pOpYRaid

@AdamBearWA and all here. Thanks a lot for all the infos. I also setup tailscale for my family, away from power forwarding. For emby it was as easy as changing the docker source to "lscr.io/linuxserver/emby:latest" (without quotes) and enabling Tailscale.

It seems the commit does not differentiate between hardware, but thats my assumption. Take it with a grain of salt.

Guys, finally unraid 7.3.0-rc1 fixed the performance issue on the i226-v. See here for details: https://forums.unraid.net/topic/182957-is-i226-v-nic-supported-by-unraid/#findComment-1589155 So i can recommend this card now for an low power (aspm enabled) unraid build. Without have to deal with plugins for realtek drivers 😀 Details incl. the commit see here: https://forums.unraid.net/topic/141349-plugin-realtek-r8125-r8126-r8168-and-r815267-drivers/page/18/#findComment-1620966

Guys, finally unraid 7.3.0-rc1 fixed the performance issue on the i226-v. So i can recommend this card now for an low power (aspm enabled) unraid build. Without have to deal with plugins for realtek drivers 😀 Details incl. the commit see here: https://forums.unraid.net/topic/141349-plugin-realtek-r8125-r8126-r8168-and-r815267-drivers/page/18/#findComment-1620966

Finally!!!!!!!!!!!!!!!!!!!!!!!!!!!! The i226-v driver bug i am reffering to in my last posts (performance loss when ASPM is on) is fixed in unraid 7.3.0-rc1 (kernel 6.18.23). Here is the commit which fixed the issue: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=0325143b59c6c6d79987afc57d2456e7a20d13b7 Basically disabling aspm L1.2, preventing the performance issue with minimal higher power consumption. So with this commit i can recommend the Intel i226-v for low power unraid builds and have ASPM with no performance loss working. The c-states rabbit hole has become smaller for me 😁 Sorry for hijacking this realtek thread, but in the search if there is an working alternative google has guided me here. Hope this helps somebody to escape the c-states rabbit hole.

Just a quick update: i enabled aspm for the i226-v (commenting out the code from my previous post in /boot/config/go) on newest unRAID 7.3.0-rc1 (which is on kernel 6.18.23). Confirming aspm enabled/disabled state with lspci (before and after) -> it was disabled before and is now enabled for the card. After spinning down all hdds and let it idling i am reaching C6 now and the network speed has not dropped on 10x iperf tests. Will test further and hoping the aspm issue is gone. If not will test also the suggestion of disabling just L1.1 L_1.2: https://forums.unraid.net/topic/141349-plugin-realtek-r8125-r8126-r8168-and-r815267-drivers/page/17/#findComment-1605492

I have the exact same issue, but with an Intel i226-v. Have to add the following to the /boot/config/go file (also where my other powertop tweaks are): # disable ASPM for I226-V, when this is enabled the network speed drops often to <300 MB/s (instead of full performance of 2500 MB/s) # with this workaround the system can reach c-state c3 (~15,8 Watt) and have full ethernet performance echo 0 > /sys/bus/pci/devices/0000:01:00.0/link/l1_aspm to get it stable at 2,5 GBE, sure not reaching lower c states than C3. Where ASPM is enabled for the card my system reaches C6, but with the exact same performance drop to about 250-300 Mbit/s (tested with iperf3). First few tests it reaches 2,5 GBE, but after a short idle time it has that drop issue Also tested a few years back an i226-T1 geniue Intel card and an rtl8125b which has exactly the issue. Sure my Mobo bios (Gigabyte H610i) is up to date So I will not test an rtl8125b card again, because than I have to also deal with drivers on every unRAID upgrade. Also updated my NVM flash of my Intel card (there is an open sense forum thread for that) without success. I am on newest unRAID 7.3.0-rc1 now, because of internal boot, which is on 6.18.23 now. Maybe test aspm again, but my hopes are low after all this year's without an working 2,5 GBE nic + proper ASPM support. Hope this will ever be fixed. pOpY

Yeah, that's a good point. Have moved all data into the data directory, to be as close to the template as it could be. Thanks for correcting the logo.

I answer my question after doing the migration. Yes, cache was prefilled in the wizard and uses the same mount directory /mnt/cache After moving all back to the cache SSD all is working as it should. Transfered the license to TPM and finally got rid of the flash drive. thx guys for the smooth transition.

@bartdejonckheere yeah, also for me the github doc pages gives me an error. I think we have to wait until it gets fixed. Maybe an github issue, if it's not solved soon, we can raise an issue on github.

thx for checking, mover is running... 😅 Have corrected the double license transfer. One question: currently my pool device is named "Cache" and lives under "/mnt/cache". I can see with "du -sh" that mover is working and used space is going down. When i remove the complete pool (like described in the video) and re-add it as boot/data afterwards, will the mount point be the same (/mnt/cache) in my case? (I hope it will be)

Hi guys, just asking here and want to switch to internal boot. For this i have 1x internal nvme SSD (1TB) currently used as "Cache". Watched this video for migrating: https://www.youtube.com/watch?v=sruWM1PqjVo And just wanted to be sure i dont mess anything up 😅 The appdata, system and domains are per my design just on the cache. For appdata i am using appdata backup, i have no vms (domains folder empty) and system (docker) can be rebuilt. In appdata is also the immich thumbnails and so on... this way my hdds can sleep most of the time :-) (-> yeah an real ssd cache prefetcher like primocache on windows would be nice, but that's another story. i know we have zfs but i think for my usecase its too much). Regarding transfering the license to TPM: i have checked my MoBo (Gigabyte H610i) which has an Firmware TPM2, see: root@MediaServer:/mnt# dmesg | grep -i tpm [ 0.000000] efi: ACPI=0x35313000 ACPI 2.0=0x35313014 TPMFinalLog=0x352e0000 SMBIOS=0x35a56000 SMBIOS 3.0=0x35a55000 MEMATTR=0x2f9e3218 ESRT=0x2ffc5818 RNG=0x35120018 TPMEventLog=0x35117018 [ 0.021464] ACPI: TPM2 0x0000000035121000 00004C (v04 ALASKA A M I 00000001 AMI 00000000) [ 0.021511] ACPI: Reserving TPM2 table memory at [mem 0x35121000-0x3512104b] root@MediaServer:/mnt# ls -l /dev/tpm* crw------- 1 root root 10, 224 Apr 25 10:42 /dev/tpm0 crw------- 1 root root 242, 65536 Apr 25 10:42 /dev/tpmrm0 root@MediaServer:/mnt# cat /sys/class/tpm/tpm0/tpm_version_major 2 So my plan will be: setup secondary storage Array for: appdata, domains & system (& ev. Videos but it's empty) stop docker & vm service manually run mover and wait for it to finish check with "du -sh" in /mnt/cache if all data is gone stop array go to "Cache" pool, click erase pool and remove pool start onboarding wizard and select internal boot (like shown in the video) Use "Cache" as name like before Reboot with old Flash drive inserted (so it copies the data to the internal boot) set ata partition to xfs (as it was before) start array format the data partiton move data back to nvme with Mover by reversing the direction manually run mover and wait for it to finish check with "du -sh" in /mnt/cache if all data is back (~83 GB) remove secondary storage Array for: appdata, domains & system (& ev. Videos but it's empty) start docker & vm service move license from flash to tpm Please can an unraid-pro double-check and confirm my plan? Thanks in advance Here are the screenshots: And my shares:

@ZappyZap really appreaciate your efforts regarding the template! This will push bootimus a lot. I have now swapped to your template and noticed two things: 1. data directory: it contains all data like: db, bootloaders folder, ... I think the default path should be "/mnt/user/appdata/bootimus" instead of "/mnt/user/appdata/bootimus/data". Is there a reason for the seperate /data on the host? Is it possible to change it? Because the container is relatively young, a change today will not force to much users to chnage something in their configuration. 2. the logo provided by the template is too low in resolution. Ex.: Emby docker has a logo resolution of 375x375 and bootimus 64x64 In the logos/ folder of the repo you can find higher resolution logos, like the "bootimus_logo_square_ulow.png" which is 256x256 pixel.

@ZappyZap thx a lot for creating the docker template and this thread!

Nice!!!!! Is that than available in the apps section of Unraid? Please note the logos folder in the repo, where I have uploaded a lot of different logo sizes, which you can use for the template and thread 😀

Sorry no template (or at least I don't know how to download the XML from my instance 😂). If anyone here can make one (incl. dedicated support thread), that would be really great and would push bootimus forward, like it deserves!!!!! Anyways, here is the docker output after adding/saving my docker. Basically it's just the docker parameters from the bootimus GitHub. I have used custom network with separate ip (reserved DHCP address) for bootp to work. This ip i have added to my openwrt. Please see bootimus documentation which is really good and has some examples. Here is the command where you can see/pick the details (sure replace IP, ...) and add an custom docker container in unraid: docker run -d --name='Bootimus' --net='eth0' --ip='192.168.0.X' --pids-limit 2048 -e TZ="Europe/Berlin" -e HOST_OS="Unraid" -e HOST_HOSTNAME="Unraid" -e HOST_CONTAINERNAME="Bootimus" -e 'TCP_PORT_3000'='3001' -e 'UDP_PORT_69'='69' -l net.unraid.docker.managed=dockerman -l net.unraid.docker.icon='https://raw.githubusercontent.com/garybowers/bootimus/logos/bootimus_logo_square_ulow.png' -v '/mnt/user/appdata/bootimus/':'/data':'rw' -v '/mnt/user/Daten/ISOs/':'/data/isos':'rw' 'garybowers/bootimus:latest' Hope this helps.

Hey guys. Because development of iVentoy seems to be stale and also there was the privacy issue, gary bowers (previous maintainer of iVentoy docker) started an own 100% open source replacement. Please give bootimus a try: https://github.com/garybowers/bootimus It's in an early stage but Gary is doing an fantastic job!!! Currently running as docker on my unraid and it can boot much more than iVentoy ever did for me. Don't forget to hit "Extract" on evety ISO details page to enable direct kernel booting. P.s.: I am helping gary on testing and did some small contributions.

You're welcome 😊 Nice to hear that it also works on other mainboards.

I am gave up on this with my H610i Gigabyte board! Or in other words, i have gave up on gigabyte. Their answer ist just they dont suppport linux. I was also stating it happens on windows. But their support driving my crazy. So my workaround is a user script Beep Fix (Standby for 5 seonds after boot) with the following content. It basically goes to standby for 5 seconds after 120 secs on every unraid reboot. This way the beeper works reliable. #!/bin/bash # sleep 120 seconds after boot, so system can settle down sleep 120 # standby and wake it in 5 seconds, with this workaround the beeper is working! thx intel rtcwake -u -s 5 -m mem # sleep again 20 seconds, so we can play our fancy sound sleep 10 # ----------------------------------------------------------- # from now on beeper is working!!!!!!!!!!!!!!!!!!!! # ----------------------------------------------------------- # play array up beep beep -l 300 -f 600 -n -l 300 -f 800 # disable ethernet leds, sadly after wakeup they are on again echo 0 > /sys/class/leds/igc-0100-led0/brightness echo 0 > /sys/class/leds/igc-0100-led1/brightness echo 0 > /sys/class/leds/igc-0100-led2/brightness

Are you guys using the official netboot.xyz docker? I have "Permission denied" error on every image it wants to pull from the internet during boot. When i store them as local assets, it works, because not certificate is needed/checked with http. It seems an https issue but i dont know. Here is an similar issue on github: https://github.com/netbootxyz/netboot.xyz/issues/1597 If anybody running an working container, please let me know your settings, menu_version, ... thx

Update. I updated my I226-V card from NVM 2.13 to 2.32 and the issue is not solved. See post #117: https://forum.opnsense.org/index.php?topic=48695.117

@runeRaid i have tried several Intel cards (i225-T1 genuine intel, I226-T1 genuine intel, I226-V china brand, ...) all had the same issue. When i activate ASPM for the card, i maybe got offline after a short idle time or the bandwidth dropped from 2,5GBE to ~300MB/s. After i disabled ASPM, the cards where all rock solid and always delivers 2,5 GBE bandwidth (meassured with iperf3). Can you please post a link to an 2,5GBE intel card with working ASPM? (the one you have) P.S.: Posted this already on several places, like: OPNsense Forum[solved] Intel i226 Firmware (see post #39)[solved] Intel i226 Firmware (see post #39) OPNsense ForumController Reg MapsController Reg Maps https://github.com/fenrus75/powertop/issues/143 the opnsense looks interesting, they actually provide eeprom/firmware files! I did not tried flash my adapter because i am need it :-)

thx, that did the trick for me as well on unraid with karakeep 0.26.0 and browserless-v2

Just want to say thank you! Treid netboot.xyz first and got it to boot, and always got permission error. Tried increas RAM... nothing helped. After fiddling around with local assets i got clonezilla and gparted to boot, but they just freeze early in the boot process. With this docker i can finally boot everything over network. thx

Thx, now it's clear to me