Alphahelix

@sibi78 I am trying to do the same, yet I did not get it to as expected. I hope it is ok to ask to see your script. /Alphahelix

I recently upgraded to 384Gb. (reg. ecc) The reason is that my plan is to run 2-3 photoshop VMs (each with 64Gb RAM) that my local photoclub can connect to via remote desktop. Until I fix the slow disk speed I plan on putting the 50Gb vdisk in RAM and a script copying it to the RAM then power up VMs at startup. That way I also ensure fresh image every boot. So I will be using most of my resources on the VMs.

Thank you! now it makes sense to me.

I don't seem to be able to find them... I even tried to download the script file again. from https://github.com/danioj/unraid-autovmbackup/blob/master/unraid-autovmbackup-0-4.sh But I could have overseen er newer revision than the 3 year old one from the link.

Did you find the error for this? I see similar behavior. I know my disk access is more than fast enough, so the bottleneck must be another place. Could it be that the docker container is set to only use 50% of the CPU? If so anyone know where to double check and correct?

@kron I found this site, maybe it can help. Yet I will warn against modding the BIOS. But maybe it can help you. https://forums.servethehome.com/index.php?threads/nvme-boot-with-supermicro-x9da7-x9dri-f.13245/ I know it is not 100% the same board you and I have, but I believe it is very close. It looks like it is short 2 LAN ports and 8 DIMM sockets and a different PCIe socket configuration.

Thanks for your assistance.

Just out of curiosity. What BIOS version do you use? Also where do you see the bifurcationoption? I am hoping to enable this on my board. /Alphahelix Sent from my SM-G955F using Tapatalk

Ok I think I have found a way to have the script delete old files. Please correct me if I have pasted the addition a wrong place, or if there is a better way to do this. But here it is: autkvmbackup-with deletion.sh

I think it was the free version of "DeleteDuplicates"

Q I used a windows alternative. I cant remember the name. Sent from my SM-G955F using Tapatalk

Sweet build. (sorry I removed the photos to keep the length in check. are your fans loud? or did your replace them all together?

First thank you for a good script! I hope it is OK to suggest improvements. I noticed that with time stamp my backup location quickly ran into space issues. So my suggestion is to add some sort of clean-up before backing up. I have done some google search myself and found this command: find ./{backup location} -mtime +{days} -type f -delete Now I am far from being able to edit this script. But it has not held me back from trying to analyse it a bit. But to me it seems like backup location can re-use the backup location already needed to be specified in the script. The amount of days a backup should be kept safe (not deleted) looks like it could be set as a variable too. I don't know if it is possible to use a derivate of this script to only check files backed-up by the script, or maybe just xml and vdi files. Again I am not an expert here. But maybe one out here with the skills can and want to take a look at it? /Alphahelix

+1 Sent from my SM-G955F using Tapatalk

I think its utilizing the excess RAM it cant find. and I am not using all my available RAM. But it is only a theory.

To be honest after 5 days I used another solution. But yes I know I had a lot of duplicates both in number of files (200.000+) and in amount of GB (200gb+) But still I love the docker. thank you for making this possible.

I don't know if this is the right forum. I have tried this docker app, yet it halt at "Almost done! Fiddling with the results..." I gave it some time, then some more time. It is now day 3 and it is still "fiddling". In "netdata" (another docker app) I can see the dupeguru at this moment is using between 10 and 15% CPU (out of 24 cores) and 81GB RAM (out of 144GB) but NO disk activity. So it does looks like it is doing something. Does anyone know if this is normal behavior?

yet the router/bridge/AP solution requires a transformer that will suck unnecessary power to run the unit. Whereas an internal wifi-card will only use the power required for the card(s). I know it is not much, but for a 24/7-365 setup it will accumulate to noticeable amounts. That said, my setup (once all my parts arrives [read once I have the coin to buy them]) will be a managed poe switch to switch two WAN to one socket (I have two public IP) and a unify AP, to reduce power consumption.

+1 even if MC via ssh is ok, its not great. I have tried Krusader via docker, but it is no longer maintained.

I dont use them. But if I could have a saying then leave tje feature for people to fine tune their system.

Maybe a slight re-write: "one box that serves all!" Another one thought is that I have friends that buy themselves poor in Synology and Qnap boxes. With that thought this could be an idea: "You bring the hardware, unRAID delivers the rest!" If they are poor I apologize... But its rather late here in little Denmark.

144GB reg. ECC. Nice to see Unraid using excess RAM as cache.