What is Unraid and how to build an Unraid media server

[Manni]

Yes, definitely running with MC, and Paul's help has been invaluable

I've just started the migration for the rest of the main share (32TB). MC is predicting over 62 hours, but I reckon it will be a bit less because he's still averaging under 100MB/s as it's just started, it should adjust this under 60 hours once its average has caught up to 110MB/s minimum.

MC final transfer.JPG (93.06 KiB) Viewed 6476 times

I'll keep you posted guys in the next few days, and thank you both, it would have been a very lonely (and less smooth) ride without the two of you.

Have a great Xmas break if I don't update this thread before next week!

[Manni]

Quick update, MC is still going but it hang for some reason yesterday morning, so I had to restart it. Apart from that, all going well.

I've installed a Windows 10 VM, in case anyone else wants to do it here is a nice tutorial to get started (it's for win 11 but it will work with any Windows 10 ISO as well, I copied it from my Synology NAS):

For some reason, left clicking on the start button using VLC does nothing (it works everywhere else). Right click works though. Quite weird. It works fine with RDC, so who cares...

Ethernet is working fine though, I've run windows update.

I've allocated 8GB and I'm still using only 62% of my total 16Gb in Unraid. The VM doesn't seem to slow down the transfer at all, so I think I should be able to give 12GB to the VM when/if needed, as I don't need any docker for now.

So all is good VM-wise, I plan to install MyMovies later today (after family duties) and see how that works.

EDIT: All is working with MyMovies, I can use the VM as a bridge for the Oppo 203, which is great as the Unraid server will always be on if I use the Oppo, given that it holds most of my 4K UHD Bluray collection. This means I can select any title on the MyMovies app on iPhone/iPad and get it to play on the Oppo 203 (or the Dune 4K Pro, but the Dune doesn't need an actual PC as a bridge, it's a direct connection). Until now, I had to switch my HTPC or a laptop to be able to control the Oppo, which was annoying.

Paul, when you have a moment, please could you sum up how you would suggest I set-up MyMovies, CCC and CMC, assuming MyMovies server in the Unraid VM and a single instance of CCC? I'd like to do it properly... It's not urgent of course.

Have a great day everyone!

[Jamie]

Thank you Manni for posting this video. I will need it when I install my windows 10 VM on my unraid server. Right now I am waiting to get another 16 GB of ram to boost the server to 32 GB. I will allocate 16 GB to the windows 10 vm. Anything you guys post here will be appreciated. Merry Christmas everyone.

Jamie

[Manni]

Thank you Manni for posting this video. I will need it when I install my windows 10 VM on my unraid server. Right now I am waiting to get another 16 GB of ram to boost the server to 32 GB. I will allocate 16 GB to the windows 10 vm. Anything you guys post here will be appreciated. Merry Christmas everyone.

Jamie

Hi Jamie,

You're welcome, glad you found the link useful. I hope you had a nice Xmas!

The content of my surviving QNAP NAS has been successfully copied to the Unraid server. The VM works very well, I can now save my MyMovies metadata in a local share which makes it a lot faster than when saving over the network.

I've also decided that I might try to use this VM for calibration duty, as it could run my calibration software and be connected to my

I've moved the array from the dead QNAP to the surviving QNAP, and all my data is back on line. Yay!

I've "pre-cleared" 2 more 4TB drivers that I was planning to use as parity, just to give them a work-out (it took 43 hours, which feels a bit excessive for 4TB drives!), but I suddenly realised that it might be a better idea to keep them as extra data/semi-hot spares. I might instead add 6TB parity drives, which means I could add 6tb drives as well if I need more data in that array. I noticed that the read speed during the pre-clear was only 60MB/s, which feels a bit slow. I hope it's because of the process itself.

I'm waiting for the new SAS controller (LSI 9201-16i) to arrive before adding the parity drives (I still have the QNAP array as a backup if everything goes wrong).

I also plan to install the 1TB SSD as a cache, once I'm sure I won't need to boot into Windows to update the new controller f/w.

Once that's done, I'll start building my "proper" Unraid server, with a 9305-24i (I found one for around £360 new, which isn't too bad), and the new 24-bay case, as well as my minimalist MB (2500K/8GB RAM). This one won't run any VM/Docker, so it should be fine. It's the one that will be upgraded when I replace my AMD Ryzen MB.

I'm waiting to hear from Paul about the best way to setup MyMovies/CCC from the VM. I guess he's busy at the moment...

Best,

Manni

[Jamie]

Hi Manni,

I have read and Paul has stated that the Parity drives should be the largest drives on your unraid array. I have two 8 tB discs for parity drives and a combination of 6 tb and 4 tb discs for the rest of the array. I hope to upgrade the 4 tb discs to 8 tb in the future. I have a 24 bay case and I am currently using 19 bays for a total of 76 TB of space. My two parity drives are seagate drives which make me a little nervous because Paul mentioned that there is an unraid bug regarding 8 tb seagate drives. I have had my unraid server for over 2 years now and I haven't noticed any issues.

I hope everyone here has a Happy new year!!! Take care.

Jamie

[Manni]

The Unraid bug with Seagate Ironwolf drives is only with Unraid V6.9.x and only with LSI controllers. Also not all Seagate drives are affected, though those that are include 8TB models. Are yours Ironwolf, and are you using an LSI controller?

There is a workaround (https://forums.unraid.net/topic/103938- ... mmary-fix/) that means changing a couple of default settings in the f/w, so if you're still with a pre 6.9.x version, you might want to do the mod for each disk before upgrading to 6.9.x. Or you can sit tight and not upgrade.

If you're already with 6.9.x, you should be fine, you would have seen your drives drop out of the array by now.

Or maybe it's only a problem with data drives and not parity drives?

[Jamie]

I am at 6.8.3. I have 2 LSI 9201-8i controllers. The drives are Iron wolf 8 tb drives. I haven't seen any issues and I will avoid 6.9.x.

[Manni]

I am at 6.8.3. I have 2 LSI 9201-8i controllers. The drives are Iron wolf 8 tb drives. I haven't seen any issues and I will avoid 6.9.x.

You won't see any issue with 6.8.x, that's why Paul stayed on (or rather reverted back to) that older version as well.

However, don't upgrade to 6.9.x without making the suggested changes to the drives f/w (they all have to be tweaked individually).

If you don't need anything from 6.9.x, easier and safer to not upgrade indeed!

Best,

Manni

[Jamie]

I have been reading about the 6.9.x unraid seagate issue and it looks like unraid may not be fixed in newer versions. Looks like it will have to be fixed in an LSI driver update or the seagate fw.

I am thinking about doing the fix to the seagate drives now even though I have no plans of upgrading at this time. I just do not want to be surprised if I upgrade later on. I usually use WD drives but the seagate drives were too tempting.

I'm surprised that this didn't get found in unraid or even linux testing since some think an update in the linux drivers caused the problem.

[Manni]

Yes, I wouldn’t expect this issue to be fixed in a future Unraid version, though it’s always possible.

The old 4TB Hitachi drives I’m using were just lying around. I decided a few years ago to move fully to WD so I only have 6TB Red, Red Pro or Red Plus in all my servers (that’s 40 6TB drives currently in service, plus a few spares). My 8-bay QNAPs and 24-bay (12+12) Synology servers were quite old, so 6TB was the maximum they could handle. Currently I only have one 10TB SEAGATE drive that I use for offsite backup of my most essential data.

Interestingly enough, I’m currently running a couple 6TB I’m not sure about through a pre-clear, and weirdly one of them seems to hang the unraid, despite the fact that both went through the WD Diag write zero process without any problem (except that it took a bit longer than usual). I’ve used a more detailed reporting from the pre-clear plugin (an email every 25% of each step) to identify which one was causing the issue, because yesterday when it crashed I had to hard power cycle the server to get it back to life. The server already crashed once a while back, without any reason.

It could be the Adaptec controller, I’ll find out when I get the new LSI controller. Not very reassuring to have the whole server hang like this, especially during a disc verification. I wouldn’t want that to happen during a parity rebuild after I lost one or two data drives.

Anyway, I’ll find out soon enough.

Happy new year!