July 16, 20232 yr I am currently running Version: 6.12.2 and I am not able to update to 6.12.3. I stopped Docker containers and disabled Docker service. Stopped all VMs. Stopped the Array. There is nothing running. 30+ GB of available RAM memory and plenty of available disk space: root@Shelf:~# df -H Filesystem Size Used Avail Use% Mounted on rootfs 17G 687M 16G 5% / tmpfs 34M 291k 34M 1% /run /dev/sda1 65G 1.1G 64G 2% /boot overlay 17G 687M 16G 5% /lib overlay 17G 687M 16G 5% /usr devtmpfs 8.4M 0 8.4M 0% /dev tmpfs 17G 0 17G 0% /dev/shm tmpfs 135M 406k 134M 1% /var/log When I run the Tools > Update OS, I am getting this: plugin: updating: unRAIDServer.plg plugin: downloading: unRAIDServer-6.12.3-x86_64.zip ... done plugin: downloading: unRAIDServer-6.12.3-x86_64.md5 ... done wrong md5 plugin: run failed: '/bin/bash' returned 1 Executing hook script: post_plugin_checks Any ideas what should I try? EDIT: Should I go with manual update? shelf-diagnostics-20230716-1853.zip Edited July 16, 20232 yr by brankko Added an link for manual update
July 17, 20232 yr Community Expert Solution 14 hours ago, brankko said: wrong md5 Reboot and try again, if the same run memtest.
July 17, 20232 yr Author Weird thing. I tried rebooting yesterday. Had the same issue. Now it worked. Thanks! Will definitely check the memory sticks, as I moved them recently from the other box. Edited July 17, 20232 yr by brankko
August 11, 20232 yr Author I had no time to investigate, as I had to open the case, connect the display and everything so I kept it turned off most of the time not to corrupt any data, but today I finally did everything by the book and run the memtest. One of the recently installed memory sticks is faulty so I removed it. Since 16GB is plenty for my current usage, I'll just keep it as is until the next upgrade (already got 10GbE and HBA cards and waiting for some SSDs so I can try ZFS pool too).
August 17, 20232 yr Author This was the first time in 20 years that I had problems with RAM memory. I was running mixed sizes, speeds, brands... new with used sticks... (it was in a various desktop machines). Many of my memory sticks came with *Lifetime warranty so I never had any suspicious for it. Turns out, memtest should not be skipped. For some future builds I may even consider ECC. This caused some erroneous files in docker file, so I had to rebuild a few images and to double check everything. Luckily everything is fine now.
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.