1 hour ago, JorgeB said:

If that's an option...

I do have an old first gen Ryzen CPU and motherboard lying about, but I think I'll try your fix first as the intel system has a lot more sata and m.2 ports. 

 

Regretting cheaping out when setting up my rack though as I just used 2 posts and the 4U server with unraid on it is kind of anchoring everything else down   Going to sort something better out with that first so I have easier access to pull the server out when needed.

If I edit the syslinux/syslinux.cfg file as you said, will any changes get overwritten by the next version update?

 

 

1 minute ago, Mundvl said:

If I edit the syslinux/syslinux.cfg file as you said, will any changes get overwritten by the next version update?

No, standard updates do not touch that file.

8 minutes ago, itimpi said:

No, standard updates do not touch that file.

That's good.  Again thanks for your help will keep an eye on the other thread to see how it goes with Binary100100 as I haven't got time to pull my server out at the moment.

Even though it goes against my nature to install any software that ends in "0" (and ".0.0" is even worse), I pulled myself together and updated my backup server to 7.0.0.  It is a fairly plain vanilla server, with Frigate and MQTT dockers running on it.  It does have a Realtek motherboard NIC that needs the plugin installed, which I left in place during the install.

 

All went without event.  Thanks to everyone who contributed to this milestone revision.

10 hours ago, Mundvl said:

New to unraid, installed 6.12.14 less than a month ago - moving all my files from my old servers to it, decided to upgrade to 7.0.0 stable for some of the new features, seemed to download okay, but when it rebooted, I could not access the server on the network, plugged in a monitor and it seems to be getting stuck on booting SMP configuration (tried rebooting a few times).

Hardware is i7-6700K on an Asrock z170 OC Formula (updated to the latest BIOS) with 16GB RAM.  Added an intel  X520-DA1 for 10gbit and a nvidia 1050ti.

 

I am hoping there is some way of fixing this without requiring a backup, as I made one before proceeding with the upgrade, but when I looked for it I notice the download had failed

 

Grateful for any advice to fix this. 

Just now seeing this.
That was my exact issue. Looks exactly like that.
For anyone else that finds this intriguing, JorgeB provided a solution. 

On 1/19/2025 at 9:49 PM, JorgeB said:

AFAIK nothing you can do for now.

What a shame. Downgraded to 6.12.14 - spindown works again for all drives. Will wait for this issue to be resolved before trying to upgrade again.

Can anyone recommend a suitable modern replacement for my 2 Supermicro AOC-SASLP-MV8 (based on the Marvell 9480 host controller) 8-port SAS/SATA cards? Ideally one that works well with Unraid 7 and future versions...

Just found an older post recommending something like this: https://www.amazon.de/dp/B098TQKPKS
Is this still a good and future-proof option?

Edited January 22, 20251 yr by AlexNathan345

39 minutes ago, AlexNathan345 said:

Can anyone recommend a suitable modern replacement for my 2 Supermicro AOC-SASLP-MV8 (based on the Marvell 9480 host controller) 8-port SAS/SATA cards?

https://forums.unraid.net/topic/102010-recommended-controllers-for-unraid/

 

20 hours ago, ConnerVT said:

Even though it goes against my nature to install any software that ends in "0" (and ".0.0" is even worse), I pulled myself together and updated my backup server to 7.0.0.  It is a fairly plain vanilla server, with Frigate and MQTT dockers running on it.  It does have a Realtek motherboard NIC that needs the plugin installed, which I left in place during the install.

 

All went without event.  Thanks to everyone who contributed to this milestone revision.

I felt the same way. x.0.0 doesn't inspire confidence, but this time it did me nothing but good things.

I upgraded my server hardware last week, and I'm going to dive into VMs for the first time. I plan to use the server as my Windows desktop.

Any idea if/when the VM documentation will be updated to reflect the improvements to Windows VMs in v7?

Many thanks.

 

EDIT: I just read this in the documentation:

USB hot-plugging is not currently supported, so devices must be attached before the VM is started in order for USB pass through to function.

Ummm does that mean I can't plug in USB devices while I'm running the VM? That's a pretty normal thing to do, right?

Can I fix this by passing through a USB hub? Or do I have to shutdown my Windows VM every time I want to copy the contents of my camera's SD card? Or do I have to go via unassigned devices or something?

Thanks for your help and insight

Edited January 22, 20251 yr by jademonkee

always best to pass through a USB controller rather than an individual device if you have the ability. that will give you full hot-plug and generally just work a lot better than trying to pass a device through separately.

Just did the upgrade.  Followed directions, updated plugins, stopped array.  On reboot, the computer did not detect the usb stick with unraid on it as a bootable device.  Rebooted multiple times, same thing.  Finally pulled the stick, plugged it into my desktop, confirmed that it was detected there, then put back into the unraid server and then it could detect it.  Really weird.  It then went through the install and everything appears to be fine now.

11 hours ago, Faceman said:

always best to pass through a USB controller rather than an individual device if you have the ability. that will give you full hot-plug and generally just work a lot better than trying to pass a device through separately.

Ah good, I was concerned that I straight up couldn't hot plug USB, but if passing through the controller gives the ability to plug/unplug then that's the answer I needed. Thanks.

From what little knowledge I have, it looks like I can separate my controllers into those for Unraid and those for the VM without issue (it looks like i have enough IOMMU groups).

56 minutes ago, jademonkee said:

Ah good, I was concerned that I straight up couldn't hot plug USB, but if passing through the controller gives the ability to plug/unplug then that's the answer I needed. Thanks.

From what little knowledge I have, it looks like I can separate my controllers into those for Unraid and those for the VM without issue (it looks like i have enough IOMMU groups).

I have a plugin that supports hot pluging via the VM tab or you can do dynamic by setting mapping. USB Manager.

I tried to update from 6.12.14 to 7.0.0 and ran into the "Wrong pool State. Too many wrong or missing devices" error.  I tried to follow the instructions, created a pool (never had one), set it to 3 (only data disks). I could not find under the pool name where to add any disks. So I went back to 6.12.14 and now there is an empty pool Normal. Again I don't see anyway to add disks. What am I missing?

 

I know I'm not a power user.

UnRAID OS Plus

ASRock Z690 Steel Legend
American Megatrends International, LLC., Version 2.10

12th Gen Intel® Core™ i5-12600K @ 3663 MHz  16Gig memory

16 Gig USB drive

12 minutes ago, Sandman416 said:

"Wrong pool State. Too many wrong or missing devices" error.  I tried to follow the instructions, created a pool (never had one)

Please re-read that post, there are two different errors, yours is the second one.

Used the gui upgrade and got the "Kernal Panic" error on restart...

Removed Flash drive and manually overwrote all bz* files on the flash.  7.0 boot up just fine after.

On "Micro-Star International Co., Ltd. MEG X399 CREATION (MS-7B92)"

and drives are not spinning down like the release notes warned... I assume there's a bug post somewhere I can follow for updates to that.  Have to consider downgrading or stay on 7.0 for the time being. 

I guess worst part of having drives not spin down is heat/power and possibly lower lifespan on the drives.

21 minutes ago, horrorhound said:

I assume there's a bug post somewhere I can follow for updates to that.

There's a kernel issue with some controllers, if you are using one, you can follow this thread, I'll post any update there.

2 hours ago, horrorhound said:

I guess worst part of having drives not spin down is heat/power and possibly lower lifespan on the drives.

I wouldn't worry about it. More wear happens when powering up rather than when keeping it spinning. Besides, I don't think it will take years to resolve this issue.

23 hours ago, JorgeB said:

Please re-read that post, there are two different errors, yours is the second one.

Found it. It woke me up this morning when it dawned on me what started it. There was a cache pool that got auto created in an update that was never populated.

 

So I can happily say update complete. 

 

Thank you.

My upgrade was a little 'eventful' but I seem to have got it all working again now...

 

Upgraded from latest 6.x version to 7 and jellyfin started having issues with transcoding, it was working fine before hand... roll back and the same issue with transcoding popped up... 

 

Only really wanted to move over because my dedicated jellyfin/tailscale setup stopped working for some reason and this would make life easier... seriously easy to set up this time.

 

So said screw it, I'll just do a clean install of unraid 7, now enjoying the pleasure of fixing jellyfin metadata again due to it not 'transferring over' from my backups, I really wish there was a simple 'backup and restore' feature in jellyfin rather than relying on unraid backups or copying stuff like the jellyfin instructions (this is the bit that failed!). 

 

The power mode options seem backwards on my system (i5 13500) for some reason, not sure why but I use more power (according to nut and my ups) when in power save than when I'm in performance lol.  Leaving it in performance matches my old power draw give or take. 

 

With the updated dashboard I really wish I could have a 4th column on my 4K display.  I have to scroll vertically even though I have space on the sides, changing to full width doesn't really change vertical requirements for me either. 

 

Performance seems a little slower to me but I'm putting that down to stuff going on in the background. 

Edited January 26, 20251 yr by LSG501

Hi, is there any official guidance on the "mover breaking hardlink issue"

 

e.g.

1) is the "Reynald Mover plugin" also affected?

2) Is there any guidance on preventing this issue? is it "don't run mover until a fix is released" or "do a,b,c, and you won't have any issues", etc

 

thanks

Edited January 26, 20251 yr by StylishQuoter

My upgrade to 7.0 didn't go so well.  Both of my thunderbolt enclosures (OWC Thunderbay 8 and OWC Flex 1U4) failed to show any drives in unRAID 7.0 and I was forced to roll back 6.12.14.  I tried to make some tweaks under system devices under 7.0 with no luck.  Aren't support for the thunderbolts controllers built into the kernel?  Working in 6.12.14 are the following: Intel Corporation JHL6540 Thunderbolt 3 Bridge and Intel Corporation JHL7440 Thunderbolt 3 Bridge and they are hooked to a Minisforum ms-01 using and Intel Corporation Raptor Lake-P Thunderbolt 4 USB Controller.  Attached logs requested during downgrade from 7.0 to 6.12.14.  Any thoughts?

unraid-diagnostics-20250125_1120.zip

Successfully upgraded from 6.12.8.  Now will be waiting for the spin down fix.  unRAID 7.0.0 seems to be running smoothly on my really old hardware that I will may not have to upgrade the hardware after all.  Just need to look into migrating from resierfs to xfs.

I recently upgraded from Unraid 6.12.10 to Unraid 7, and initially, everything seemed to be running smoothly. However, after a few days, I started noticing severe slowdowns, especially in the GUI. Eventually, the system became completely unresponsive, forcing me to perform a hard reboot.
 

Issue:
Upon inspecting the logs, I found multiple warnings such as:

Unraid php-fpm[7718]: [WARNING] [pool www] child 3095784 exited on signal 9 (SIGKILL) after 15.407785 seconds from start

After searching online, many discussions suggested that plugins could be the cause of such issues. I went through a tedious process of uninstalling plugins, rebooting, and testing again, but the issue persisted. Just before I was about to downgrade back to Unraid 6.12.10, I accidentally noticed high transfer rates on my Unraid USB stick.
 

Root Cause & Solution:
I recalled reading the Unraid 7 changelog, which mentioned a known issue: "Excessive flash drive activity slows the system down." The recommended fix was to create a file named fastusr in /boot/config/ and reboot:

touch /boot/config/fastusr

After applying this fix, my system has been running smoothly with no further slowdowns.
 

Final Thoughts:
The issue arises because when too much RAM is used, the OS gets pushed out of memory, causing data to be read and written directly from the flash drive, which significantly slows down the system. My server has 64GB of DDR4 RAM, running around 50 Docker containers and 5 vdi`s. My typical RAM usage is between 80-90%, which likely contributed to the problem.

Why is this configuration file not created by default in Unraid? There is no reason for the OS to ever be pushed out of RAM. Ensuring that the system remains responsive at all times should be a fundamental priority.

Edited January 29, 20251 yr by Cout99

2 hours ago, Cout99 said:

Why is this configuration file not created by default in Unraid?

I agree, and asked for that for 7.0, maybe LT can do it for next release.