15 hours ago, limetech said:

Note: some motherboards may present a strongly worded warning along the lines of "The system found unauthorized changes on the firmware, operating system or UEFI drivers."  In this case look for a "Secure Boot" BIOS setting and change to "Other OS".

 

Not everyone has a keyboard/monitor attached to their systems though. This change is going to generate tons of support requests, do the benefits outweigh the hassle this is going to cause? It might make sense on new systems, but for systems that are already working fine I'm not sure I see the benefit.

 

I think it would be better for the upgrade to place the EFI directory on the flash drive as EFI.notused and then include directions along these lines:

• If you would like to boot in UEFI mode:
  • rename EFI.notused on your flash drive to EFI
  • rename syslinux on your flash drive to syslinux.notused (optional)
• Reboot into your BIOS and enable UEFI. Also look for a "Secure Boot" BIOS setting and change it to "Other OS".

Something like this would give unRAID the future-proofing it needs without wreaking havoc on the installed base. By only having one option enabled at a time users can be sure they know how their systems are going to boot.

I have an existing machine and it just started like normal. I had to manually configure my bios to boot uefi and that was simple as well.


Sent from my iPad using Tapatalk

Looking forward to running 6.4 but won't be installing beta software. So many thanks to all of you that are testing this!

Im having a problem with the web gui, it will load to the main tab then it will just become unresponsive. I get this error in syslog after update.

 

Jun 18 18:08:09 Floki emhttpd: shcmd (69): /usr/local/emhttp/plugins/dynamix.docker.manager/scripts/dockerupdate.php
Jun 18 18:08:43 Floki root:  Updating templates...  Updating info...  Done.
Jun 18 18:08:43 Floki emhttpd: nothing to sync
Jun 18 18:14:45 Floki ntpd[1705]: kernel reports TIME_ERROR: 0x41: Clock Unsynchronized
Jun 18 18:15:07 Floki php-fpm[10981]: [NOTICE] [pool www] child 17097 exited with code 0 after 382.858963 seconds from start
Jun 18 18:15:07 Floki php-fpm[10981]: [NOTICE] [pool www] child 28771 started
Jun 18 18:15:57 Floki kernel: BTRFS critical (device loop2): corrupt leaf, bad key order: block=542326784, root=1, slot=171

 

29 minutes ago, Porkie said:

Im having a problem with the web gui, it will load to the main tab then it will just become unresponsive. I get this error in syslog after update.

 

Jun 18 18:08:09 Floki emhttpd: shcmd (69): /usr/local/emhttp/plugins/dynamix.docker.manager/scripts/dockerupdate.php
Jun 18 18:08:43 Floki root:  Updating templates...  Updating info...  Done.
Jun 18 18:08:43 Floki emhttpd: nothing to sync
Jun 18 18:14:45 Floki ntpd[1705]: kernel reports TIME_ERROR: 0x41: Clock Unsynchronized
Jun 18 18:15:07 Floki php-fpm[10981]: [NOTICE] [pool www] child 17097 exited with code 0 after 382.858963 seconds from start
Jun 18 18:15:07 Floki php-fpm[10981]: [NOTICE] [pool www] child 28771 started
Jun 18 18:15:57 Floki kernel: BTRFS critical (device loop2): corrupt leaf, bad key order: block=542326784, root=1, slot=171

 

 

Did you set the correct time + timezone on your server?

It is recommended to use NTP for time synchronization.

My box is an hour behind so I dont think it adjusted for BST. I reverted back to previous version using a back up I had, which has worked but its really slow.

 

It was my docker.img cause the issues, I deleted it and its back to normal now.

Edited June 19, 20179 yr by Porkie

Booting in UEFI mode from the flash drive stops my NVidia 980ti from working. I get the dreaded code 43 in windows. Back to normal legacy boot and all is OK.

Any ideas what might be wrong?

1 hour ago, mikeyosm said:

Booting in UEFI mode from the flash drive stops my NVidia 980ti from working. I get the dreaded code 43 in windows. Back to normal legacy boot and all is OK.

Any ideas what might be wrong?

 

Did you google this  code 43 nvidia   ?  (It always helps when you tell what things you tried to attempt to solve the problem.)

Yes I searched the site. Hyper v on or off does  not make any difference. Why it works i  legacy but not uefi is not clear to me

Still not realizing the performance of the new mover.. Is it only from cache to array or is it from share to share too? Also does turbo write plugin still apply or is it redundant now due to the I/O mover changes.

 

Thanks!

39 minutes ago, robbiered said:

Also does turbo write plugin still apply

The plugin will still work, and will still improve write speeds if all the drives are spun up vs when they are not.

 

40 minutes ago, robbiered said:

Is it only from cache to array or is it from share to share too?

The only real time you're going to notice a big difference is from SSD to SSD Share.  Anything involving a hard drive basically won't change 

1 hour ago, robbiered said:

Still not realizing the performance of the new mover.. Is it only from cache to array or is it from share to share too? Also does turbo write plugin still apply or is it redundant now due to the I/O mover changes.

 

Thanks!

Mover only moves from cache to array for cache-yes user shares and only moves from array to cache for cache-prefer user shares. There is no moving from share to share. Turbo write is independent of mover. Any writes mover does uses whatever parity method is set.

Understood, thanks!

colors from disks are gone

 

1 hour ago, killeriq said:

colors from disks are gone

 

Have you tried re-selecting it? Sometimes settings get accidentally changed or lost.

Settings -> Display Settings -> Used/Free Columns: Bar (color)

thx for tip, anything i try from that menu - no change look still the same.

4 minutes ago, John_M said:

 

Have you tried re-selecting it? Sometimes settings get accidentally changed or lost.

Settings -> Display Settings -> Used/Free Columns: Bar (color)

 

On 18/06/2017 at 5:37 PM, limetech said:

Also note: the /mnt/user0 mount point is no longer needed and has been removed.

 

I understand /mnt/user0 isn't needed for mover anymore, but I actually use it for other purposes too. I've seen other users referring to user0 in their docker folder mappings too, so I don't think I'm alone in this.

Would you consider keeping /mnt/user0 unless there are other reasons for removing it?

 

Example use case:

A user share is set to cache=yes, so when adding files manually from my desktop the transfer is fast. In a single cache disk scenario this leaves files without parity protection until mover runs.

For automated downloads, the transfer speed is not a factor and I like to transfer the file to the user share directly, bypassing the cache disk and reaping the benefits of instant parity protection.

46 minutes ago, Jorgen said:

Would you consider keeping /mnt/user0 unless there are other reasons for removing it?

 

As soon as you think you can actually simplify some code, someone comes along with a reason why you can't

 

Thank you for the report.  This was something I was wondering about.. we'll put it back.

 
As soon as you think you can actually simplify some code, someone comes along with a reason why you can't [emoji35]
 
Thank you for the report.  This was something I was wondering about.. we'll put it back.

I'm all for simplification so please don't put it back if I'm the only one that want to keep it!
I should really solve my use case with a cache pool.


Sent from my iPhone using Tapatalk

22 minutes ago, Jorgen said:

I'm all for simplification so please don't put it back if I'm the only one that want to keep it!
I should really solve my use case with a cache pool.


Sent from my iPhone using Tapatalk

 

+1.  I also prefer simplifying things.  I've always found the user0 mount points to be confusing anyway.

12 hours ago, killeriq said:

thx for tip, anything i try from that menu - no change look still the same.

 

You might want to check your USB flash device for corruption. Shut down, remove it and plug it into a Windows/Linux/Macintosh computer and run the appropriate disk check utility on it.

7 minutes ago, John_M said:

13 hours ago, killeriq said:

thx for tip, anything i try from that menu - no change look still the same.

 

You might want to check your USB flash device for corruption. Shut down, remove it and plug it into a Windows/Linux/Macintosh computer and run the appropriate disk check utility on it.

 

It's a known bug:

 

 

 

@johnnie.black Thanks

On 18/06/2017 at 8:37 AM, limetech said:

Important: If you want to report an issue, please start a new topic in this board.

 

UEFI support (-rc4)

A new directory is included on the USB flash boot device called 'EFI'.  The presence of this directory along with its contents, and along with some

additional linux kernel options permit UEFI boot.  This is done in such a way that you could choose either BIOS (legacy) or UEFI to boot off your USB flash device.

 

 

oh YES YES YES, thank you!! Now I can see about some Xeon overclocking too.... You may have saved me from needing to buy a new CPU/board etc...

 

The new UEFI boot feature seems nice, but I run a system without keyboard and monitor attached so its hard to change bios options or operate a boot menu. Is there a way to update to 6.4 RC and keep legacy booting active. Would a hack like renaming the uefi directory after upgrading but before rebooting work? Is there a more elegant way to prevent UEFI boot in the first place?