embiem Posted October 28, 2022 Share Posted October 28, 2022 (edited) Hi, I've been following the "Remove Drives Then Rebuild Parity" method from this guide: https://wiki.unraid.net/Shrink_array to remove 2 SSDs from the array and only keep the same HDD parity and one HDD disk. Before I started I set all shares to cache:yes and ran the mover (as suggested here), then I moved all files from the SSDs to the HDD that I wanted to keep in the array. All worked fine, I re-started the array and it started the Parity-Sync. I can confirm that all files are also perfectly fine on the remaining disk. The issue: no user shares are showing anymore (appdata, system etc and my custom shares) and only one disk share is showing where all the files are located now. I have not re-started the docker service yet. Should I just re-create the user shares manually? Will they pick-up the existing folders on my drive? I want the system shares to be cache:prefer again, but ofc they should pick-up the existing data. I'm unsure how to best proceed to get to that result. Thank you for any tips! Edited October 28, 2022 by embiem Quote Link to comment
JorgeB Posted October 28, 2022 Share Posted October 28, 2022 Please post the diagnostics. Quote Link to comment
embiem Posted October 28, 2022 Author Share Posted October 28, 2022 46 minutes ago, JorgeB said: Please post the diagnostics. Just tried via the UI and after ~15 minutes the browser crashed "out of memory". Will retry now using the terminal and `diagnostics` command. BTW, I see that "Log" on the Main page is showing 100%, so I assume my storage for logs is full. Quote Link to comment
JorgeB Posted October 28, 2022 Share Posted October 28, 2022 If the diags also don't work in the console get at least the syslog: cp /var/log/syslog /boot/syslog.txt Quote Link to comment
embiem Posted October 28, 2022 Author Share Posted October 28, 2022 @JorgeB thank you for the tip! Attached is the ZIP tower1070-diagnostics-20221028-1119.zip I only edited /logs/syslog.1.txt, as it was showing sensitive directory paths and was essentially just 10k lines of the same error, just for different files. Quote Link to comment
JorgeB Posted October 28, 2022 Share Posted October 28, 2022 Nothing obvious but the last syslog is empty, try rebooting and if there are still issues post new diags after array start. Quote Link to comment
embiem Posted October 28, 2022 Author Share Posted October 28, 2022 tower1070-diagnostics-20221028-2026.zip Sadly the shares are still gone after reboot & array start. Attached are the diagnostics after array start. Thanks again for checking. Quote Link to comment
JorgeB Posted October 28, 2022 Share Posted October 28, 2022 Post output of: ls -la /mnt/user Quote Link to comment
embiem Posted October 28, 2022 Author Share Posted October 28, 2022 (edited) total 16 drwxrwxrwx 1 nobody users 0 Oct 28 09:30 ./ drwxr-xr-x 7 root root 140 Oct 28 20:23 ../ It's still a relatively new server. I can backup all files temporarily, reset the Unraid server and copy all files back if that's the safest approach to fix this? Edited October 28, 2022 by embiem Quote Link to comment
Solution JorgeB Posted October 29, 2022 Solution Share Posted October 29, 2022 Those are OK, think I've found the problem, click on Settings -> Global Share settings and remove disk2 from the included disks. Quote Link to comment
embiem Posted October 29, 2022 Author Share Posted October 29, 2022 ohhhh you're my hero!! Thank you!! I totally forgot I had this enabled from the "Shrink Array" guide. It was actually showing as "All" in the UI, but I selected the new Disk1, applied it, then deselected it again to include All, applied it. Started Array and now the shares are showing up! Seems like the UI has a bug here, or at least shows "All" even though an old & now unassigned Disk is still selected. Quote Link to comment
JorgeB Posted October 29, 2022 Share Posted October 29, 2022 33 minutes ago, embiem said: Started Array and now the shares are showing up! Great 33 minutes ago, embiem said: Seems like the UI has a bug here, or at least shows "All" even though an old & now unassigned Disk is still selected. Yes, it does sound like a bug. Quote Link to comment
JorgeB Posted October 29, 2022 Share Posted October 29, 2022 I can confirm it's a bug, but it's kind of a corner case, and this is #1 for the procedure you've followed: "Make sure that the drive or drives you are removing have been removed from any inclusions or exclusions for all shares, including in the global share settings. Shares should be changed from the default of "All" to "Include". This include list should contain only the drives that will be retained." So as long as the instructions are followed there won't be a problem. Quote Link to comment
embiem Posted October 30, 2022 Author Share Posted October 30, 2022 Yes that's true. My specific case was: set Include Disk 2 only on all shares & global share settings during Shrink process I unassigned Disk 1 & Disk 3 also set Disk 2 as Disk 1 have I kept the retained Disk as Disk 2, the problem wouldn't have appeared Quote Link to comment
phanb Posted November 14, 2022 Share Posted November 14, 2022 (edited) weird share issue just popped up tonight all my shares were fine but somehow all my folders in my media shares are now user shares? No reboot since I upgraded to new OS and that was 3 days ago... and now all my NFS shares are stale on my clients Edited November 14, 2022 by phanb more info Quote Link to comment
phanb Posted November 14, 2022 Share Posted November 14, 2022 rebooted after all users were offline and still same thing Quote Link to comment
itimpi Posted November 14, 2022 Share Posted November 14, 2022 All top level folders on any drive are User Shares so this implies you accidentally copied/moved part of your Media share to the top level of a drive. Quote Link to comment
phanb Posted November 14, 2022 Share Posted November 14, 2022 (edited) I haven't done anything like that... here is a screenshot of my shares from a different system Krusader shows the exact same thing as the screenshot below and I have that mounted as /mnt/user/ so I can't explain this but it's not the scenerio of accidentally copied/moved to top level, but odd that Conspiracy Theory shows up there sorry attached wrong screenshot... again sorry if I am understanding your implication incorrectly, but this cannot be something initiated by a user as I am the only person that has access to the backend & frontend... all other users are simply plex from a different server/ interface. Edited November 14, 2022 by phanb wrong screenshot Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.