DoItMyselfToo

Members
  • Posts

    340
  • Joined

Everything posted by DoItMyselfToo

  1. During the past couple of years (COVID years) the price of drives, even shuckables, has been much higher. Whether this is related to COVID, Chia mining, etc., doesn't really matter to the home server enthusiast. So what happens when a disk fails but you don't have another disk to replace it with? Wouldn't it be cool to be able to spread that data over the free space of the remaining disks in the array? It would be very helpful to be able to recover the data from a failed disk (emulated via parity) to free space on another disk(s). I saw a post in another forum about SnapRAID having this functionality and would love to see something like it in Unraid.
  2. this morning SAB w/ wireguard would not connect even though it has been working fine for months. so then changed back to OVPN and worked flawlessly but slower. kept getting this error: "Unable to successfully download PIA json to generate token..." will retest tomorrow. suspect PIA was messing server side.
  3. Keep in mind that the E-2244G (and other Xeons in this class) have a limit of 16 pcie lanes. As for risks transferring to different chassis, basically every time one moves stuff around there is a risk of messing something up. Maybe inadvertent static discharge, or improperly seated cables, or a bad cable. Whatever. I have found that once my system is up and working, I do better to NOT tinker with it.
  4. That Xeon (E-2244G) is a 9th gen processor, which is 4c/8t. They sell for just under $300 and support ECC RAM. You could also get an i3-9100 (supports ECC and has Igpu for transcoding), which is 4c, so no hyper-threading. But the cost is about $120 on the eBay used market, but can also be purchased brand new for about $150. As for motherboards, I'm not familiar with the Asus board,. Some of us here use the ASRockRack E3C246D4U, which overall has been very good. See this forum post for more information. If you go this route, the board can be purchased separately from the eBay seller listed in that post. You could message the seller and ask how much for just the motherboard with RAM, which is what I did, allowing you to source your own processor. Also, RAM should be installed in pairs, so you'll need two sticks. You mention wanting 16gb. I would aim for 32gb (2x16gb) because you'll probably end up using it. You do not need ECC RAM w/ Unraid. But if you want it, that's totally cool. Just be sure that the motherboard and cpu support it. I would go with a 500gb SSD over the 256gb, because metadata for something like Plex or similar programs can grow quite large (depending on settings) and if you use the cache drive for downloading or have a bunch of docker containers, you'll want some headroom. Also, a larger drive should offer better endurance due to the ability to wear level over a greater number of cells. A 500-600w power supply should be plenty. I've seen systems 3x your goal operating in the 225-275w range and never more than 350w. But you do want some overhead to account for startup and also expansion. So IMHO 500-600w is a nice size. As for chassis, I'v had several Fractal cases and find their quality to be excellent. Just be sure to get something that offers enough drive slots for the disks you want to start with and more for expansion. It takes a fair bit of time to install the server components into the chassis. Swapping chassis is a pain. And also opens the possibility of causing problems. Sometimes it's better to "buy once and cry once" than to mess around. Good luck. Building one's first Unraid server is exciting and rewarding. Enjoy!
  5. Heard back from WIlliam at Asrock, who said that there is no reason to change to the 2.34 bios if the 2.21a is running fine. Here is an exact quote: ”There is some minor modification in the BIOS to give little more support for a certain PCIe video card and there isn’t an additional feature added that would suit your need. If your system is current running fine with the current BIOS, you don’t need to use that beta BIOS. Thank you!”
  6. I really like these recommendations, which all have been previously implemented in my setup as a result of either watching SIO videos or various UnRAID forum posts. But having this security info in one thread really makes better sense than users having to assemble it on their own. Also, as another user posted in this thread, UnRAID should default to more secure settings out of the box. I’m not an IT professional, but my reading of the forums suggests that many if not most computer problems come from user error. I suspect security issues even more so.
  7. thank you. this makes sense. i seem to have tracked my problem down and the disk rebuild seems to be fine, though knowing more about this can only help in future. one "bad" thing about Unraid is that it works so well that when it doesn't i cannot remember how to fix anything. lol.
  8. so to do this looks like one would have to stop the array, remove the new disk. and replace with the old disk. and then use the "recipe" from Unraid 6/Storage Management to reset the array configuration in maintenance mode. anything else?
  9. i reseated the rear backplane connector, in both the backplane and hba then double checked all other connections. it's been nearly 7 hours since the rebuild started and so far so good. i am still interested to learn the answer to my original question if anyone knows more about parity syncing/rebuilding. thanks in advance.
  10. help please: my new to me Supermicro 36 bay chassis was running fine for a number of days; decided to upgrade the size of a disk; did the normal disk upgrade process: 1. stop array 2. unnassigne smaller disk 3 assign larger disk to same slot 4 start array (note: all dockers/vms off.) disk upgrade/rebuild started. everything seemed fine. then during the first hour a few UDMA CRC errors appeared on 6 disks that are in the rear backplane. prior to this there were no disk connection issues. so, i acknowleged the errors and watched if more would appear. got a few more on each of the same 6 drives. so, i paused the rebuild. powered down. and checked hardware to find the problem. my understanding with UDMA CRC Error Count errors is they are an indication that a physical connection between the drives and host is compromised in some way, often being bad cables or poor connections. i think the issue is with the one SFF-8087 to SFF-8087 cable that connects the rear backplane to the sas hba. it two spots on its length that are suspect, especially when compared to the other cable. the other cable connects the same sas hba to the front backplane, with no problems there. the connectors on the backplanes and sas hba all appear fine and are tight. (note: no dockers/vms were run at all during this process.) (2nd note: i have a replacement cable ordered, plus a couple extras for spares. but i would like to run the array in the meantime. and the normal reading of the array disks did not induce UDMA CRC errors.) Q: can i just unassign the new disk and replace the old disk and not have parity issues? thank you in advance.
  11. i've been running the 2.21a bios and 1.80 bmc since i got my E3C246D4U. seeing there is a 2.34 bios. anyone know what the changes are from 2.21a? anything critical? microcodes?
  12. had similar problems as others with Sonarr after updated SABnzbdvpn. added Unraid server to "ignore addresses" field in Sonarr (settings - general - proxy). then restarted SAB, NZBHydra2 and Sonarr. still wouldn't yield search results. so went into Sonarr indexer and downloader settings to "verify" them, though environment did not change, so settings were not changed. then clicked "test" for each. after testing both, now searches work.
  13. New - I was going to use in a second server build. Tested; work great. Four sticks available. Prefer to sell in multiples of two. $160 each, shipping included. Shipping: US lower lower 48; fast shipping. (Willing to ship to Canada if buyer pays shipping difference.) Paypal. Post here; or PM - questions, interest.
  14. Happy Birthday! Cheers!
  15. thank you for offering your input. i appreciate this. i always update all sas controllers to the latest (20.00.07.00) firmware before running in my server. the cables are new. still problems with this disk as an UD. but i am currently preclearing this disk. did a full read of the disk and now at 20% zeroing without any hardware change. update: the disk pre-cleared without issues. but i did double check the firmware on my two LSI SAS HBA's. the 8i was updated to the latest (20.00.07.00) but the 16e was not (20.00.06.00). my bad. not sure how i missed that. but thank you again to @JorgeB for their post.
  16. changed the UUID in UD. and did a little bit of reading up on this. post if more issues. thank you.
  17. how can i figure out what is causing the mount points to duplicate? i just deleted the mount point and reformatted the disk. the same mount point returns.
  18. the drive is running off a 9201-16e via sas-to-sata breakout cables. prior to this issue never had a problem. a bunch of drives running off this controller now without any problems. for some reason this drive in UD is a problem. drive itself passes smart tests.
  19. this disk continues to be a problem in UD. ever since this issue, i cannot use this disk in UD. it appears fine, then i copy files to it. the disk then drops out, in, out, in, out, etc. then some kind of error. here is one i got today.
  20. no, i did not enter a "?" mark in the mount point. a physical disk. so, the "?unassigned_disk4" mount point does not show up in the UD. it is the one i changed. new is in circled in red above. but i can see old one when navigating (/mnt/disks/) Krusader or MC. but cannot delete it due to permissions. was going to change permissions, then delete but wanted to get your input first.
  21. i changed the mount point of an unassigned disk. noticed that the old mount point with a "?" mark in front of it is showing when browsing MC and when browsing Krusader. how to get rid of this? thanks.
  22. testing privoxy in this docker was getting dns different than endpoint using wireguard. found these pia nextgen gen custom dns here. edited container, adding 10.0.0.242 as first dns. now dns leak test shows dns matching endpoint.
  23. need to add a host path to this container. here's an example: otherwise everything else as default. works just fine. thanks to the linuxserver.io team for another excellent container.