After updating BaseRow to 1.29.2 The login screen asks me to create a new user and No databases exist.

[n0cturnal79]

I have tried to restore to the backup before the update, and get this error in Postgresql the app errors out. Any help would be greatly appreciated. I will have serious issues in 11 hours if i cannot get this back running.:

clear: failed to clear database. error: database "db12e4ccd803504fd8bb15f4e0bc2bb78a" is being accessed by other users
Mar 26 18:05:36 at Parser.parseErrorMessage (/app/code/node_modules/pg-protocol/dist/parser.js:287:98)
Mar 26 18:05:36 at Parser.handlePacket (/app/code/node_modules/pg-protocol/dist/parser.js:126:29)
Mar 26 18:05:36 at Parser.parse (/app/code/node_modules/pg-protocol/dist/parser.js:39:38)
Mar 26 18:05:36 at Socket.<anonymous> (/app/code/node_modules/pg-protocol/dist/index.js:11:42)
Mar 26 18:05:36 at Socket.emit (node:events:517:28)
Mar 26 18:05:36 at addChunk (node:internal/streams/readable:335:12)
Mar 26 18:05:36 at readableAddChunk (node:internal/streams/readable:308:9)
Mar 26 18:05:36 at Readable.push (node:internal/streams/readable:245:10)
Mar 26 18:05:36 at TCP.onStreamRead (node:internal/stream_base_commons:190:23) {
Mar 26 18:05:36 length: 181,
Mar 26 18:05:36 severity: 'ERROR',
Mar 26 18:05:36 code: '55006',
Mar 26 18:05:36 detail: 'There are 5 other sessions using the database.',
Mar 26 18:05:36 hint: undefined,
Mar 26 18:05:36 position: undefined,
Mar 26 18:05:36 internalPosition: undefined,
Mar 26 18:05:36 internalQuery: undefined,
Mar 26 18:05:36 where: undefined,
Mar 26 18:05:36 schema: undefined,
Mar 26 18:05:36 table: undefined,
Mar 26 18:05:36 column: undefined,
Mar 26 18:05:36 dataType: undefined,
Mar 26 18:05:36 constraint: undefined,
Mar 26 18:05:36 file: 'dbcommands.c',
Mar 26 18:05:36 line: '942',
Mar 26 18:05:36 routine: 'dropdb'
Mar 26 18:05:36 }
Mar 26 18:05:36 clear: disconnected

[nebulon]

Try to restart the prostgres service in the services view. Seems like there is some dangling connection preventing the restore

[n0cturnal79]

I did try that, but no glory. I also tried to re-install and restore from the backup just previous to the update. It was about 1/3 of the size of my other backups, and did not work. I ended up restoring a backup from the previous day, and pushing my saved data from the form view to rebuild the day's work. It sucked, but got us back going.

I plan to retry the update this weekend (With a full VMS backup beforehand). Has anyone else had this issue?

[nebulon]

When you retry the update, maybe create a clone of that app instance first and try the update there to see if it was a one-off issue or if it is reproducible so we can try to fix something.

[rmdes]

Similar issue here, but my logs are a bit different , I have tried to restore from previous backup but havent managed to connect with my existing user of recovered any of my databases. Honestly a bit worried because one of these were a journalist project about to launch

trying a restore now with a postgres service restart

same situation

[rmdes]

If I clone from a previous backup, any backup, I end up in the same first admin creation screen

[rmdes]

I'm starting to think the problem may have happened even before, and that I have been making empty backups

I'm not sure how I can confirm this tho, but no matter what I restore, all I get is an empty Baserow

[nebulon]

The backups would contain the sql dump file, so you can check that in the raw format within your backups to confirm that or not. This certainly would narrow the issue angle. If the database sql dump is already mostly empty, we at least know the restore logic isn't the issue.