unRAID Server Release 6.2.0-beta18 Available

[Furby8704]

has anyone had success with nfs  shares??

[dlandon]

has anyone had success with nfs  shares??

 

Not working.

[jjslegacy]

has anyone had success with nfs  shares??

 

Not working.

 

Why is this not on the first page?

[bman212121]

I'm very concerned about the total lack of security in the unRAID GUI boot mode, specifically the ability to use the launched FireFox application to browse the internet.

 

Um, then don't boot in that boot mode?

 

Better yet, what do you suggest?

 

This isn't for me, this is for your basic user who doesn't know about security and best practices or the pitfalls of browsing the internet as super-user on a server. I can already see the horror support threads about unknowing consumers' systems being pwned.

 

So is the suggestion to simply block access to the internet from the browser aside from a connection to the local web interface?

 

Jon, I'm just playing with this in a virtual machine.  To further what BRiT is saying, not sure it's a great idea to be able to install plugins in FireFox either.

 

We're aiming at a whole new crowd nowadays that is starting to be a little more divergent from the core users of old...

 

But on the whole I think it's a great idea to have access locally and can see the benefits.  I'm no great browser expert, but was wondering if Chromium may be a better choice as it's sandboxed by design isn't it?

 

So I'm not sure how some of the old users are using this product, but to me it seems like a moot point to be worrying about a browser hijack with the default configuration of unRAID. There is absolutely no security in the default configuration, so providing a web browser running under root isn't going to make or break anything. If you were wanting to lock this down, there is a lot more that needs to be done than just changing how firefox is launched.

[archedraft]

So I'm not sure how some of the old users are using this product, but to me it seems like a moot point to be worrying about a browser hijack with the default configuration of unRAID. There is absolutely no security in the default configuration, so providing a web browser running under root isn't going to make or break anything. If you were wanting to lock this down, there is a lot more that needs to be done than just changing how firefox is launched.

 

I would agrue there is a difference. Currently in 6.1 the user would have to use the CLI to access the interwebs or open up ports on their router, ect. Meaning it currently takes some effort to expose yourself. With the GUI any user with any level of experience can head on out to the interwebs, and lets face it, the Firefox GUI on unRAID doesn't use internet condoms and unRAID itself has no protection, so this user is now sticking exposed "parts"  all over the place and bound to get a nice "virus" of some sort.

[BRiT]

So I'm not sure how some of the old users are using this product, but to me it seems like a moot point to be worrying about a browser hijack with the default configuration of unRAID. There is absolutely no security in the default configuration, so providing a web browser running under root isn't going to make or break anything. If you were wanting to lock this down, there is a lot more that needs to be done than just changing how firefox is launched.

 

I would agrue there is a difference. Currently in 6.1 the user would have to use the CLI to access the interwebs or open up ports on their router, ect. Meaning it currently takes some effort to expose yourself. With the GUI any user with any level of experience can head on out to the interwebs, and lets face it, the Firefox GUI on unRAID doesn't use internet condoms and unRAID itself has no protection, so this user is now sticking exposed "parts"  all over the place and bound to get a nice "virus" of some sort.

 

Yup. Archedraft has it right. There is a huge difference in security risk vulnerabilities between 6.1 and 6.2 unrestricted web browser as root.

 

[Frank1940]

So I'm not sure how some of the old users are using this product, but to me it seems like a moot point to be worrying about a browser hijack with the default configuration of unRAID. There is absolutely no security in the default configuration, so providing a web browser running under root isn't going to make or break anything. If you were wanting to lock this down, there is a lot more that needs to be done than just changing how firefox is launched.

 

I would agrue there is a difference. Currently in 6.1 the user would have to use the CLI to access the interwebs or open up ports on their router, ect. Meaning it currently takes some effort to expose yourself. With the GUI any user with any level of experience can head on out to the interwebs, and lets face it, the Firefox GUI on unRAID doesn't use internet condoms and unRAID itself has no protection, so this user is now sticking exposed "parts"  all over the place and bound to get a nice "virus" of some sort.

 

Yup. Archedraft has it right. There is a huge difference in security risk vulnerabilities between 6.1 and 6.2 unrestricted web browser as root.

 

As one example, I can download a file in my Windows Firefox and execute the file directly from the browser.  Now in my Windows setup, I have the antivirus scan of the files while being download. And Windows throws up a warning box that asks my permission to modify my computer by allowing this file to execute.  (If I am not the administrator, I can't even do that!)  There is some protection against my doing something stupid.  But I would bet that the execution would take place with unRAID Linux OS with any notice. 

[MyKroFt]

All of a sudden starting getting the following in my log file - this happens with any drive that spins down - any idea on what to look for?

 

Mar 13 21:27:04 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:27:04 Tower kernel: mdcmd (74): spindown 8

 

Here is a force spindown command

 

Mar 13 21:40:28 Tower emhttp: Spinning down all drives...
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: mdcmd: write: No such device or address
Mar 13 21:40:28 Tower emhttp: shcmd (7072): /usr/sbin/hdparm -y /dev/sdd &> /dev/null
Mar 13 21:40:28 Tower kernel: mdcmd (76): spindown 0
Mar 13 21:40:28 Tower kernel: mdcmd (77): spindown 1
Mar 13 21:40:28 Tower kernel: mdcmd (78): spindown 2
Mar 13 21:40:28 Tower kernel: mdcmd (79): spindown 3
Mar 13 21:40:28 Tower kernel: mdcmd (80): spindown 4
Mar 13 21:40:28 Tower kernel: mdcmd (81): spindown 5
Mar 13 21:40:28 Tower kernel: mdcmd (82): spindown 6
Mar 13 21:40:28 Tower kernel: mdcmd (83): spindown 7
Mar 13 21:40:28 Tower kernel: mdcmd (84): spindown 8
Mar 13 21:40:28 Tower kernel: mdcmd (85): spindown 9
Mar 13 21:40:28 Tower kernel: mdcmd (86): spindown 10
Mar 13 21:40:28 Tower kernel: mdcmd (87): spindown 11
Mar 13 21:40:28 Tower kernel: mdcmd (88): spindown 12
Mar 13 21:40:28 Tower kernel: mdcmd (89): spindown 13
Mar 13 21:40:28 Tower kernel: mdcmd (90): spindown 14
Mar 13 21:40:28 Tower kernel: mdcmd (91): spindown 15
Mar 13 21:40:28 Tower kernel: mdcmd (92): spindown 16
Mar 13 21:40:29 Tower emhttp: shcmd (7073): /usr/sbin/hdparm -y /dev/sdc &> /dev/null

 

Thanks

Myk

[MyKroFt]

Another one

 

When reinstalling a docker, the AppData Config Path defaults to the docker name field for the final folder even if you had a different path.

 

I like my paths lower case - ex docker name is BTSyncFree - i use btsyncfree for the end path, but leave the name alone so CA sees it correctly - but it is getting auto changed to whatever the name field is

 

Myk

 

[bennymundz]

has anyone had success with nfs  shares??

 

Not working.

 

Why is this not on the first page?

 

Agreed people banging on about perceived security flaws when such a major function of the system not working.

 

NFS is basically paramount to how i use unraid. :-(

[gfjardim]

has anyone had success with nfs  shares??

 

Not working.

 

Why is this not on the first page?

 

Agreed people banging on about perceived security flaws when such a major function of the system not working.

 

NFS is basically paramount to how i use unraid. :-(

 

Guys, please read the BETA statement in the main OP. This version is mainly to debug, so it's normal things break and they will be in the next beta. Please be patient.

[Dephcon]

i think many were suckered in due to it being beta18 and assumed it'd be pretty much prod ready.  i fully expected to roll back after trying it out, but it's been fine for me.

[Furby8704]

Guys, please read the BETA statement in the main OP. This version is mainly to debug, so it's normal things break and they will be in the next beta. Please be patient.

 

its understood on my part. was just asking if there was a temp fix at the moment.

everything else i rely on is running just fine and not a problem using smb for a bit

i shall remain patient until the kinks are worked out. no prob.

[jjslegacy]

I certainly understand it's Beta and things break - no problem but wouldn't it also make sense to keep a running tab of the "major" issues wrong in the release on the main page?  NFS not working seems like a major one to me.

 

Again no complaints just think we could cut down a lot of extra noise if we had a Known issues that was kept somewhat current. 

[StevenD]

It's certainly easy to tell the "new" unraid users when a new release comes out...especially one that's very beta.

[trurl]

6TB dual parity check. Same as my usual results with single parity.

 

Duration: 14 hours, 57 minutes, 44 seconds. Average speed: 111.4 MB/sec

[JorgeB]

I’m still doing more testing but my initial findings are that the 2nd parity itself does not slow down a parity check, it can, if adding an extra disk limits your bus/controller speed (same would happen if adding an extra data disk) or in very big arrays when the parity check is CPU limited, in those cases there will be a little slowdown.

 

Two examples, due to the number of disks the CPU (Pentium G2030, [email protected]) is the limit here:

 

sp + 23d = 191.7MB/s

dp + 22d = 164.2Mb/s

 

sp + 28d = 164.2MB/s

dp + 27d = 138.6MB/s

 

Note that I purposely used the same number of total disks to compare the actual penalty speed when adding the 2nd parity.

 

In my view this is a perfectly acceptable slowdown (about 15%) for the extra protection, and like I said, this will only happen on big arrays with an entry level CPU, lucky ones with Xeons and alike probably won’t notice any difference even in big arrays.

 

[FalafelFighter]

I got it working switching around with NVMe disk and the HDD between Cache and Disk Drives. Letting UNRAID reformat.

But atm. i have the following error:

"SG_IO: questionable sense data, results may be incorrect"

 

Other forums does mention the Samsung 950 PRO NVMe disk may not be formatted properly.

 

UNRAID does still remember when it were formatted as a Cache drive and a Disk Drive.

 

- I have no idear where UNRAID does hide those settings?

 

Does other have problems with the new Samsung 950 PRO? Mine hangs and BSOD while updating/upgrading Windows 10 with a driver issue.

Not clear what you're asking, but unRAID saves configuration on the flash. Most of the configuration files are text so you can read them yourself. It only remembers the current disk assignments.

 

I were just wondering where UNRAID store those configurations. I've tried replacing all HDD and the USB with a different, even when formated multiple times in NTFS and back to FAT.

 

The funny part was that the previous USB i used, still had those SHARES setting from last time i used it and notification settings. I did format it as well before i started. That USB were used as a daily driver for my stuff.

 

My question is:

 

- How do i reset UNRAID to stock/standart settings? (Like i started all over again)

[itimpi]

Is there any reason when creating a vdisk for a VM via the GUI that you are limited to raw and qcow2?  Could not any of the other KVM supported types be listed.  For instance I regularly use .vdi files which seem to provide a good compromise between space and performance and allow me to easily interchange the vdisk files with Virtualbox on my desktop PC.

[peter_sm]

 

Is there any reason when creating a vdisk for a VM via the GUI that you are limited to raw and qcow2?  Could not any of the other KVM supported types be listed.  For instance I regularly use .vdi files which seem to provide a good compromise between space and performance and allow me to easily interchange the vdisk files with Virtualbox on my desktop PC.

There is also LVM :-)

[FalafelFighter]

My second issue is following:

When installing Win10 as a VM in UNRAID 6.2.0 BETA. The installation cannot update to newest Build 1511 without crashing with BSOD afterwards.

The BSOD shows different driver related issues. I did try different driver versions under the VM TAB including the latest version of "virtio-win-0.1.113"

 

I did try in IDE and AHCI with the new BIOS as well as the previous.

 

Sometimes the VM shuts down randomly and shows that the process were terminated.

 

I've used weeks on this previous with HDD array, still the same result.

 

- Do some know the answer or experienced simular problems?

 

[itimpi]

- How do i reset UNRAID to stock/standart settings? (Like i started all over again)

Proceed just as you would if you have never used unRAID before!

• Make sure you have a copy of your .key file somewhere safe
  
• Format the USB stick as FAT32 and label "UNRAID"
  
• Download the ZIP of unRAID from the Limetech site and extract the contents onto the USB stick
  
• Run the 'make_bootable.bat' file as Administrator to make the USB stick bootable
  
• Copy the .key file you saved into the config folder on the USB stick
  
• Boot your unRAID server off the USB stick and it will come as if it were a new installation
  

Is that what you were asking for or have I missed something?

[itimpi]

My second issue is following:

When installing Win10 as a VM in UNRAID 6.2.0 BETA. The installation cannot update to newest Build 1511 without crashing with BSOD afterwards.

The BSOD shows different driver related issues. I did try different driver versions under the VM TAB including the latest version of "virtio-win-0.1.113"

 

I did try in IDE and AHCI with the new BIOS as well as the previous.

 

Sometimes the VM shuts down randomly and shows that the process were terminated.

 

I've used weeks on this previous with HDD array, still the same result.

 

- Do some know the answer or experienced simular problems?

How many CPU's do you have assigned to the VM?  There have been reports that you often need to reduce it to 1 while doing the update, and then can set it back to a larger number afterwards.

[FalafelFighter]

- How do i reset UNRAID to stock/standart settings? (Like i started all over again)

Proceed just as you would if you have never used unRAID before!

• Make sure you have a copy of your .key file somewhere safe
  
• Format the USB stick as FAT32 and label "UNRAID"
  
• Download the ZIP of unRAID from the Limetech site and extract the contents onto the USB stick
  
• Run the 'make_bootable.bat' file as Administrator to make the USB stick bootable
  
• Copy the .key file you saved into the config folder on the USB stick
  
• Boot your unRAID server off the USB stick and it will come as if it were a new installation
  

Is that what you were asking for or have I missed something?

 

I have done the exact same as you mentioned.

All other settings were still there without the "data" on the drives ofc.

 

The only thing i notised that i could create a password again. The SHARES and notifications + where the drives previous were assigned were still showing. Kinda weird...

[itimpi]

I have done the exact same as you mentioned.

All other settings were still there without the "data" on the drives ofc.

unRAID will recognise previously used drives and leave there data intact.  If you want to force them to be cleared then stop the array and change the format.  When you start the array unRAID will now offer to format the drives wiping existing data.  If necessary you can repeat this process to get back to the original format but with the data gone.

 

The only thing i notised that i could create a password again. The SHARES and notifications + where the drives previous were assigned were still showing. Kinda weird...

There should have been no drives assigned if you did what I suggested!  You would have to have copied across a configuration file to get that behaviour.

Shares 'coming back' is expected behaviour if you have not cleared the drives as a share is automatically created for each top level folder on the drives with default settings.