DivideBy0

Not to be a debbie downer but I wonder if this is going to cannibalize the free support from UnRaid we get on this great forum Why offer free when you can get $$$ for it But I understand this is capitalism and $$$ must be made somewhere Again nothing negative about it, great idea if you can afford it.

Something is going on with this new build system, version: 6.12.4. My CPU fun was doing 9000 RPM so I got an alert. Got into the system and notice CPU running high and lot of "timeout" processes, 8663 of them? I stoped all dockers and managed to reboot gracefully. Attached my diagnostic and the ps -ef > timeout.log root@NAS-UNRAID-2:/boot/logs# cat timeout.log |wc -l 8663 201 32741 25606 0 15:50 ? 00:00:00 [timeout] <defunct> 201 32749 25606 0 17:08 ? 00:00:00 [timeout] <defunct> 201 32752 25606 0 17:48 ? 00:00:00 [timeout] <defunct> 201 32754 25606 0 16:54 ? 00:00:00 [timeout] <defunct> 201 32757 25606 0 17:15 ? 00:00:00 [timeout] <defunct> 201 32758 25606 0 15:20 ? 00:00:00 [timeout] <defunct> timeout.log nas-unraid-2-diagnostics-20230927-1814.zip

I could play with the syslog file size? Now is set to 5Megs. Maybe if I drop it to 1 or 2 Megs will work better? Thinking if I limit the size then syslog will start writing new files on cache and when the movers kick in it will just move the completed ones. Well I ended up with a custom script that does it all for me daily. #!/bin/bash sh /etc/rc.d/rc.rsyslogd stop for file in /mnt/cache/NVME-LOGS/*; do mv "$file" "/mnt/disk7/LOGS/$(date +"%Y%m%d_%H%M%S")_$(basename "$file")" done sh /etc/rc.d/rc.rsyslogd start

Or write a script to daily: Stop syslog Execute Mover Start Syslog

Anybody?

Here is my intended scenario that partially works now. I am using this UnRaid as my main syslog repository with syslog server configured to store logs into a dedicated share named LOGS. Fort that I configured a share LOGS with "primary storage" to cache and "secondary storage" to array and 'mover action" cache -->array scheduled weekly. My expected behavior will be that syslog will constantly write to cache and not array therefore my array & parity drives won't be spinning all the time, except when the mover schedule take place every week. But that is not the case. I have about 10 devices sending logs to this UnRaid and some of them are written to cache and some of them to the array, they're are very inconsistent and some of my array and parity drives are constantly spinning. Am I doing something wrong or this is the expected behavior? Any advise or guidance please

Just looking at my logs and I see this? But why as I have no VPN configured? [ 103.260366] docker0: port 3(veth435a22c) entered blocking state [ 103.260370] docker0: port 3(veth435a22c) entered disabled state [ 103.260395] device veth435a22c entered promiscuous mode [ 103.260426] docker0: port 3(veth435a22c) entered blocking state [ 103.260427] docker0: port 3(veth435a22c) entered forwarding state [ 103.386306] eth0: renamed from vethb126c12 [ 103.408346] IPv6: ADDRCONF(NETDEV_CHANGE): veth435a22c: link becomes ready [ 127.405858] elogind-daemon[1594]: New session c2 of user root. [ 192.838314] elogind-daemon[1594]: New session c3 of user root. [ 192.839703] elogind-daemon[1594]: New session c4 of user root. [ 323.633381] wireguard: WireGuard 1.0.0 loaded. See www.wireguard.com for information. [ 323.633383] wireguard: Copyright (C) 2015-2019 Jason A. Donenfeld <[email protected]>. All Rights Reserved. [ 8046.123683] elogind-daemon[1594]: New session c5 of user root.

Just making sure and many thanks Jorge

I only get one of this error after the boot, just one, no more after. Should I worry? Is just noise? [ 179.871820] mpt3sas 0000:01:00.0: invalid VPD tag 0x00 (size 0) at offset 0; assume missing optional EEPROM

Thanks JorgeB. Worked like a charm. I now have a screaming new UnRaid server and all my data is intact. I am doing a parity check just to be sure nothing is corrupted.

For the love of my life can't figure out how to remove this old cache, named "Cacheold"? I have no drives assigned to it anymore as I added new SSD drives to the other new cache. I can't see a delete button anywhere?

Thanks Sir. I used 2 SSD's as my parity and array and did a quick sync . But is good to know where the logs are now.

I am reseting this UnRaid this system to sell it and since the buyer is new to UnRaid I had to setup few things for him on a new UnRaid demo USB drive. I have a need to to reset the Parity error count and I was hoping when you do the New Config that will take care of it. Which file is keeping these counts and can I manually deleted it? Or any other procedure? I don't want to just blow the entire USB drive with a fresh install as I have few custom things setup for the new user.

Thanks guys

Arrrrgghhhh silly me. It was disabled in BIOS. Loud face palm

I have an Intel® UHD Graphics 770 on my i9-12900K. Just recently built a new system with this mobo ASUS Pro WS W680M-ACE SE and I am new to the iGPU thingie. I am not seeing a GPU in the system, no idea why? Do I need any special plugins, configuration or anything? So far installed the Intel GPU Top and GPU Statistics Do I need the HDMI Dummy Plug 4K Display Emulator? and lspci gives me this? I am not seeing the i9-12900K CPU quicksync GPU here, the 770? root@U-NAS:/dev/dri# lspci -Dmm | grep VGA 0000:09:00.0 "VGA compatible controller" "ASPEED Technology, Inc." "ASPEED Graphics Family" -r52 -p00 "ASPEED Technology, Inc." "ASPEED Graphics Family" I have no GPU in my system display ? Where do I start?

LOL, you fast, you beat me to it So yeap that line in syconfig fixed it Should I be worried about this error? [ 179.871820] mpt3sas 0000:01:00.0: invalid VPD tag 0x00 (size 0) at offset 0; assume missing optional EEPROM

I am setting up a new system based on Asus Pro WS W680M-ACE SE motherboard with an LSI 9240-8i FW:P20 IT Mode controller. I've seen some UnRaid forum posts with successful UnRaid implementations of this mobo and so I got everything working correct except the last part adding the LSI controller and few SAS drives. I get these errors and not sure if is something missing or needs changing in my BIOS? Despite these errors that continuously run in my syslog (tail -f /var/log/syslog) I can see the drives and assign them and even do a parity sync right now. But with these constant errors in the background UPDATE: Researching this --> [ 1005.824110] pcieport 0000:00:01.0: AER: Corrected error received: 0000:01:00.0 [ 1005.824125] mpt3sas 0000:01:00.0: PCIe Bus Error: severity=Corrected, type=Physical Layer, (Receiver ID) [ 1005.824130] mpt3sas 0000:01:00.0: device [1000:0072] error status/mask=00000001/00002000 [ 1005.824135] mpt3sas 0000:01:00.0: [ 0] RxErr [ 1005.837359] pcieport 0000:00:01.0: AER: Multiple Corrected error received: 0000:01:00.0 [ 1005.837379] pcieport 0000:00:01.0: PCIe Bus Error: severity=Corrected, type=Data Link Layer, (Transmitter ID) [ 1005.837384] pcieport 0000:00:01.0: device [8086:460d] error status/mask=00001000/00002000 [ 1005.837389] pcieport 0000:00:01.0: [12] Timeout [ 1005.837395] mpt3sas 0000:01:00.0: PCIe Bus Error: severity=Corrected, type=Physical Layer, (Receiver ID) [ 1005.837399] mpt3sas 0000:01:00.0: device [1000:0072] error status/mask=000000c1/00002000 [ 1005.837403] mpt3sas 0000:01:00.0: [ 0] RxErr [ 1005.837407] mpt3sas 0000:01:00.0: [ 6] BadTLP [ 1005.837410] mpt3sas 0000:01:00.0: [ 7] BadDLLP [ 1005.837412] mpt3sas 0000:01:00.0: AER: Error of this Agent is reported first [ 1005.843141] pcieport 0000:00:01.0: AER: Corrected error received: 0000:01:00.0 [ 1005.843162] mpt3sas 0000:01:00.0: PCIe Bus Error: severity=Corrected, type=Physical Layer, (Receiver ID) [ 1005.843167] mpt3sas 0000:01:00.0: device [1000:0072] error status/mask=00000001/00002000 [ 1005.843172] mpt3sas 0000:01:00.0: [ 0] RxErr [ 1005.862001] pcieport 0000:00:01.0: AER: Corrected error received: 0000:01:00.0 [ 1005.862018] mpt3sas 0000:01:00.0: PCIe Bus Error: severity=Corrected, type=Physical Layer, (Receiver ID) [ 1005.862022] mpt3sas 0000:01:00.0: device [1000:0072] error status/mask=00000001/00002000 [ 1005.862027] mpt3sas 0000:01:00.0: [ 0] RxErr

Awesome and thanks Jorge. Do you sleep by the way

And one last question LOL. I plan to use a different HBA Controller in IT mode for my new system, but same vendor, LSI. Moving the array + parity drives to the new system should still retain the drive ID and assignment correct, so the array and parity will come up without issues in the new system? I will also make copies of the drive assignment so if it does not come up automatically I can manually assign them? I really appreciate the guidance on this gents.

Hey quick question on the Docker. When I disable the Docker should I also select to delete the /mnt/user/system/docker/docker.img before moving my array+ parity drives to the new system? And when I re-enable docker on the upgraded system will it download the latest docker.img?

Copy that Sir and many thanks So you can actually tell the mover to move from pool to array? I didn't know that The reason for my settings on step 4 was to inform the mover what to do What do I know I did change the steps to reflect the correct order so others can see the right solution.

I wanted to make sure I got my procedure correct. I am planning to move all my data divers (array & parity) but no cache/pool drives to a new upgraded system. The reason I am leaving my pool drives behind is because I will be using new & bigger size cache drives in the new system. So System-1 is old system and System-2 is new system and these steps are purely to move all my cache data to the new system and leave my old cache/pool drives behind. In System-1 I currently have the shares appdata, domains and system, setup as primary storage -> cache and secondary storage -> array. And here are my planned steps: 1. Shutdown all dockers and I have no VM's running 2. Disable VM 3. Disable Docker 4. Change the 3 shares (appdata,domains, system) to primary storage -> array and secondary storage -> none 4. Keep the 3 shares (appdata,domains, system) to primary storage -> cache and secondary storage -> array and mover action set to move from pool to array 5. Invoke mover, wait for it to finish and verify data indeed was moved. 6. Move all drives (array & parity) to the new System-2 ensuring drives assignment is correct 7. Setup the new Cache / Pool drives in System-2 8. Redo step 4 but with primary storage -> cache and secondary storage -> array 8. Same as 4 but reverse mover action. 9. Invoke mover and move all data in System-2 10. Enable Docker and VM and restart all dockers. Did I get it right?

Thanks gents. I am paranoid but I have backups. The question is do we really know when corruption happens due to non-ECC ram? Because if data is corrupted you will copy corrupted data to your backups, right? I do file integrity checks but I want to be 1000% I am not copying corrupted data to cloud.

I am building a new UnRaid based on the on the Raptor Lake CPU and I have a mobo that supports ECC or non-ECC memory. Which one should I get and is there a benefit to ECC memory over non ECC? Obviously the ECC is almost double in price if I go that route. Is it worth the extra $$$ ?