Hello.

 

I'm a synology refugee on my 1st Unraid build. 

 

Motherboard here : https://vi.aliexpress.com/item/1005008053795350.html?spm=a2g0o.productlist.main.1.8da45e61QZC63T&algo_pvid=e619ea87-c3f4-4a98-b8de-74c91593e9a7&algo_exp_id=e619ea87-c3f4-4a98-b8de-74c91593e9a7-0&pdp_npi=4%40dis!VND!5516459!3530503!!!217.33!139.09!%402101584917365109254202091e8133!12000043457837939!sea!VN!0!ABX&curPageLogUid=Yg5OThrZXK9R&utparam-url=scene%3Asearch|query_from%3A#nav-specification

 

I'm using Krusader docker to transfer the files.

 

On migrating data, unraid crashes repeatedly. Not all the time, say 50% with 100GB - 700GB transfers and 100% on 2TB to 5TB transfers.

 

It only crashes on transfer. Perfect otherwise.

 

I have run memtest+ and the RAM passes. 

 

I've tried it on NFSv3 and 4. Same problem.

 

I have switched between my 2 intel nics. Same problem.

 

Removed the cache which I wasn't using yet with one old SSD and cable which had a smart error. Same problem.

 

I have now increased the NFS threads to 32 and increased the Number of fuse File Descriptors to 100,000 per advice seen on the forum and will test now.

 

So either:

 

1. I am a clown who has made an error in my build

2. There is an Unraid NFS issue unresolved.

3. Unraid doesn't properly support my nics: 2*2.5G LANs chipset: Intel i226-V

4. I have a hardware problem with this board and need to return it to Ali

 

I only have a few days left before the 15 day limit expires so need to diagnose this accurately and quickly.

 

I tried using the 1*10G LAN chipset: Marvell AQC113C and encountered a separate issue where I could not access the gui or ping it.

 

Find it hard to believe that a brand new board could have 3 nics all with HW issues when the rest of it works like a charm but I guess it's possible.

 

Any help to get to the bottom of this would be greatly appreciated.

 

 

 

 

gecko-diagnostics-20250210-1350_Anon.zip gecko-syslog-20250210-0656.zip

Update: Increased the NFS threads to 32 and increased the Number of fuse File Descriptors to 100,000, did a reboot and it crashed again on transfer.

 

Currently using both 2.5 nics on a 1gb network that is perfectly stable.

 

Chassis is a new Jonsbo N4, it has a new PSU, New sata cables, new cat cables.

 

This time gui was unreachable as well, had to pull power to reboot.

 

I see btrfs error in the previous syslog I assume that referes to the cache I removed.

 

Apart from that, can't see anything pertaining to the crashes during transfers in the syslogs.

 

 

gecko-syslog-20250211-0934.zip gecko-syslog-previous-20250211-0933.zip gecko-diagnostics-20250211-1630.zip

Both syslogs only cover a few minutes of uptime, when was the last time it crashed?

An hour ago, I then download both syslogs.

 

I am currently setting the access to private and setting rules. I had it set to public. I was unaware of this best practice guidance.

Enable the syslog server and post that after the next crash, and mention the approximate time code it happened.

Local syslog server was and is enabled, Mirror syslog to flash was enabled as was Copy syslog to flash on shutdown.

 

I have changed nfs to private with the rule outlined here and will test again now.

 

OK, applying private as above on all shares is an improvement. Small transfers all worked up to 750GB transfer which went smoothly but a 1.5TB one failed. 

 

Krusader crapped out, couldn't access the gui or ping it, didn't respond to a soft shutdown from the front panel. Nothing in the logs.

 

The 1.5TB is my last transfer so I'll just do it in smaller chunks as before.

 

Pretty comfortable now that it isn't the HW. So pressure off me.

 

For now I'll continue to use my Synology for all my backups, camera recordings etc and just use the Unraid for media.

 

I'll continue to monitor the forum for any additional tweaks I can do to make this go away as it is annoying not being able to just set a large transfer to run over night and forget it.

 

Monitoring every transfer is boring and time consuming.

 

 

So I moved all my media apps to unraid, removed all NFS shares. 

 

Still having problems. once or twice a day the GUI is unreachable and I can't ping it. Spoke to my Ali vendor who questioned i226-V support on Unraid.

 

Does Unraid fully support Intel i226-v nics?

 

What about the Marvell AQC113C?

 

I've moved it to the marvell nic for now to see what happens.

 

Are there any tweaks needed to make either nic work properly in unraid? 

 

Anyone else using these nics and what is their experience?

 

 

Both NICs should be supported, I see that you never posted the persistent syslog, it may be worth doing it, in case there's something logged there that points to the issue.

It crashed again. This time on the Marvell nic. Syslog attached but there is nothing in it of note as far as I can see.

syslog-192.168.1.5.log

Unfortunately, there's nothing relevant logged, this can also be a hardware issue, one thing you can try is to boot the server in safe mode with all docker containers/VMs disabled, let it run as a basic NAS for a few days, if it still crashes it's likely a hardware problem, if it doesn't start turning on the other services one by one, including the individual docker containers. 

 

Additionally, look in the BIOS for a "Global C-States" or similar setting and disable that to retest, it's been known to be a problem with some boards, with both Intel and AMD CPUs.

It crashed all the time long before I enabled docker. I have never enabled VMs. I'll try C-States thing but my patience is at an end now. I know that this board and all it's nics work perfectly with TrueNAS scale so if this doesn't fix it I'll just switch OS and then I will have NFS and my old workflow back as well.

OK. C-States were disabled by default all along so it's neither that or virtualisation.

 

Feel I'm at the end of the line here now.

I'm afraid that without anything relevant logged don't have other ideas, you can also try this just in case it's related, but that should not make the server crash, though it can appear unresponsive under load:

https://docs.unraid.net/unraid-os/release-notes/7.0.0/#excessive-flash-drive-activity-slows-the-system-down

This has become a nightmare that never ends. I replaced the motherboard with a similar one from a different vendor. The problems were the same but worse. I know the ram is fine so I decided to replace the USB stick. a brand new, 1st use, tested kingston 32gb with an identical one bought at the same time from same vendor. 

 

For the first few days it seemed to have done the trick and I thought I had a stable system at last to get my data off it. Yesterday I had crashes and unreachable gui and ping and today it won't boot after a crash with the waiting for device named unraid for 30 secs and then rebooting to infinity. tried putting it in different usb slots no change so now I have a dead server and no way to get the remaining 1.5tb of data off it.

 

2 failed usb sticks on an OS that mandates their use for it's business model.

 

What am I supposed to do?

 

I'm at my swap the usb limit.

 

I really need support.

You can contact support and they can do a manual replacement.

I have contacted support and explained my difficulty and all I have been told is this...

 

"For us to help transfer your registration, please send us a copy & paste of the information within Tools - Registration (no screenshots, please) when booting Unraid with the replacement flash drive. This will include your GUID."

 

I cannot boot the system, I have told support this and have received no response.

 

My data is trapped on the system until I get support and a new working USB.

 

This is unacceptable.

You need to boot with the new flash drive, not the old one:

 

1 hour ago, BuilderVN said:

booting Unraid with the replacement flash drive