The "check network connection and try again" problems to 5.20 (solved)

First, congratulations on finding this community forum board that might have the solution to your problem. If only Dell/EMC could create upgrade scripts that WORKS then we wouldn't be here.

This post are the steps I took for the VMware VE edition 5.16 to 5.20

1. Get into a command line shell - either the remote console or ssh with the root account (not webUI admin account).
2. The app is in a docker container, get into it via docker exec -it esrsde-app bash
3. Check the update log file (command less is not installed) cat /var/log/esrs/esrsupdate.log | more
4. Is there an error code right there? "in log files I noticed  that it is looking for /root/esrsverepos/SCG-Update-5.20.00.10.zip" pbednarsh's post, gatisa reply
5. If you're lucky, you can just make the zip file as gatisa did, REBOOT, and then go back to the web UI to finish the update. The message on the update page said the file was downloaded and ready to apply.
6. I was not that lucky, as I had tried to start the update from the UI a few times and rebooted at an inappropriate time since the UI did not tell me what was going on in the background...
   1. From that command prompt, already inside the docker container - check and clear stuck processes in the database (taken from "Unable to upgrade from GUI - 5.12 to 5.16")

   2. su -l postgres -c "psql -d vapp"

   3. update mft_sessions set enddatetime=current_timestamp where enddatetime is null;

   4. update remote_session_access_audit set endtimestamp=current_timestamp where endtimestamp is null;

   5. delete from version_details where download_completed = 'N';

   6. \q  (to exit)
   7. I had a row return for the version_details but not the other two.
   8. exit out of the docker container
   9. reboot the virtual appliance
7. Once rebooted and ideally a fresh start of the update, get back into a command line

8. hostname# docker exec -it esrsde-app bash -c 'tail -f /var/log/esrs/esrsupdate.log'

9. Now that you are watching the update log, get into the web UI and start the download/update again
10. BE PATIENT! The log file will sometimes sit there for a bit with no obvious changes when downloading the update. Maybe 10 minutes or so (depending on your bandwidth) of no log file updates is when I would start to worry that something isn't right. Watch the virtual server advanced performance stats to watch for network traffic/cpu/etc
11. This is the point where I had to make the zip file, reboot and finish (steps 4/5 above).
12. There will be multiple reboots of the docker container - kicking your tail of the logs and at least one reboot of the full virtual server - kicking your command line. Get back in and continue to be patient.
13. You'll know things are hopefully still updating when the web UI is unavailable. For me this happened after the full reboot and took a while (5-10minutes?) to come back up.

I did not need to run ./SCGUpdateFix.sh and did not need to modify any files (just the db record) to get it to work.

Good luck