CS01-HS

Members
  • Content Count

    240
  • Joined

  • Last visited

Everything posted by CS01-HS

  1. Force a dirty shutdown and consequent parity check on startup Let parity check complete Stop array If you start the array you're forced to run a parity check because the completed run's not recognized. Reboot gets around it.
  2. Ah I see, that's still good power savings. Yes, I have the J5005-mITX with 6 HDDs (4 on the card, 2 on the board) and 2 Cache SDDs. More information is linked in my signature. If you want to see great power savings look at this: https://translate.google.com/translate?sl=auto&tl=en&u=https%3A%2F%2Fwww.computerbase.de%2Fforum%2Fthreads%2Fselbstbau-nas-im-10-zoll-rack.1932675%2F He also posts here:
  3. I'm running a j5005 with an H1110 HBA (cut to fit the x1 slot) and 8 disks: 2 SSD Cache (on the integrated intel) 3 Array disks and and a BTRFS Raid 5 Pool of 3 disks (all distributed between the integrated ASMedia and the HBA) It's not a performance server but I don't notice any slowness although your idle wattage with the new setup (22W) is close to mine. Do you remember what your idle wattage was with the J4105 and HBA? Benchmarking the x1 controller in DiskSpeed does show limitations but not enough to affect usability (although if my controller ran 8 disks i
  4. I saw a few of these hard resetting link errors during my mover run. Thankfully (?) no CRC errors reported. ata3 is a spinning disk attached to an integrated ASM1062 controller. I wonder if it might be related to the power-saving tweaks because nothing else changed. For now I've disabled them and will see if they reappear. Maybe coincidence but I'm posting in case others have the same issue. Nov 3 23:17:30 NAS move: move: file /mnt/cache/Download/movie_1.mp4 Nov 3 23:17:33 NAS kernel: ata3.00: exception Emask 0x10 SAct 0x80 SErr 0x4050002 action 0x6 frozen Nov 3 23:17:33 NAS
  5. I don't search often so I can't say for certain it's the beta but it used to work consistently and now it doesn't, even in safe mode. No results returned no matter how long I wait. I'm running the latest MacOS Catalina. I also tested in MacOS Mojave (unraid VM), same result. I have a Raspberry Pi shared over SMB where search from the same two clients works fine. Diagnostics from safe mode attached. nas-diagnostics-20201027-1319.zip
  6. I wouldn't take any meaningful risk to save 4W. Would you distinguish between any of the power-saving tweaks (SATA links, I2C, USB, PCI and increasing dirty_writeback) in terms of risk, assuming a UPS/no unexpected power loss?
  7. Your go files additions saved me also about 4W. And it might be coincidence but my SSD temps dropped about 4°C post-tuning. I've never seen them so cool. There has to be some cost to this, no? I haven't noticed a difference but nothing's free.
  8. Sure maybe if/when it gets cleaned up. Just to be clear this isn't related to spin-up groups (which I don't use either), just standard drives. I wanted a way to easily track whether my drives were sleeping/waking too frequently. Someone cleverer might be able to integrate total wakes over the specified time range.
  9. I have a visually pleasing (but technically dirty) solution to my quest for a spin-up graph. I'm new to grafana and find it frustrating so if anyone has improvements feel free to post them. This is the end result: Setup: 1. Start with a User Script script to track drive activity and temperature in influx set to run every 5 minutes (borrowed from this php version.) Replace every XX with your system's settings (default influx port is 8086) #!/bin/bash # User settings INFLUX_IP="XX" INFLUX_PORT="XX" HOSTNAME="XX" # Drive IDs (in /dev/disk/by-id
  10. Updated the script to work with with 6.9.0-beta30 I don't know if it's the new kernel or unraid itself but now the drive's diskstats seem to increment even without access (maybe SMART polling?) Keying off the partition's diskstats seems to solve it. Note, this looks for reads/writes on the first partition - if you have a multi-partition drive it won't work properly (and may sleep you drive while you're accessing it,)
  11. Follow up: As of (at least) 6.9.0-beta30 adding intel_iommu=igfx_off to syslinux.cfg is no longer necessary with my j5005.
  12. That's great, thanks for finding it. My setup's a lot cleaner now. I think that was my only sdX dependency so problem solved.
  13. Thanks. I don't see it in the dashboard (which instead shows a configurable "cache devices" setting) but I'll dig into the thread. Clever trick installing smartmontools in the container, I had been using HDDTemp docker.
  14. What's everyone doing to ensure persistent drive labels for reporting (Grafana), etc. ? My sdg and sdh swapped labels with sdi and sdj. I thought I could take advantage of hotswap to correct it - DON'T DO THIS (I'm rebuilding parity now because of it.) So is there a fix or do we just accept some level of randomness?
  15. SSH keys suddenly stopped working. The cause was root ./ and root parent(?) ../ were owned by nobody:users with group/other write permissions. I didn't notice any other directories affected. Not sure if the beta caused it but it's never happened before. A reboot fixed it, back to normal: root@NAS:~# ls -la / total 8 drwxr-xr-x 20 root root 440 Oct 1 23:02 ./ drwxr-xr-x 20 root root 440 Oct 1 23:02 ../ EDIT: Just to add, the new multi-pool support which, with restrictions, can approximate multi-array is an amazing feature. I'm rebuilding parity while runnin
  16. Not critical but VNC Remote for VMs fails to load with the following error in Safari 14 (latest version) for MacOS. Works properly in Brave.
  17. No need to change the shares' caching setting or run mover? Wow, that's easier than I expected, thanks.
  18. II have a RAID 1 btrfs cache (with equally sized disks.) I figure I can convert it to a single-drive pool then use the newly-freed drive as my "btrfs destination." But is there a way do that without adding it to Array Devices (which will introduce new problems), maybe through unassigned devices? Otherwise I'm not sure how most plan to do this. NoCOW seems important so I don't want to lose it.
  19. I searched the thread and didn't find an answer so apologies if this question's been asked. I'm new to this plugin and have my first verification running after the initial check. It's been going for around 8 hours and I'm wondering how far it's progressed. Is there an indicator somewhere? The initial check showed a handy progress bar in the page under Tools. I'm also confused about whether "check" and "verify" are used interchangeably in the documentation. The help says Use the Check command to verify files against a previously exported file but the plugin can be config
  20. Fancier options are plug-and-play but I have automatic restart working on my low-end Cyberpower. It requires the NUT plugin and some custom configuration. I wrote a how-to for my networked setup but the NUT plugin provides all the necessary files on unRAID so adapting it for a direct (USB) setup shouldn't be difficult.
  21. Right, the BIOS has to be set to start automatically when powered but that won't be triggered unless the UPS (a) cuts power after the computer shuts down and (b) restores power when the outage ends. For example, if he'd set Turn off UPS to NO and the outage ended after the computer shut down but before the UPS was depleted the computer wouldn't know to boot because the outlet never lost power. (At least as I understand it.) This doesn't make sense to me. The unRAID UPS setting doesn't change your BIOS.
  22. Using the built-in UPS management with Turn Off UPS set to yes? Interesting, it didn't work with my cheaper model but good to know.
  23. I had the same inaccessible webui problem. It's caused by an initialization failure either because port-forwarding fails on the server end or the container script that detects a successfully-forwarded port fails. Either way, disabling port-forwarding fixed it:
  24. My hot/cool: Sep 8 06:55:01 NAS parity.check.tuning.php: TESTING: parity temp=32 (settings are: hot=40, cool=35)) and yours Sep 8 18:20:02 Tower parity.check.tuning.php: TESTING: parity temp=26 (settings are: hot=0, cool=6)) are very different. Are you maybe setting absolute temperatures on the plugin page? They should be relative: