onyx00

I have a new 847 waiting to be populated so interested to hear what others are thinking. My initial reaction is to go with Xeon for the PCI lanes as I suspect that will benefit with 30+ drives, yes? No plex needs now as I have another system for that. Really looking for the best bang for the buck on Xeon platform.

I feel stupid for asking this but will the rack below work for a homelab setup? I have a cse847, a UDM pro, and a Dell PowerConnect 7048P 48-Port Gigabit switch. I don’t see why not, but beyond more space I am not sure if there a reason to get something else. Thanks in advance! VEVOR 12U Open Frame Server Rack, 23''-40'' Adjustable Depth

Woke up to an error via email and looking at the unraid dashboard it showed disk4 (sdj) was disabled and showing "1" in the r/w error column. I've attached the diagnostics file, and below is what I saw in the syslog. A short SMART test passed without issue and the extended test is currently running. Also, will point out that a parity check found 0 errors on 5/1. Based on provided details is there any suggestion of how I proceed? Especially if the extended smart test passes without issue. Do I look at cables...? Or just ignore and rebuild the disk? May 10 01:01:44 Tower emhttpd: read SMART /dev/sdd May 10 01:03:12 Tower kernel: sd 1:0:6:0: attempting task abort!scmd(0x000000008ab633e3), outstanding for 15295 ms & timeout 15000 ms May 10 01:03:12 Tower kernel: sd 1:0:6:0: [sdj] tag#3223 CDB: opcode=0x85 85 06 20 00 00 00 00 00 00 00 00 00 00 40 e5 00 May 10 01:03:12 Tower kernel: scsi target1:0:6: handle(0x000f), sas_address(0x4433221105000000), phy(5) May 10 01:03:12 Tower kernel: scsi target1:0:6: enclosure logical id(0x500605b004cd7430), slot(6) May 10 01:03:16 Tower kernel: sd 1:0:6:0: task abort: SUCCESS scmd(0x000000008ab633e3) May 10 01:03:16 Tower kernel: sd 1:0:6:0: [sdj] tag#3276 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=0s May 10 01:03:16 Tower kernel: sd 1:0:6:0: [sdj] tag#3276 Sense Key : 0x2 [current] May 10 01:03:16 Tower kernel: sd 1:0:6:0: [sdj] tag#3276 ASC=0x4 ASCQ=0x0 May 10 01:03:16 Tower kernel: sd 1:0:6:0: [sdj] tag#3276 CDB: opcode=0x8a 8a 08 00 00 00 00 ae ac 00 d0 00 00 00 08 00 00 May 10 01:03:16 Tower kernel: blk_update_request: I/O error, dev sdj, sector 2930507984 op 0x1:(WRITE) flags 0x20800 phys_seg 1 prio class 0 May 10 01:03:16 Tower kernel: md: disk4 write error, sector=2930507920 May 10 01:03:16 Tower kernel: sd 1:0:6:0: [sdj] tag#5 UNKNOWN(0x2003) Result: hostbyte=0x0b driverbyte=DRIVER_OK cmd_age=0s May 10 01:03:16 Tower kernel: sd 1:0:6:0: [sdj] tag#5 CDB: opcode=0x85 85 06 20 00 00 00 00 00 00 00 00 00 00 40 98 00 May 10 01:03:16 Tower emhttpd: read SMART /dev/sdj May 10 01:03:26 Tower crond[1174]: exit status 1 from user root /usr/local/sbin/mover &> /dev/null May 10 01:03:28 Tower emhttpd: read SMART /dev/sdh May 10 01:03:28 Tower emhttpd: read SMART /dev/sdg May 10 01:03:28 Tower emhttpd: read SMART /dev/sde May 10 01:03:28 Tower emhttpd: read SMART /dev/sdf tower-diagnostics-20230510-1630.zip

Not sure entirely how to tell what drive is at fault here. Can anyone provide some quick guidance? Diagnostics attached. tower-diagnostics-20230320-2313.zip

To the top.

Having an issue where I changed my IP address schema, and docker seems to hold on to the old IP address for port mappings on containers that were installed at the time. For example, if I install new or clean install existing docker (e.g., delete appdata folder) the old IP address is gone and everything is okay. Otherwise, docker seems to retain the old scheme in the port mappings. I've looked around and been unable to find a fix for this, or understanding if it is a bug. I've attached a screenshot for an example showing what I mean, but unclear how to even determine the issue. First and last items are *wrong* as they existed before I changed the IP scheme Middle one is *right* but I had to uninstall the docker, delete the directory in appdata, and re-install to get to this point. Any thoughts or advice? Logs I need to pull?

Regarding the instructions to use v3 of Sonarr, I believe I followed Binhex's instructions on Github but when I start the Sonarr container I receive an error (full log here): Am I missing something or is there something else wrong in Unraid/Docker that would explain why I am seeing this?

Yeah good point...no sense rolling the dice. Thx!

During a recent parity check I received a notification about a SMART error on one of the disks (below). I observed no other issues (read/write errors, etc.) and the parity check completed successfully without any errors or corrections. Despite being ~8 years old the drive itself is fine according to SMART, with no reallocated sectors or pending, etc., and a subsequent parity did not trigger this error again (although I did move all data off the disk just in case so not sure if that is why...). I searched the forums and seems this error is less of a concern unless I am observing other issues. Any suggestions on what I do? Would it be wise to swap the drive and then do some pre-clears on it to confirm there are no issues? Thanks in advance! Here is the error I received for reference: Event: Unraid Disk 1 SMART health [1] Subject: Warning [TOWER] - raw read error rate (failing now) is 4294933651 Description: Hitachi_HDS723020BLA642_MN1220F327PH3D (sdc) Importance: warning

No help on this? I've tried blowing away my entire docker/appdata, confirmed docker puts everything on the cache drive, and followed Space Invader's youtube videos exactly...same issues. Any other ideas short of starting from scratch on my unraid configuration?

Hi all, seeing "database is locked" errors in SQLite for Sonarr on downloads. Sometimes Sonarr imports fine, sometimes I get the error and while the file is moved, Sonarr never recognizes this because of the database locked error. While attempting to resolve this I have confirmed/tried a few things: Check all my paths between SAB and Sonarr (both have exactly the same path set in docker, i.e., /data -> /mnt/cache/Downloads/) Ensured all docker containers /config maps to /mnt/cache/appdata Looked at permissions for my /mnt/cache/Downloads files and folders, which were all set to 777 Log file attached as an example. Any ideas of what else I can try short of just blowing Sonarr away and starting over? 20200524 Sonarr log with issues - truncated.txt

Hi, running on an old Sempron CPU and in need of an upgrade. I've looked through threads but trying to see if there are any general recommendations for hardware? I don't have a specific budget but have basic needs - a few dockers, only ~15TB of storage across five drives. VMs with hardware pass-through would be great if it doesn't create complexity. Are there any guides / recommendations out there for hardware? I would love to have something I can just go pull the trigger on and be done.

(nevermind - was able to figure it out - please delete)

Have searched, read, etc. which has helped me narrow down choices. Wanted to bounce my thinking off folks and hear some guidance. My goal is to consolidate HW to a single chassis (currently have unraid w/ Sempron 145 and separate Phenom 2 X4 955 w/ Win10 + Win10 VM) that supports my needs. Overall I think my general needs are: + 1-2 Win10 VMs for general browsing and other tasks but no gaming (but would like HW pass-through to future proof) + Support 1-2 other VMs at times (Linux, etc.) but low usage + Some basic dockers (e.g., Deluge+VPN) I was thinking a Xeon E3-1275 v6 would be good with, but worried I wouldn't have much breathing room - or am I overly worried given my needs? If not, then I was considering i7-5820K but it wouldn't allow ECC (...I've survived last 6 years on Unraid without it though), or on the higher end the E5-1650 V4. And then depending on CPU, what motherboard? Thanks in advance for help/suggestions!

Hi, having repeated problems creating a directory in a specific directory on a user share, and while trying to solve this I also noticed that even though my user shares are all set to public, I need to login in to access them from my Windows machine. I've attached a diagnostic file and a screenshot of the directory I'm having issues with. Questions I have are: 1a.) Why did new perms script not fix this directory? 1b.) Is there some command I can run on all my drives to fix issues such as this? 2.) Is it intended that even set to public I need to log in to access my shares from a windows machine? Thanks in advance for any help! tower-diagnostics-20161020-1812.zip

Sorry all, got this resolved. I ended up cross-posting (I know, but I was desperate for help) in the 5.0 rc forum, and listed all the things I tried. http://lime-technology.com/forum/index.php?topic=27145.msg238140#msg238140 The issue seemed to just resolve itself over night. Not sure how or why (no reboots of any PCs, or any other special steps). I was going to keep an eye on it just to see if the issue crops up again.

Thanks for the reply - so after all of that, now randomly it started working (no reboot of either system). I'll just keep an eye on it I guess. Maybe write a quick script to run on my Windows 7 machine to testing creating, modifying, and deleting a file from the shares and run that a few times a day just to see if the issue crops up again. So odd...

Upgraded to Unraid 5.0 rc12b from 4.7 and followed the migration guide. All went fine per the guide and I ran the new permissions utility. Today, I noticed I could not delete files. From my Windows 7 machine I try to delete and get: "You need permission to perform this action." Next line reads: "You require permission from TOWER\nobody to make changes to this file" Things I know or have tried: - Have run New Permissions script 10+ times - My user shares are set to Public - I checked that I had no stored credentials in my Windows 7 machine. - I also rebooted my Windows 7 machine as well as Unraid, but this did not help at all. - Tried resetting share.cfg and disk.cfg taking the 5.0 rc12b version) - Confirmed permissions on my user shares as shown in the attached picture. Does someone have any ideas? This occurs on all user shares.

As I mentioned in my original post, there are no saved credentials...please see above for all the things I have already checked/done. This whole unRaid 5.0 thing is turning in to a nightmare. This should be a simple thing - a share set as public shouldn't behave like it has some kind of permissions on it. Or am I mis-understanding things?

Upgraded to Unraid 5.0 rc12b from 4.7 and followed the migration guide. All went fine per the guide and I ran the new permissions utility. Today, I noticed I could not delete files. From my Windows 7 machine I try to delete and get: "You need permission to perform this action." Next line reads: "You require permission from TOWER\nobody to make changes to this file" My user shares are set to Public, and I checked that I had no stored credentials in my Windows 7 machine. I also rebooted my Windows 7 machine as well as Unraid, but this did not help at all. Does someone have any ideas? This occurs on all user shares.

Form the above I got the impression that continuously writing to the boot flash drive was not a good idea as you will wear out the device. Not that flash drives are expensive, but you don't really want to purposely degrade the flash device you are booting with. I was going to try doing this with a second USB flash drive by mounting it and redirecting syslog as you and others described above. I was also going to try getting my Tomato-based router to write its logs to that same flash drive/share. I'll check back in once I finish trying to set this up.

Sorry if this is the wrong place to post (please let me know if there is a better place). I'm having issues installing the vim package in unMenu 1.5. I installed "C" compiler & development tools and set it to re-install on reboot. I also then rebooted. I installed the vim package following this, and I get a message on package install saying "Installed, but version is different. Current version=' ' expected '7.2' ." I try to run vim from the command line but get the following error: "vim: /lib/libc.so.6: version `GLIBC_2.11' not found (required by vim)" Any ideas? I tried searching the web and the forums, but didn't find anything substantial. Would appreciate anyone who could help or point me in the right direction. Thanks in advance!

Good deal, thanks all - I may try the booting in to unRAID on a spare PC and then running the preclear that way.

Cool revco, no problem - I want to do it right, so I'll just bite the bullet and wait. Thanks for the fast reply!