Questions before first purchase


Recommended Posts

Thank you for having me here, this is my first post and also my first attempt to build a NAS on my own. 

 

Right now, I have a PC running windows 10, serving 2 8TB hard drives over my home network, basically for Plex and SMB access. 

Both hard drives are formated as NTFS and almost 12TB of files.  One hard drive is maxed out and the other is 3,71TB deep.

 

After reading some discussions here, it's not possible to "convert" the file system for unRaid, so I would have to build my NAS and use another computer on my home network to send all the files to my server running unRaid. 

 

My main goal, and why I want to use unRaid is for parity, so if a drive is lost, I would just swap it and have it rebuild, at least this is how I understand it, correct me if I'm wrong. 

 

The thing is, how many extra drives do I need to buy to accomplish this?

Just one more 8TB drive?

 

Putting this as simple as possible, also for my own understanding.

1. Build unRaid with the newly purchased drive. 8TB free space.

2. Fill up 8TB over the network from already in use HD. 

3. Result another 8TB free for unRaid.

4. add this hard drive to unRaid. Fill up with the extra 3,71 TB files. Free the second HD.

5. Add the extra and final drive, and build parity on it?

 

This would have me having 2x8TB drive of data and 1x8TB for parity in case one dies. 

Is this possible?

 

Now the more simple questions:

1. If the parity hard drive fails and I add another, am I correct to assume unRaid will just build another one and everything will be ok?

2. Can I run Plex with unRaid without windows?

3. Is SMB possible with unRaid so I can map my drives on windows 10 from another machine?

4. Is NFS possible with unRaid? I need this for my OPPO 203. 

 

Sorry for the lenght post but I want to be sure before bitting the bullet and jumping into unRaid for the first time. 

Edited by TheFreemancer
Link to comment

You can do it this way. There is an element of risk involved, but if you don't currently have the data backed up or parity protected then you are already at risk of data loss so it is more than likely an acceptable level of risk at this point. I would suggest not completely filling your 8TB drive on unraid, maybe move some to your other drive first so that when you add the other drive to the array unRAID will balance the load across the drives for you.

 

Also the answer is yes to all your other questions. SMB/NFS/AFP/FTP are all supported by unRAID, with the above mentioned setup you will be protected from a single drive failure. Also Plex can be run in a docker container on unRAID itself.

Link to comment
10 minutes ago, Poprin said:

unRAID will balance the load across the drives for you

Not sure what is meant here since Unraid doesn't do this.

 

The main thing that stands out to me in the OP is the question of BACKUPS.

 

Parity is not a substitute for backups. You must have a copy of anything important and irreplaceable on another system.

 

You can put the NTFS drives in your server and copy them using Unassigned Devices plugin without going over the network.

 

The answers to each question

  1. Yes
  2. Plex Server runs on Unraid and serves to your Plex Clients, including the web player.
  3. Yes
  4. Yes
  • Like 1
Link to comment

@trurl Sorry not much detail in my swift answer, I should have said if you add the second drive to the same share and it is set to 'high water' if one drive is empty and the other is only 50% full it will start filling the empty drive. I know unRAID doesn't specifically spread the load between drives but it will fill them relatively evenly if you are using all your drives across a share and have it set to high water.

Link to comment
27 minutes ago, Poprin said:

would suggest not completely filling your 8TB drive on unraid, maybe move some to your other drive first so that when you add the other drive to the array unRAID will balance the load across the drives for you.

You posted while I was typing but I will go ahead. Some of what I will say is the same as your reply.

 

I think I know what you mean here. If there is only one data disk in the array then Unraid will have no choice but to put all the data there, whereas if there is more than one data disk in the array then it can choose the other disk depending on the User Share settings.

 

But I would also say it doesn't matter. There is no good reason to have the data evenly distributed.

Link to comment

It seems unRaid is everything I need. 

 

Going to order the third 8TB hard drive now. 

 

I just don't understand how's that parity isn't safe? I mean, if a hard drive dies, you replace it and the data is restored. 

If the parity drive fails, you add another and it gets reconstructed again. 

I can only see this failing if both parity and another drive dies at the same time. 

 

 

Now I have one last question, at least before start building my own NAS. 

 

What about rclone? Can it be done?

I'm wondering if I could mount my google drive with rclone inside unRaid and have my files streamed to PlexMediaServer. 

 

I've seen people acomplishing this but not sure under unRaid. 

 

Link to comment
7 minutes ago, TheFreemancer said:

I just don't understand how's that parity isn't safe?

Lots of even more common ways to lose data besides drive failure, including user error. And we have certainly seen a lot of user error over the years.

 

If you accidentally delete a file, parity can't help. If you get filesystem corruption, parity can't help. If you make a serious mistake in the Unraid webUI such as formatting a disk (lots of people have made this very mistake), parity can't help.

 

You don't have to backup everything, but you absolutely must have another copy of anything important and irreplaceable. You get to decide what qualifies. 

Link to comment
16 minutes ago, TheFreemancer said:

I just don't understand how's that parity isn't safe?

 

3 minutes ago, trurl said:

If you accidentally delete a file,

 

3 minutes ago, trurl said:

absolutely must have another copy of anything important and irreplaceable.

Think of it this way, what would happen if you accidentally deleted all of your kid's baby pictures?  Not a very happy wife.  (And yes, I did manage to do that a number of years ago - luckily I did have the bulk of them on a CD)

Link to comment
14 minutes ago, Squid said:

 

 

Think of it this way, what would happen if you accidentally deleted all of your kid's baby pictures?  Not a very happy wife.  (And yes, I did manage to do that a number of years ago - luckily I did have the bulk of them on a CD) 

That is my single biggest fear.  So to elaborate to the OP.  If you delete all your pics, the parity will be updated to reflect that you did so.  Like a double record of your error.

 

I back mine up to a cloud source and a USB drive.  So if the cloud source goes down, blows up, ceases to exist, I still have two copies, one on my Unraid server and one on a cold USB HDD.

 

I think of it as like an insurance policy for my marriage...... :D

 

Link to comment
33 minutes ago, TheFreemancer said:

What about rclone? Can it be done?

I'm wondering if I could mount my google drive with rclone inside unRaid and have my files streamed to PlexMediaServer. 

I believe this thread will answer that question for you.

 

So far, there is nothing you have mentioned which unRAID will not do for you.  😀

 

By the way, perhaps you already know this, unRAID will support one or two parity drives, but as has been mentioned several times, parity is not a data backup.  it only protects against drive failures, not data corruption, deletion, user error, etc.

Link to comment

And it's not if you delete something unintended, it's when, I'm sure all of us have done something stupid in the past, I know I have......

rm -rf / --no-preserve-root

Need I say more? I was very tired, and in my sleep deprived, addled brain, it kind of made sense.  Only for a split second though before my laptop screen went blank...... then, I knew....

 

Worst thing is, I actually ran

rm -rf /

first, got a big ass warning, telling me if I wanted to proceed I needed to add

--no-preserve-root

and I still did it.

 

Edited by CHBMB
  • Like 1
Link to comment
1 minute ago, trurl said:

You do need to remind her several times a day but you have the rest of that backwards re: who is the lucky one.

Her..... definitely her.

 

I acknowledge she might give a different answer, but that's my appraisal of things, and I'm sticking with it :D

 

Link to comment
  • 4 weeks later...

Hey, finally received 2 more 8TB hard drives. 

 

Going to purchase unRaid tomorrow when I'm building the server. 

I need to know how to copy the files, I need to copy 12TB of files, so sata <-> is the best option. It will take the entire weekend. 

 

How do I do that? 

I leave the hard drive with my stuff un assigned and unRaid will see the NTFS content?

 

I really need to know how to not screw it up and play it safe. I know over lan would be safer but will take ages over gigabit. 

 

Can you guys explain how to do it?

Link to comment
9 minutes ago, TheFreemancer said:

I know over lan would be safer but will take ages over gigabit.

A healthy gigabit network is very close to the top speed of the disk array in Unraid. I would do it the safe way.

 

If you don't have parity built yet, wait until after you load the data. For the initial data load set all your destination shares to cache:no if you have a cache drive installed.

Link to comment
8 minutes ago, TheFreemancer said:

I have a 60gb SSD doing nothing here. 

 

Would be a good idea to use it as a cache drive? I mean, after copying everything. 

If you are planning on using dockers or a VM, it's practically a requirement. For a drive that small, I wouldn't use it to cache writes to the array, just use it for your app drive.

Link to comment

Bought unraid, installed both 8TB drives. Copying from my network is killing everything else. Can't even download a movie while transfering. 

 

So I added the other 8 TB drive into unRaid. 

Installed unassigned devices plugin. 

Installed Krusader.

 

Mounted my drive that is in exfat format. 

Now I'm copying from the drive to /media/user/unRaid (made this last folder)

 

So the thing is: both drives are xfs now, and set as high water, which should populate the other when the first is getting full. 

 

I'm trying to understand this: if I put a file into /mnt/disks1 the fille will be only at disk1.

if I copy to /mnt/disks2 the file will be only on disks 2.

 

If I copy to /media/user, it will populate one and will start populating the other?

Let's say I'm copying a 10TB folder named movie. 

UnRaid would create a folder in /mnt/disks2/movies to keep copying everything?

 

And this would seem like a single share from the network?

 

Also, if nothing is written to disk2 yet, can I change that split directory rule I think I left it offline and it will be bad for full movies like BDMV folder and tv shows.

Link to comment

Be careful and don't mix disks (array or cache) and user shares when moving or copying. Linux doesn't know that the source and destination could be the same file and will try to overwrite the file it is reading. It's OK with Unassigned since they are never part of user shares.

 

High-water is a little complicated to explain, but is the default setting for a reason. It will eventually switch to the other disk, but it won't constantly switch back and forth between the disks just because one disk is fuller.

https://wiki.unraid.net/Un-Official_UnRAID_Manual#High_Water

Link to comment

Ok, so copy only to /media/user/

 

My unassigned drive mounted showed some erros and I can't actually understand.

First it's showing a 8TB as a SSD.

 

Here's the log:

Mar 15 21:54:06 Tower kernel: ata5: SATA max UDMA/133 abar m2048@0xd0910000 port 0xd0910200 irq 89
Mar 15 21:54:06 Tower kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
Mar 15 21:54:06 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 21:54:06 Tower kernel: ata5.00: ATA-9: WDC WD80EZAZ-11TDBA0, 7SGZ2Y4C, 83.H0A83, max UDMA/133
Mar 15 21:54:06 Tower kernel: ata5.00: 15628053168 sectors, multi 0: LBA48 NCQ (depth 32), AA
Mar 15 21:54:06 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 21:54:06 Tower kernel: ata5.00: configured for UDMA/133
Mar 15 21:54:06 Tower kernel: sd 5:0:0:0: [sdd] 15628053168 512-byte logical blocks: (8.00 TB/7.28 TiB)
Mar 15 21:54:06 Tower kernel: sd 5:0:0:0: [sdd] 4096-byte physical blocks
Mar 15 21:54:06 Tower kernel: sd 5:0:0:0: [sdd] Write Protect is off
Mar 15 21:54:06 Tower kernel: sd 5:0:0:0: [sdd] Mode Sense: 00 3a 00 00
Mar 15 21:54:06 Tower kernel: sd 5:0:0:0: [sdd] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Mar 15 21:54:06 Tower kernel: sdd: sdd1
Mar 15 21:54:06 Tower kernel: sd 5:0:0:0: [sdd] Attached SCSI removable disk
Mar 15 21:54:22 Tower emhttpd: WDC_WD80EZAZ-11TDBA0_7SGZ2Y4C (sdd) 512 15628053168
Mar 15 21:59:39 Tower unassigned.devices: Adding disk '/dev/sdd1'...
Mar 15 21:59:39 Tower unassigned.devices: Mount drive command: /sbin/mount -t exfat -o auto,async,noatime,nodiratime,nodev,nosuid,umask=000 '/dev/sdd1' '/mnt/disks/My_Book'
Mar 15 21:59:39 Tower unassigned.devices: Successfully mounted '/dev/sdd1' on '/mnt/disks/My_Book'.
Mar 15 22:36:18 Tower kernel: ata5.00: exception Emask 0x10 SAct 0x8000000 SErr 0x0 action 0x6 frozen
Mar 15 22:36:18 Tower kernel: ata5.00: irq_stat 0x08000000, interface fatal error
Mar 15 22:36:18 Tower kernel: ata5.00: failed command: READ FPDMA QUEUED
Mar 15 22:36:18 Tower kernel: ata5.00: cmd 60/00:d8:e8:e3:46/01:00:27:01:00/40 tag 27 ncq dma 131072 in
Mar 15 22:36:18 Tower kernel: ata5.00: status: { DRDY }
Mar 15 22:36:18 Tower kernel: ata5: hard resetting link
Mar 15 22:36:18 Tower kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
Mar 15 22:36:18 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 22:36:19 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 22:36:19 Tower kernel: ata5.00: configured for UDMA/133
Mar 15 22:36:19 Tower kernel: ata5: EH complete
Mar 15 22:49:05 Tower kernel: ata5.00: exception Emask 0x10 SAct 0x20000000 SErr 0x400000 action 0x6 frozen
Mar 15 22:49:05 Tower kernel: ata5.00: irq_stat 0x08000000, interface fatal error
Mar 15 22:49:05 Tower kernel: ata5: SError: { Handshk }
Mar 15 22:49:05 Tower kernel: ata5.00: failed command: READ FPDMA QUEUED
Mar 15 22:49:05 Tower kernel: ata5.00: cmd 60/00:e8:e8:9a:d8/01:00:2b:01:00/40 tag 29 ncq dma 131072 in
Mar 15 22:49:05 Tower kernel: ata5.00: status: { DRDY }
Mar 15 22:49:05 Tower kernel: ata5: hard resetting link
Mar 15 22:49:05 Tower kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
Mar 15 22:49:05 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 22:49:05 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 22:49:05 Tower kernel: ata5.00: configured for UDMA/133
Mar 15 22:49:05 Tower kernel: ata5: EH complete
Mar 15 22:55:49 Tower kernel: ata5.00: exception Emask 0x10 SAct 0x30000000 SErr 0x400000 action 0x6 frozen
Mar 15 22:55:49 Tower kernel: ata5.00: irq_stat 0x08000008, interface fatal error
Mar 15 22:55:49 Tower kernel: ata5: SError: { Handshk }
Mar 15 22:55:49 Tower kernel: ata5.00: failed command: READ FPDMA QUEUED
Mar 15 22:55:49 Tower kernel: ata5.00: cmd 60/00:e0:80:20:45/01:00:3c:01:00/40 tag 28 ncq dma 131072 in
Mar 15 22:55:49 Tower kernel: ata5.00: status: { DRDY }
Mar 15 22:55:49 Tower kernel: ata5.00: failed command: READ FPDMA QUEUED
Mar 15 22:55:49 Tower kernel: ata5.00: cmd 60/00:e8:80:21:45/01:00:3c:01:00/40 tag 29 ncq dma 131072 in
Mar 15 22:55:49 Tower kernel: ata5.00: status: { DRDY }
Mar 15 22:55:49 Tower kernel: ata5: hard resetting link
Mar 15 22:55:49 Tower kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
Mar 15 22:55:49 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 22:55:49 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 22:55:49 Tower kernel: ata5.00: configured for UDMA/133
Mar 15 22:55:49 Tower kernel: ata5: EH complete
Mar 15 23:20:45 Tower kernel: ata5.00: exception Emask 0x0 SAct 0x80000001 SErr 0x1400000 action 0x6 frozen
Mar 15 23:20:45 Tower kernel: ata5: SError: { Handshk TrStaTrns }
Mar 15 23:20:45 Tower kernel: ata5.00: failed command: READ FPDMA QUEUED
Mar 15 23:20:45 Tower kernel: ata5.00: cmd 60/00:00:80:21:0e/01:00:32:01:00/40 tag 0 ncq dma 131072 in
Mar 15 23:20:45 Tower kernel: ata5.00: status: { DRDY }
Mar 15 23:20:45 Tower kernel: ata5.00: failed command: READ FPDMA QUEUED
Mar 15 23:20:45 Tower kernel: ata5.00: cmd 60/00:f8:80:20:0e/01:00:32:01:00/40 tag 31 ncq dma 131072 in
Mar 15 23:20:45 Tower kernel: ata5.00: status: { DRDY }
Mar 15 23:20:45 Tower kernel: ata5: hard resetting link
Mar 15 23:20:45 Tower kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
Mar 15 23:20:45 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 23:20:45 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 23:20:45 Tower kernel: ata5.00: configured for UDMA/133
Mar 15 23:20:45 Tower kernel: ata5: EH complete
Mar 15 23:25:19 Tower kernel: ata5.00: exception Emask 0x10 SAct 0x40 SErr 0x400000 action 0x6 frozen
Mar 15 23:25:19 Tower kernel: ata5.00: irq_stat 0x08000000, interface fatal error
Mar 15 23:25:19 Tower kernel: ata5: SError: { Handshk }
Mar 15 23:25:19 Tower kernel: ata5.00: failed command: READ FPDMA QUEUED
Mar 15 23:25:19 Tower kernel: ata5.00: cmd 60/00:30:80:c4:c9/01:00:33:01:00/40 tag 6 ncq dma 131072 in
Mar 15 23:25:19 Tower kernel: ata5.00: status: { DRDY }
Mar 15 23:25:19 Tower kernel: ata5: hard resetting link
Mar 15 23:25:20 Tower kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
Mar 15 23:25:20 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 23:25:20 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 23:25:20 Tower kernel: ata5.00: configured for UDMA/133
Mar 15 23:25:20 Tower kernel: ata5: EH complete
Mar 15 23:25:27 Tower kernel: ata5.00: exception Emask 0x10 SAct 0x180000 SErr 0x0 action 0x6 frozen
Mar 15 23:25:27 Tower kernel: ata5.00: irq_stat 0x08000008, interface fatal error
Mar 15 23:25:27 Tower kernel: ata5.00: failed command: READ FPDMA QUEUED
Mar 15 23:25:27 Tower kernel: ata5.00: cmd 60/00:98:80:3a:d6/01:00:33:01:00/40 tag 19 ncq dma 131072 in
Mar 15 23:25:27 Tower kernel: ata5.00: status: { DRDY }
Mar 15 23:25:27 Tower kernel: ata5.00: failed command: READ FPDMA QUEUED
Mar 15 23:25:27 Tower kernel: ata5.00: cmd 60/00:a0:80:3b:d6/01:00:33:01:00/40 tag 20 ncq dma 131072 in
Mar 15 23:25:27 Tower kernel: ata5.00: status: { DRDY }
Mar 15 23:25:27 Tower kernel: ata5: hard resetting link
Mar 15 23:25:28 Tower kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
Mar 15 23:25:28 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 23:25:28 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 23:25:28 Tower kernel: ata5.00: configured for UDMA/133
Mar 15 23:25:28 Tower kernel: ata5: EH complete
Mar 15 23:29:18 Tower kernel: ata5: limiting SATA link speed to 3.0 Gbps
Mar 15 23:29:18 Tower kernel: ata5.00: exception Emask 0x0 SAct 0x60000000 SErr 0x1400000 action 0x6 frozen
Mar 15 23:29:18 Tower kernel: ata5: SError: { Handshk TrStaTrns }
Mar 15 23:29:18 Tower kernel: ata5.00: failed command: READ FPDMA QUEUED
Mar 15 23:29:18 Tower kernel: ata5.00: cmd 60/00:e8:80:04:15/01:00:35:01:00/40 tag 29 ncq dma 131072 in
Mar 15 23:29:18 Tower kernel: ata5.00: status: { DRDY }
Mar 15 23:29:18 Tower kernel: ata5.00: failed command: READ FPDMA QUEUED
Mar 15 23:29:18 Tower kernel: ata5.00: cmd 60/00:f0:80:05:15/01:00:35:01:00/40 tag 30 ncq dma 131072 in
Mar 15 23:29:18 Tower kernel: ata5.00: status: { DRDY }
Mar 15 23:29:18 Tower kernel: ata5: hard resetting link
Mar 15 23:29:18 Tower kernel: ata5: SATA link up 3.0 Gbps (SStatus 123 SControl 320)
Mar 15 23:29:18 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 23:29:18 Tower kernel: ata5.00: supports DRM functions and may not be fully accessible
Mar 15 23:29:18 Tower kernel: ata5.00: configured for UDMA/133
Mar 15 23:29:18 Tower kernel: ata5: EH complete

 

And error:

Unraid device sdd SMART health [199]: 15-03-2019 23:29

Warning [TOWER] - udma crc error count is 2
WDC_WD80EZAZ-11TDBA0_7SGZ2Y4C (sdd)

 

199 UDMA CRC error count 0x000a 200 200 000 Old age Always Never 2

From smart data.

 

This is the hard drive with the files it's currently copying 1TB to the array.

 

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.