I made the upgrade yesterday and want to share what i went through and all the fixes I applied.
I started with the link:
https://enhance.com/docs/getting-started/upgrading-from-v11.html#prerequisites
When running the commands naturally the terminal would tell me "Next run xxyyzz" -
I IGNORED THIS and proceeded to follow the guide. - I'd say it's safe to assume this is where most people made an error.
-- Main confusion was mysql upgrade command // this one really makes you think but I figured there must be a reason we've been asked to run it last.
-- Step 4 was confusing because it doesn't specify WHERE to run it i think someone with less experience would not know where to run it, maybe should include that in a separate step saying run it on every server with webserver on it/ftp.
I did not run into any errors upon my installation process.
After installation I tested the core functions and everything seemed fine except for the backups.
- Could not see any backups
So I took the COMMON problem fixes advice:
I ran the command:
v12-upgrade fix-backup-permissions /<MYBACKUPLOCATIONHERE>
It produces a bunch of these
chown -R e44e1183034541daaf99fa5fc543fa95:e44e1183034541daaf99fa5fc543fa95 /mnt/blockstorage/e44e1183-0345-41da-af99-fa5fc543fa95
useradd -d /mnt/blockstorage/43d81831-a121-46cd-96ce-fc8b6acf8771 43d81831a12146cd96cefc8b6acf8771
I attempted to run all of these at once but the server couldn't handle it, so I made a small .sh file to slowly execute them one by one.
nano slow_useradd_chown.sh
File Contents:
#!/bin/bash
while read -r line; do
echo "Running: $line"
eval "$line"
sleep 1
done <<EOF
useradd -d /mnt/blockstorage/833bb4f0-27f6-44f4-ae6d-ccaae293aa9a 833bb4f027f644f4ae6dccaae293aa9a
chown -R 833bb4f027f644f4ae6dccaae293aa9a:833bb4f027f644f4ae6dccaae293aa9a /mnt/blockstorage/833bb4f0-27f6-44f4-ae6d-ccaae293aa9a
useradd -d /mnt/blockstorage/35e586d8-18f5-43a8-b8be-ba5c31518b44 35e586d818f543a8b8beba5c31518b44
chown -R 35e586d818f543a8b8beba5c31518b44:35e586d818f543a8b8beba5c31518b44 /mnt/blockstorage/35e586d8-18f5-43a8-b8be-ba5c31518b44
useradd -d /mnt/blockstorage/bbe2c28b-f825-4e11-ab8a-282accfdeb66 bbe2c28bf8254e11ab8a282accfdeb66
<AND ALL OF THEM HERE>
EOF
This took hours and did nothing...
I attempted to run:
apt update && apt install ecp-core
After running these - I could SEE MY BACKUPS, but could not restore or anything. Manual backups also were not being created.
I deployed a fresh new server
- Set it as backup only
- Decomissioned old backup server
- Re-assigned all websites using MOVE SERVER to new backup server
Then backups were still not working correctly, manual backups were displaying as 250bytes - 300bytes, restores were failing.
On the CONTROL PANEL SERVER:
I had to run this (as per suggestion by support) - This command took over 2 hours to actually run.
v12-upgrade update-all-website-backup-locations
After this backups were working, visible, restoring, manual backups all tested and fine.
Then i discovered that my automatic backups weren't engaging. But this was expected as the minimum backup age i had set was 4 hours so there was a 4 hour delay, and they began running as intended.
In the morning I discovered we were NOT RECEIVING emails from anyone outside of our cluster.
(ie. @gmail.com > couldnt send to my email)
As per support recommendation:
/var/spool/postfix/etc/resolv.conf
Updated to use: 8.8.8.8
rather than the systemd-resolved IP
postfix stop; postfix start
Emails are now operating just fine.
So far my only issue is backups are taking extra space but this is not an alarming issue and was expected with the backup issues already being widely reported.
- Median RAM usage has dropped from 55-60% to 40-41% staying solid (even in peak hour here).
- Backups seem to be faster and using LESS CPU usage.
- Backups are TAKING extra space - I think they're not incremental anymore.
