nealbscott

Members
  • Posts

    22
  • Joined

  • Last visited

About nealbscott

  • Birthday 03/26/1966

Converted

  • Gender
    Male
  • Location
    Manassas, VA

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

nealbscott's Achievements

Noob

Noob (1/14)

0

Reputation

  1. Currently running Machinaris v0.6.8 on my cache drive, /mnt/user/appdata/machinaris, which is spinning rust. I get plenty of good connections. 8 pages of them in fact. Problem is after a few hours, maybe 4 to 8 hours, more and more of them decide to get an 'SB Height' of 0 when they certainly don't start out that way. Whether they just 'go' to zero, or good connections drop to be replaced with bad... who knows? I havent even begun to actually make new plots. I just wanted to farm the plots I copied from my Windows machine. So may have to get an SSD anyway. I fully expect that running all this on spinning rust is sub optimal. But come on, I expect that the only penalty should be slowness, not failure. I see a bunch of this: "It took 10.17s to apply 816 additions and 10 removals to the coin store. Make sure blockchain database is on a fast drive" I know the blockchain database is not on a fast drive. Heck the wallet is not either. I know my system is older-ish, but 32gb ram, 1gb service from Verizon Fios, and an i5 going 3.4mhz should be enough, right? Sure smells like someone did not develop Chia while on spinning rust or they would have seen this problem. Why do my plots go to an 'SB Height' of 0? My system only shows 10% ram used. Is there any way to tell this docker, or sqllite to go ahead and use more to speed things up? My unraid 6.9.2 system: Model: N/A M/B: Gigabyte Technology Co., Ltd. Z87X-UD5H-CF Version x.x - s/n: To be filled by O.E.M. BIOS: American Megatrends Inc. Version 10c. Dated: 06/12/2014 CPU: Intel® Core™ i5-4670K CPU @ 3.40GHz HVM: Enabled IOMMU: Disabled Cache: 256 KiB, 1 MB, 6 MB Memory: 32 GiB DDR3 (max. installable capacity 32 GiB) Network: eth0: 1000 Mbps, full duplex, mtu 1500 eth1: interface down Kernel: Linux 5.10.28-Unraid x86_64 OpenSSL: 1.1.1j Uptime:
  2. 6 days ago I had GPU folding working great. Today I see: 2:58:36: GPU 0: Bus:1 Slot:0 Func:0 NVIDIA:7 TU116 [GeForce GTX 1660 SUPER] 02:58:36: CUDA: Not detected: cuInit() returned 804 02:58:36: OpenCL: Not detected: clGetPlatformIDs() returned -1001 so therefore my GPU folding is broke and flagged as disabled. I deleted/reinstalled the docker (keeping and reusing the FAH appdata folders) and it made no difference. So I am left wondering, did UNRAID NVIDIA stop working when they pulled the plug on it? I didnt think that was supposed to happen? If the only fix here is to install the unraid beta w/the new nvidia plug in, then is the beta ok to install on top of nvidia unraid?
  3. I love that unraid works with almost any hardware, and can upgrade with you. But raid is not a backup. Can you make it easy for unraid servers to back up to each other? Even better over the internet for off-site backup goodness.
  4. My HighPoint Rocket 2720SGL, also Marvell based, had a nasty habit of not detecting one of its drives at boot time. But was very happy if you plugged in the sata cable AFTER the whole system was booted up. I gave up on it and replaced with an LSI card from ebay.
  5. You can also use the 'unassigned devices' plug in. It will let you mount an SMB share, so basically its like mapping a network drive in Windows. Be sure to launch unraid server#1 (the one with unassigned devices plugin) in GUI mode. Make an SMB share on server#2, then go to server#1 and mount your new share. Then in the bottom left corner you will find MC, Midnight commander. It will happily copy files around from one machine to the next.
  6. Run the plugin 'fix common problems' and see if it finds anything. I had files on my cache drive, but the share they belonged to was not enabled to use that cache, therefore the mover never did anything with them. The good news is that the plugin caught it.
  7. So I was thinking about bit rot and the small small possibility of data files being corrupted and having no way to even know about it. Silent bit rot. I dont think XFS uses CRC checksums to detect it. Right? Wrong? And I was going to bone up on using Dynamix File Integrity plug in to guard against it. Then it occurred to me that an encrypted drive might do the same thing. A single flipped bit from a nearby magnet (or cosmic ray!) should cause a failure of some sort and you'd at least know something was wrong at the encryption level. I know at work we use an awful lot of Bitlocker full disk encryption, and seldom ever see any bit rot on the 1000's of hard drives we have encrypted. From this I know bit rot is a very minor issue. But still, it seems to me that full drive encryption is a super way to guard against it, assuming the decryption process can self detect success and failure. Would there be any reason to still use the Dynamix File Integrity tool? Is there any way to validate a drive fully decrypts OK?
  8. ICY DOCK FatCage MB155SP-B 5 in 3 cage: $90 https://www.amazon.com/ICY-DOCK-MB155SP-B-Hot-Swap-Backplane/dp/B00DWHLFMA/ref=sr_1_3?keywords=MB155SP-B&qid=1558546593&s=gateway&sr=8-3 less $25 May 2019 rebate https://www.icydock.com/product_news_info.php?id=741 brings it to $65. I am very tempted.
  9. I also have the Antec 1200 and was looking at the Icy Dock Fatcage 5 in 3 https://www.newegg.com/Product/Product.aspx?Item=N82E16817994155&cm_re=FatCage_MB155SP-B-_-17-994-155-_-Product But I am waiting to see if there is any February promotion: https://www.icydock.com/promotions.php?type=monthly That said... I'm open to any reason to NOT do that too. This Norco server case sure tempted me. Norco RPC-4224 https://www.newegg.com/Product/Product.aspx?Item=N82E16811219038&Description=Norco RPC-4224&cm_re=Norco_RPC-4224-_-11-219-038-_-Product This video review of it is from 8 years ago, but it appears the case is still available:
  10. FYI there were reasons SMB 1 keeps being turned off in Windows. Namely, its an old vulnerable protocol that was exploited by WannaCry virus. https://www.zdnet.com/article/windows-10-tip-stop-using-the-horribly-insecure-smbv1-protocol/
  11. For a single NTFS based drive, well, connect that drive to your unraid system. Use the 'unassigned devices' plug in and mount that extra drive into your unraid system. At the unraid prompt, type: mc This will run Midnight Commander, a text based front end that makes it easy to copy files around. The only gotcha I think was a few folks found it difficult to mount NTFS partitions into unraid. But if your Windows system is using some hardware RAID (even JBOD), then all bets are off. Taking a single drive out would not neccessarily be readable in any machine, let alone an unraid machine.
  12. Oh I certainly do know one unclean shutdown. Blinking clock radio a few weeks ago in the house tells me we lost power. Yet my other unraid server has no parity errors. I fully expected the drives were probably already spun down at the time of the power outage, since both systems were probably idle. OK, new mental note... good idea to manually run parity check after unclean shutdown.
  13. Its still confusing to see 'parity valid' and also see errors listed. Diagnostics attached, thanks for looking! tower2-diagnostics-20161026-0641.zip
  14. OK thats confusing. I did a noncorrecting parity check last night. I'm told I have valid parity in one part of the web page, but also told the 'last check completed' found 396 errors. Thats confusing. Is it telling me that there is a drive with errors, and that the errors will be recreated if there is a data drive rebuilt? I have about 100 lines of this in my system log: Oct 25 18:38:53 Tower2 kernel: md: recovery thread: PQ incorrect, sector=81804808 Oct 25 18:38:53 Tower2 kernel: md: recovery thread: PQ incorrect, sector=81804824 Oct 25 18:38:53 Tower2 kernel: md: recovery thread: PQ incorrect, sector=81804840 . . Oct 25 18:38:53 Tower2 kernel: md: recovery thread: PQ incorrect, sector=81808136 Oct 25 18:38:53 Tower2 kernel: md: recovery thread: PQ incorrect, sector=81808152 That wording from the parity check is confusing. If I put a check in there to correct errors, would it help anything?