adammerkley

Replaced the failed Sabrent Rocket 4.0 nvme drive with a Samsung 980 Pro drive. Thanks to zfs the procedure went without a hitch.

Yup, rebooting brought the device back, then I cleared the pool error and ran a scrub. Shortly after, the device was offline again and the pool degraded. I'm going to replace the nvme drive.

Randomly looked in my syslog this evening and saw that one nvme drive in my cache pool was dropped. Now I see in the UI that my cache pool is degraded. The nvme that was dropped is the older of the 2 drives. I have posted diagnostics. What's my next step before rebooting? I have added nvme_core.default_ps_max_latency_us=0 pcie_aspm=off to my go file as advised in a similar thread I found. tower-diagnostics-20240509-2104.zip

There is a replacement now from dlandon in CA.

Upgraded from 6.12.6 to 6.12.8 without issue.

I was able to launch tmux by deleting /tmp/tmux-0 first, but any idea why this would be neccessary? rm -r /tmp/tmux-0/

Getting this when trying to run tmux like I have a thousand times before: directory /tmp/tmux-0 has unsafe permissions I ran Docker Safer New Permissions but that didn't resolve (Not sure that touches /tmp anyway.)

The info for the RTL8168(B)/RTL8111(B) PCI Drivers plugin on the App store calls out 8411 too .

Upgrade from 6.12.4 went fine. Didn't follow the instructions on stopping the array, but I encountered no issues anyway.

Did you ever get an answer to this? I see this message upon boot too. Doesn't seem to be harming anything, but just knowing it's there is annoying

I replaced all of the SAS-SATA cables for drives connected to my HBA, and the SATA cables for drives connected to the motherboard. I also removed the molex splitters I had deployed to deliver power to the backplane/drive cages. Just completed a parity check with no errors.

Ah good point. I'll check that too. Thanks!

Looking for a sanity check here. Parity check kicked off last night at 6PM. A few minutes into the parity check a bunch of errors like the below popped up. I've ordered replacement SAS->SATA cables for my HBA drives, and replacement SATA cables for my motherboard-connected drives. Anything else I should be looking for? I've attached diagnostics. Oct 6 18:10:08 Tower kernel: ata3.00: failed command: READ FPDMA QUEUED Oct 6 18:10:08 Tower kernel: ata3.00: cmd 60/f8:a8:68:53:ab/03:00:0c:00:00/40 tag 21 ncq dma 520192 in Oct 6 18:10:08 Tower kernel: res 40/00:80:60:3f:ab/00:00:0c:00:00/40 Emask 0x50 (ATA bus error) Oct 6 18:10:08 Tower kernel: ata3.00: status: { DRDY } Oct 6 18:10:08 Tower kernel: ata3: hard resetting link Oct 6 18:10:14 Tower kernel: ata3: found unknown device (class 0) Oct 6 18:10:18 Tower kernel: ata3: softreset failed (1st FIS failed) Oct 6 18:10:18 Tower kernel: ata3: hard resetting link Oct 6 18:10:24 Tower kernel: ata3: found unknown device (class 0) Oct 6 18:10:26 Tower kernel: ata3: SATA link up 6.0 Gbps (SStatus 133 SControl 300) Oct 6 18:10:26 Tower kernel: ata3.00: configured for UDMA/133 Oct 6 18:10:26 Tower kernel: ata3: EH complete tower-diagnostics-20231007-0824.zip

Upgrade went fine.

Apologies if this has been asked previously, I searched the thread and couldn't find it. I started a 3 cycle pre clear on a 16TB drive a few days ago. It's almost finished with the post-read on cycle 1. Is it possible to change the settings to just 1 cycle at this point?

I am currently running a preclear of a new 16TB drive. I just got a notification that UD has an update available. Should I update or hold off? If I have to hold off, I suppose I'll need to tell CA Auto-Update to not run until the preclear completes.

Suddenly getting "Conversion failed. A required codec could not be found or failed to install." when trying to play media. This warning appears in the plex console when playing fails: [Transcode] HTTP error requesting GET https://plex.tv/api/codecs/h264_decoder?build=linux-x86_64-standard&deviceId=cfb78a4e-ef5f-4d89-8c6c-ecb622afcfa9&oldestPreviousVersion=1%2E2%2E7%2E2987-1bef33a&version=fa235d6-3855 (6, Couldn't resolve host name) (Could not resolve host: plex.tv) I've tried deleting the codecs folder, and also chmod'ing it to 777 but still get the error. Edit: Changing VERSION from plexpass to public has resolved this behavior.

Upgraded just fine from 6.7.2. Now to look into replacing OpenVPN with WireGuard... Thanks for all your hard work.

Can someone walk me through how to get this running behind an NGINX reverse proxy? I'm using the linuxserver.io LetsEncrypt container.

I'm so curious about trying this. I have done the first few steps and rclone is working great to my Google Drive. I just can't bring myself to fully rely on Drive in case something goes wrong!

Upgraded from 6.6.5 to 6.6.6 without issue.

Another thank you. The tape trick worked perfectly on my eBay 310. Now I don't have the Marvell warning from Fix Common Problems anymore

Ah, thank you. I will plug it into one of my open drive caddies. Edit: Plugged a single drive into one of my caddies and it worked right away. Thanks so much!

Trying to get some data off of my old Unraid server. The hardware no longer boots so I have pulled the drives and plugged them into an external USB enclosure. I've plugged the drives into my current Unraid server, but I'm unable to mount any of them so far in UD. The drives are formatted in ReiserFS. I get these messages in the log when trying to mount: Oct 14 11:12:35 Tower unassigned.devices: Adding disk '/dev/sda1'... Oct 14 11:12:35 Tower unassigned.devices: Mount drive command: /sbin/mount -t reiserfs -o auto,async,noatime,nodiratime '/dev/sda1' '/mnt/disks/old1' Oct 14 11:12:35 Tower kernel: REISERFS (device sda1): found reiserfs format "3.6" with standard journal Oct 14 11:12:35 Tower kernel: REISERFS (device sda1): using ordered data mode Oct 14 11:12:35 Tower kernel: reiserfs: using flush barriers Oct 14 11:12:35 Tower kernel: REISERFS (device sda1): journal params: device sda1, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30 Oct 14 11:12:35 Tower kernel: REISERFS (device sda1): checking transaction log (sda1) Oct 14 11:12:35 Tower kernel: sd 0:0:0:0: [sda] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08 Oct 14 11:12:35 Tower kernel: sd 0:0:0:0: [sda] tag#0 Sense Key : 0x5 [current] Oct 14 11:12:35 Tower kernel: sd 0:0:0:0: [sda] tag#0 ASC=0x24 ASCQ=0x0 Oct 14 11:12:35 Tower kernel: sd 0:0:0:0: [sda] tag#0 CDB: opcode=0x2a 2a 08 00 01 00 cf 00 00 08 00 Oct 14 11:12:35 Tower kernel: print_req_error: critical target error, dev sda, sector 65743 Oct 14 11:12:35 Tower kernel: print_req_error: critical target error, dev sda, sector 65743 Oct 14 11:12:35 Tower kernel: Buffer I/O error on dev sda1, logical block 8210, lost sync page write Oct 14 11:12:35 Tower kernel: REISERFS warning (device sda1): journal-837 _update_journal_header_block: IO error during journal replay Oct 14 11:12:35 Tower kernel: REISERFS warning (device sda1): reiserfs-2006 journal_init: Replay Failure, unable to mount Oct 14 11:12:35 Tower kernel: REISERFS warning (device sda1): sh-2022 reiserfs_fill_super: unable to initialize journal space Oct 14 11:12:35 Tower unassigned.devices: Mount of '/dev/sda1' failed. Error message: mount: /mnt/disks/old1: wrong fs type, bad option, bad superblock on /dev/sda1, missing codepage or helper program, or other error. Oct 14 11:12:35 Tower unassigned.devices: Partition 'ST315003 41AS' could not be mounted...

Can't think of much benefit to doing that. I'm relying on CA Backup for this. But if you do backup using Github and you use DNS verification, you'll want to exclude the file with your DNS provider API key.