  • +0/-0
Re: SME 9.2.0 64 bits - Activité disque dur et bruits...
« Reply #14 on: May 08, 2020, 01:32:14 AM »
Hello and thank you for your comments.
Bonjour et merci pour vos commentaires.

  • server outlet
Qu'est-ce que c'est ça ?
Je n'ai aps de problème matériel.


I have read documentation, spent a lot of time, carried out tests but I really do not understand:
I am fortunate to have a machine to test with two 3TB drives.
These two discs are Hitachi Deskstar 7K3000 HDS723030ALA640 with the famous 'TLER'.
It is therefore activated :

J'ai lu des documentations, passé beaucoups de temps, réalisé des essais mais je ne comprend vraiment pas :
J'ai la chance d'avoir une machine pour faire un essai avec deux disques de 3To.
Ces deux disques sont des Hitachi Deskstar 7K3000 HDS723030ALA640 avec le fameux 'TLER'.
Il est donc activé :


Quote
# smartctl -l scterc /dev/sda
SCT Error Recovery Control:
           Read:     70 (7.0 seconds)
          Write:     70 (7.0 seconds)
# smartctl -l scterc /dev/sdb
SCT Error Recovery Control:
           Read:     70 (7.0 seconds)
          Write:     70 (7.0 seconds)
         
BUT I still have this famous problem :
MAIS j'ai tout de même ce fameux problème :
Quote
  TID  PRIO  USER     DISK READ  DISK WRITE  SWAPIN      IO    COMMAND
  358 be/3 root        0.00 B/s    0.00 B/s  0.00 %  3.51 % [jbd2/md1-8]
  358 be/3 root        0.00 B/s   26.75 K/s  0.00 %  3.53 % [jbd2/md1-8]
  358 be/3 root        0.00 B/s    3.82 K/s  0.00 %  6.03 % [jbd2/md1-8]
It is totally inconsistent with what we can read.
There may be something else, but what to look for ?


C'est totalement incohérent avec ce que l'on peut donc lire.
Il y a peut etre autre chose, mais quoi chercher ?


Important Note:
-We can adjust the value of these two discs: what value to put?
-I read that it takes at least 8 seconds. It's amazing as a value because any current disc is fast enough to respond faster than 8 seconds ... I don't understand.
-As long as I don't understand the problem more precisely, the 2x8TB RAID1 server will remain turned off.


NB:
-Nous pouvons régler les valeur de ces deux disques : quelle valeur mettre ?
-J'ai lu qu'il faut au minimum 8 secondes. C'est étonant comme valeur car n'importe quel disque actuel est suffisament rapide pour répondre plus vite que 8 secondes... Je ne comprends pas.
-Tant que je ne comprends plus précisément le problème, le serveur de 2x8To RAID1 restera eteinds.


What do you think of all this?
Que pensez vous de tout ceci ?