rorton

Just moved my disks into a new server build - its all come up good or so it seems, disk are now on a HBA controller, new cache drive running NVME on an m.2 slot and some unassigned drives, all dockers running vm good etc. The only odd thing i just noticed, was my schedules are gone - i didn't have a lot setup, a schedule for the mover, a schedule for trim, and another for parity, but got me concerned why they didn't carry over and if there is anything else i may miss that didn't carry over? I bought over the same USB flash drive - is there anywhere i can check what my old settings were? I took a copy of the flash before i put it in the new build if that helps? Thanks

Thanks - i think the non plus is the one prior to the plus, so fingers crossed should be OK! Otherwise it can go back to Amazon!

I was going to get a but did a quick search on the forum and noted people having trim issues with this brand. Started to look for other brands/drives, and its a minefield. I sort have focused now on a Samsung 970 EVO Plus 500 GB NVMe M.2 Internal SSD 500gb drive to use as cache - will trim be OK with this?

Cheers. I’m in the uk unfortunately. I was looking at that drive though on amazon, I may go for the 512gb model. Read.write says around 3000MB/s, unraid handles that ok?

Ok cheers. I’m using the plugin to backup app data already, need to see how vm backups can be automated. No idea why, but my current cache as a stand-alone is formatted as btrfs

cheers - so don't bother with a cache pool then?

Upgrading current system (new m/b cpu ram) currently have a single 250gb SSD as cache in running system. New motherboard has 1 Fast NVME m.2 slot, and a slower m.2 slot. My initial thought was - buy 2 slower m.2 drives (500gb) and put them in a cache pool - then stick the old 250gb ssd in as an unassigned drive - job done. Then i got thinking that i was wasting the speed of theNVME slot, so should i put a fast single m.2 in the NVME slot, and then buy a second 250gb 2.5 SSD to match the existing - put the 2 2.5 SSDS in a cache pool, and then have the NVME as an unassigned drive that will be fast Or do i just put in 2 slower ssd's on the motherboard m.2 slots, add in the existing 250gb 2.5 SSD, and pool all of them Cache used as a download folder for sabnzb before getting moved to the array as part of sonar and radar, docker image is here, and a couple of VM's

Thanks, that’s a great help. My shares are set to cache only, not prefer, should I set them to prefer prior to copying the data, is that the advised way to run shares you only want on the cache drive and thus keep them on prefer when I have the m.2 cache pool running

Sorry, replied before I saw your second reply. ok, so I use the mover to move the data to the array on the old hardware, move disks to the new hardware. Introduce the m.2 drives and set them up as cache pool, then change the relevant shares again to be cache only, and mover will move all the data back? i plan to then wipe the old 256gb ssd and put that back in the system, no idea what for, perhaps an unassigned drive. Seems a waste not to use it?

Thanks, sounds much simpler your way:) do I need to use the automated backup restore app data plugin, or just manually copy the data, and manually copy it back? I have app data, docker image, a download folder where temp downloads go to, and a vm folder with my vms and the libvert image. Not a big cli guy, so unsure if I can just grab the lot, move it, then move it back so it all works

Will be moving into a new ryzen based system. I understand I just plug the existing array disks in, boot, and everything will work on the new hardware? I have 4 3tb disks in the array and a 256gb ssd cache. I want to make the cache a pool, using 2 x m.2 ssd on the motherboard, am I best moving all my old disks to the new motherboard/cpu/ram/case and getting it running using the existing 256gb ssd cache, Then put in the m.2 drives and move the cache data to them? what’s the best way to do that? Backup the cache to the array, then restore it to the m.2 cache cluster once configured?

On the quest to building a new unraid box, and currently looking at a Ryzen 3400g and suitable motherboard. I had an idea of getting a board with dual m.2 slots, and sticking in 2 NVME drives (512gb) to create a cache pool. I dont know a lot about ryzen, but been reading stories about how the 2 slots run at different speeds due to PCI Express lanes - so question is, will this work ok? I guess i could run 2 2.5" SSD drives, just fancied the nvme drives on the board, 2 less sata and power cables.

Hi, still for sale?

I have an old HP Microserver with a AMD Turion II Neo N40L Dual-Core @ 1500 MHz cpu and 8gb ram, with 4 x 3tb disks. Had it for 9 or 10 years CPU is struggling now, i have a number of dockers running, a VM running some SNMP monitoring. Its gotta be 12 years since i last built a PC - i used to build a lot, but have next to no knowledge on todays hardware Im looking (i think) to goto a 6 or 8 core XEON cpu (is the number of cores important?) - great auction site sees things like - Intel Xeon X5675 Six Core 3.06GHZ Processor SKT LGA1366 Very Fast CPU for £35 which seems worlds ahead from what i have now - is this very power hungry/any good or should i be looking at something different? I will be running stuff like SAB, Sonnar, Raddar, Unifi in dockers, one or 2 Linux VM's running LibreNMS or Zabbix and was hoping to run a Win 10 VM - really just to run some windows based apps as I'm mainly a mac user, so its needs to be just a basic user type win 10 machine. Where do i start? Id like a smallish case like a Silverstone D380, but that restricts me too motherboards sizes? Anyway, if i can get an idea of CPU/Board/Ram, then i can move onto the other parts of the forum for SATA cards if its best not to use the motherboard ones?

ah gotcha - good luck

cool I just switched SMB on, re mounted my drive, and pointed time machine to the same share that way all the history was there, and i could go back in time over 3 years if i needed

Sorry mounting is the correct terminology, connect to server, enter smb://ipaddressofnas

I had similar issues but it’s been rock solid since I changed. The key was that once smb enabled, you have to remap the share, then time machine will see it.

AFP has been depreciated, so eventually won’t exist

thanks for this - works perfectly

hi, yep, it worked fine, and its now seen all the old data, so im really happy. I think for me, the bit i was missing is initially mapping then as share on the mac - i just opened time machine and expected it to appear - makes sense i guess, but after i mapped it, and the mac knew about it, it appeared, and worked as expected!

yeah, if i could help it, i didn't want to blow away the 3 years of backups i have on there, I've started it off, and it seems to have found it i think, has half hour left to run, so will report back later and let you know the status Seems that TM only understand what's on the disk when it starts the backup, goes off to interrogate or something like that!

With your tip about mapping the drive first - i too can now see my SMB drive in time machine - that was the key there! Only problem i see now, is that I'm setting it to backup, and it doesn't seem to see the data that's on there, so i have to create a whole new backup which doesnt seem right

Im struggling to get my SMB time machine share to appear also. Ive done all the settings in SMB and when I go into Time Machine, there is no share to find - if i switch on AFP - it works fine. Ive tried rebooting unraid - rebooting the mac in question, neither makes the share appear

thanks - where does all the data go then, i dont get that in 12 hours, my LibreNMS VM has written 14.55 GB worth of data according to this iota app, and yet the available space on the SSD hasn't reduced by 14GB, Unifi has apparently written 7gb etc,