barrysagittarius

Well, the short response would be ignorance. I tried to look this up and found the following. The 650W PSU I have has the following specs +3.3V@20A, +5V@20A, +12V@54A, [email protected], [email protected] From this post I'm under the impressions that the +12V@54A is the power I have for all of the hard drives which seems like it should be enough. Should my next step be a different PSU cable setup? the documentation says I have SATA/PERIF: 1 x SATA 5-Pin x 4, 1 x SATA 5-Pin x 2, 1 x Molex 4-Pin x 4 Well, a few updates. I checked and I had 2 PSU cables. So the answer to your question is "pure ignorance" on my part. Since then I was able to run an extended SMART test without problems. However, after replacing the power cable for the previously faulty drive, I powered on the server and now 2 different hard drives were missing. I powered it off, replaced the other PSU cable, looked to make sure nothing was loose or out of place and now having powered it on and off, all hard drives are missing. So my new questions are 1) Would this be the PSU if 2 brand changing the cables didn't improve anything/no pattern for missing hard drive to new cable. If the whole PSU is failing shouldn't the computer not load up? 2) I have SATA data cables for 2 drives to the motherboard, and for to an LSI. There is no pattern here as well. So why would all the drives show up as missing now with no other changes? Here's one more diagnostics file from today if it helps. Thanks! towerbackup-diagnostics-20260125-2130.zip

I recently received an error that disk1 has been disabled, and has read errors. Looking at the SMART logs there are no read errors or reallocated sectors, but "199 UDMA_CRC_Error_Count" was at 19. After running a SMART test and not seeing anything (I think) that shows the drive is failing and looking around the forum my current understanding is this is a data or power cable issue. I used a different data cable, and tried to unassign, start the array, stop, then assign to rebuild the drive. When the array finally starts the drive is still disabled and nothing is being rebuilt and I believe the same error is popping up, but new since the CRC error count has increased. My questions are 1) Am I missing something or is the above assessment correct? 2) Since I changed the data cable, and the same problem persists, does this mean I need to look into it being a power issue? As a side note, I have a splitter cable power all the hard drives. If it's a power issue with one drive would it not affect the other drives? I'm currently not sure how to proceed, and am not an expert at reading the diagnostics and haven't learned much more about the UDMA CRC error count than what I've already mentioned. Any help would be greatly appreciated! Thanks! towerbackup-diagnostics-20260122-1614.zip towerbackup-diagnostics-20260123-2057.zip

Thanks for your response! I ended up removing the drive and trying to run a SMART extended test but couldn't get that to work. I ended up reformatting and adding back to the cache pool and was able to run a SMART extended test there. It came back with no errors. Not sure why the drive would have dropped, but at least according to the SMART test the drive is fine. As a side note, the drive runs consistently "warmer" at ~40-50 celsius while idle if that means anything. I've attached the diagnostics. If you could, let me know if I should be concerned about anything or if I should just wait to see if I get more errors in the future. Thanks! tower-diagnostics-20251002-1753.zip

I got the above error for fix common problems and was prompted to post my diagnostics here. I installed a new ssd to create a cache pool several weeks ago and while the ssd always ran hot even when idle (40-50 C) it didn't seem to be a concern. Checking the logs it looks like there are BTRFs errors and I'm not sure if it's a failing drive, or something I might have done setting up the cache pool but this is the first I'm getting this error and it looks like it's been the past few days. I'm not sure if I've diagnosed this drive correctly as the problem or next steps with using Scrub since this thread seems to have a lot of cautions (https://www.reddit.com/r/unRAID/comments/qbwoa8/how_do_i_get_rid_of_the_btrfs_error_device_sde1/). I think I want to know 1) Is the drive dying/bad 2) Steps to not have my cache data corrupted/revert back to the single SSD if the 2nd is failing. I don't want to do anything too stupid and hope someone who knows more than I do can point me in the right direction. Thanks! tower-diagnostics-20250928-0925.zip

Previously I had no problem spinning down SATA disks. Created new build (different mother board, CPU, PSU, RAM) but everything else, including configuration remained the same. Now all discs will not spin down, or spin down and immediately spin back up. Not sure what could have changed besides maybe something about the motherboard? Tried all the usual suggestions from other posts, nothing seems to work. tower-diagnostics-20240708-1940.zip

I probably worded it poorly but I assume it's restarting after a crash. How would I know the difference? I'm not seeing any errors before it does an unclean restart. If it is hardware is there anything in the diagnostics I can look at to determine? I see nothing in the syslog.

Problem: No problems with Unraid or hardware for several years. Server was turned off around July, then turned back on in September. Server began to randomly restart resulting in needing to run a parity check all the time. Eventually began to pinpoint it to using Jellyfin and server crashing while watching higher quality files that might have been more intensive for transcoding. Turned off transcoding and it felt like the crashes stopped. Starting a few weeks ago the server would restart in the middle of the night or randomly in the day when not using Jellyfin. So while it definitely crashed while using Jellyfin, that's either not the main cause, or there's now additional causes. Things I've tried changing transcoding settings. Updating Unraid/hoping an update might fix the problem if software Changing macvlan to ipvlan Syslog to server. I don't see anything showing an error before it restarts. For example (see below), It looks like it updated plugins at midnight. A little later drives spun down, a trim process happened an hour or two later, things spun down again, and then around 6:00 the server restarted (I'm assuming that's what this indicates) Jan 5 20:17:39 Tower monitor: Stop running nchan processes Jan 5 22:00:07 Tower emhttpd: read SMART /dev/sdh Jan 6 00:00:01 Tower Plugin Auto Update: Checking for available plugin updates Jan 6 00:00:09 Tower Plugin Auto Update: NerdTools.plg version 2024.01.06 does not meet age requirements to update - 0 days old Jan 6 00:00:09 Tower Plugin Auto Update: Checking for language updates Jan 6 00:00:09 Tower Plugin Auto Update: Community Applications Plugin Auto Update finished Jan 6 00:05:41 Tower emhttpd: spinning down /dev/sdc Jan 6 00:16:01 Tower emhttpd: spinning down /dev/sde Jan 6 00:17:00 Tower emhttpd: spinning down /dev/sdd Jan 6 00:17:02 Tower emhttpd: spinning down /dev/sdg Jan 6 01:02:28 Tower root: /var/lib/docker: 11.3 GiB (12100575232 bytes) trimmed on /dev/loop2 Jan 6 01:02:28 Tower root: /mnt/cache: 38 GiB (40803491840 bytes) trimmed on /dev/sdf1 Jan 6 02:02:22 Tower emhttpd: spinning down /dev/sdh Jan 6 06:03:32 Tower file.activity: Starting File Activity Jan 6 06:03:32 Tower emhttpd: Starting File Activity... Jan 6 06:03:32 Tower file.activity: File Activity inotify starting Jan 6 06:03:32 Tower inotifywait[14355]: Setting up watches. Beware: since -r was given, this may take a while! I'm currently not sure if this is a software problem (it seemed like there were problems with Unraid OS update over the summer/fall for some people) or a hardware problem (I've run a ram test. not sure why there's not an error before shutdown). No idea what else to try or how why I'm not finding errors, but I don't know how to read the diagnostic logs to understand this further. Any help would be greatly appreciated! tower-diagnostics-20240106-0643.zip

@purplewalrus I solved my problem, which I think was the same as your problem. When installing the docker, I set Network Type: Bridge Then I added the following by Path/Port/Variable Port -Config Type: Port -Name: WebUI -Container Port: 8080 -Default Value: 8080 -Connection Type: TCP Variable -Config Type: Variable -Name: OPENPROJECT_HTTPS -Key: OPENPROJECT_HTTPS -Value: False -Default: False I think the key for not getting the SSL error was the OPENPROJECT_HTTPS variable. To get it to work otherwise, with https, I think requires setting up Swag/nginx which I was able to do for Nextcloud but is beyond my paygrade for this (didn't have time to figure out). More info is here https://www.openproject.org/docs/installation-and-operations/installation/docker/ Hope that solves your problem.

I'm unable to troubleshoot what is causing slow transfer speeds to my server. Previously I was able to transfer from my computer to server at more than 100mbps. Now it's around 10. The only changes I made were installing an LSI 9207-8i SAS card, Switching a new hard drive in for parity, and using the old parity drive as a new drive in the array (this is all functioning fine). I've checked the following: 1) When copying to the server, the Cache drive is being used, and is an ssd card (which was previously the case) 2) Using both open speed test, and speed test dockers, the DL network speed for the server is ~100mbps. Using ethtool eth0 I get "Speed: 1000Mb/s" 3) Using Diskspeed Docker, all HDs, including SSD are operating as expected, and one isn't a bottleneck. Am I missing something? It seems like even with installing the SAS and new HDs, with nothing changing to the SSD Cache (it's not connecting through the SAS card) that transfer speed to the server shouldn't have changed.