• Posts

  • Joined

  • Days Won


Everything posted by Vr2Io

  1. If no spare cable then no solution. As WOL were base on magic-packet ( MAC address ), so nothing need to block at router. I use optical and use suggested method for WOL. I also notice Tplinlk say support WOL and most other 10G NIC not mention this. For WOL on those power hungry NIC, unless they have advance power management design, otherwise that would draw too much power in S5 and I think that's why 10G NIC usually haven't WOL. ** edit : If you have Wireless NIC support WoWLAN ( wake on WiFi ) then you could try setup under VM **
  2. Simple connect the onboard NIC ( without assign IP ) for WOL purpose.
  3. No, you must fix it or try use other USB SATA bridge, otherwise USB storage file or file system will corrupt.
  4. Yes, dynamic allocate to different disk according different setting. Yes, if a file with same name in other disks, it will writing too, as result duplicate file in different disk.
  5. It may save a bit of power, but reliable & stable more importance. Pls do a simple test, "rsync /mnt/disk1/<share>/<folder> /mnt/disk1/test/<share>/<folder> " to check does it crash, then delete /mnt/disk1/test
  6. There are no link aggregation in SATA, I think you talking about aggresive sleep feature, btw you should set to disable and I don't think it will be the cause.
  7. Your setup quite simple and seems no abnormal found. First, are you sure only rsync will got crash, how about use command cp or mc ? BTW, I guess problem cause by transfer to USB device and USB brisge/device have problem, also, what filesystem of the USB storage ? FYR, I never got crash with rsync, even with USB storage.
  8. I use Intel X520 and Connect X3 ( need flash from IB to Eth firmware ), both work well with Unraid. https://forums.unraid.net/topic/84705-mellanox-connectx-2-mhrh2a-xsr-10gb/ I also have Emulex old gen OCE-10xxx adapter, it work with Unraid ( be2net driver ) by flash it to another firmware something like 4.x.x ( can't remember ). 49Y7952 should be newer gen OCE-11xx adapter. You may search / try any firmware would make it work. https://forums.servethehome.com/index.php?threads/beware-emulex-10gbe-virtual-fabric-adapter-ii-x8-pci-e-slot.3491/page-3
  9. Enjoy !! BTW only two cable for 5 cage for total 20 bay still not quite enough.
  10. I haven't found attributes 200 on WD disk, just finish re-arrange disks and now all array disk were WD. 😈 I have try monitor attributes 1, but sometimes it will change to some no. and return to zero in short time ( can't remember WD or not ), so I finally haven't monitor it.
  11. Note, I just re-install UD and reboot, I have 10 data disk , 16 UD disk and 2 parity. Below is the screen capture when array start, Disk 1-10 counter figure show at Dev 1-10, i.e. Dev 1=Disk 1, Dev 2=Disk2 ..... and Dev 11-16 will be zero. For about spindown issue, I also reproduce by - Start array - Spindown all disk - Clear all counter by "clear stats" - Let say made activity on Disk 1,5 then Dev 1,5 also spinup - Same counter figure show on UD @bonienl @limetech @dlandon So, if I perform parity check/sync, I have 10 data disk, then UD dev 1-10 will spinup and waste 100W+ uncessary.
  12. I got same issue when switch to R/W counter mode. All figure is copy from array R/W figure and places in UD different disk. Or more clear state the symptom, array Disk 3,4 figure copy to Dev 3,4 ........ ( due to delay in capture and update, they will show different figure, but actually they are same number ) Array: UD: And this also prevent UD Dev 3,4 to spindown.
  13. I am not sure the outlook of zpools in UD, I have two RAID0 pool in UD, I will mask member disk by set to passthrough, then got a clear look.
  14. There are some interest founding about UD disk spindown issue after I rearrange most disk. My setup was 10 data + 2 parity and 16 UD. When 10 array data disk have activity, then UD dev 1-10 won't spindown, but dev 11-16 will spindown. After only array disk 3,4 have activity, then UD also only dev 3,4 in spinup. I notice UD state spin down not control by UD after 6.9+, so does it is Unraid OS issue ? FYR, parity sync in progress and no other disk activity, OS 6.10.0-RC1.
  15. If you build those power cable, pls use spare disk to test before put in production. Sometimes we will make mistake and burn all stuff. I have that and burn 4 of Blray drive, lucky they all cheap and I have fix it.
  16. It likely Seagate disk with LSI HBA, wakeup / power-save issue, most read error occur during disk wakeup.
  17. Could you try two test 1. If only insert 1 NVMe ( you have two ), try swap M2 slot. 2. If have another brand M2, pls try mix different model.
  18. Yes, those are capacitor, use to minimise ripper voltage, it also share the loading of PSU inside capacitor, it should little bit help to extend the PSU life.
  19. I will build something use DIY component. One side coming from PSU and the other side connect to harddisk. 4 terminal are 12v G G 5V. Each "Insulated Terminal Barrier Strip" ( red black yellow ) will insert 3 or 4 wire and I will soldering to ensure reliable connection. I will build different harddisk wire set ( black wire with correct diameter ) for different need, just detach / attach at terminal block. Each wire will connect max 4 disk, and inset to DIY SATA plug at the end. Anyway, max 16 disks, never use Y power splitter. Or something like that, if connect point have two "Insulated Terminal Barrier Strip", then it can increase the no. of connect device or to change the direction.
  20. May be try eliminate the Y power splitter and try again to verify problem reproduce or not first.
  21. Pls source different supplier.