Grafana 10.2 upgrade doesn't fix db tables on sqlite

Grafana Installation
1

Hello Community,
I’m having troubles starting Grafana oss after an upgrade from 9.5 to 10.2 on RedHat 7.9

The postscript already failed to start the service, and in the log there are the following lines

logger=migrator t=2023-11-02T16:41:11.562312858+01:00 level=error msg=“Executing migration failed” id=“copy correlation v1 to v2” error=“no such column: config” duration=126.263µs
logger=migrator t=2023-11-02T16:41:11.562343522+01:00 level=error msg=“Exec failed” error=“no such column: config” sql=“INSERT INTO correlation (target_uid\n, label\n, description\n, config\n, uid\n, source_uid) SELECT target_uid\n, label\n, description\n, config\n, uid\n, source_uid FROM correlation_tmp_qwerty”
logger=migrator t=2023-11-02T16:41:12.114993591+01:00 level=error msg=“Executing migration failed” id=“copy correlation v1 to v2” error=“no such column: config” duration=116.596µs
logger=migrator t=2023-11-02T16:41:12.115024839+01:00 level=error msg=“Exec failed” error=“no such column: config” sql=“INSERT INTO correlation (uid\n, source_uid\n, target_uid\n, label\n, description\n, config) SELECT uid\n, source_uid\n, target_uid\n, label\n, description\n, config FROM correlation_tmp_qwerty”
logger=migrator t=2023-11-02T16:41:12.608657424+01:00 level=error msg=“Executing migration failed” id=“copy correlation v1 to v2” error=“no such column: config” duration=114.461µs
logger=migrator t=2023-11-02T16:41:12.608687242+01:00 level=error msg=“Exec failed” error=“no such column: config” sql=“INSERT INTO correlation (description\n, config\n, uid\n, source_uid\n, target_uid\n, label) SELECT description\n, config\n, uid\n, source_uid\n, target_uid\n, label FROM correlation_tmp_qwerty”
logger=migrator t=2023-11-02T16:41:13.11222051+01:00 level=error msg=“Executing migration failed” id=“copy correlation v1 to v2” error=“no such column: config” duration=113.648µs
logger=migrator t=2023-11-02T16:41:13.112251884+01:00 level=error msg=“Exec failed” error=“no such column: config” sql=“INSERT INTO correlation (config\n, uid\n, source_uid\n, target_uid\n, label\n, description) SELECT config\n, uid\n, source_uid\n, target_uid\n, label\n, description FROM correlation_tmp_qwerty”
logger=migrator t=2023-11-02T16:41:13.613869402+01:00 level=error msg=“Executing migration failed” id=“copy correlation v1 to v2” error=“no such column: config” duration=114.097µs
logger=migrator t=2023-11-02T16:41:13.613898003+01:00 level=error msg=“Exec failed” error=“no such column: config” sql=“INSERT INTO correlation (target_uid\n, label\n, description\n, config\n, uid\n, source_uid) SELECT target_uid\n, label\n, description\n, config\n, uid\n, source_uid FROM correlation_tmp_qwerty”

Is there any script that can be run manually to adjust the DB according to the new structure?
I’ve also tried to start the version 10.2 without a DB, and manually adjusting the DB, but it still complains since correlation_tmp_qwerty doesn’t exist even there.

Have you already stumbled on this issue?

Kind regards,

2

welcome @gb_frt

What steps did you take to upgrade?

3

Hello @yosiasz ,
very simple:
I’ve stopped the service, backed up the database, plug-ins and conf files and installed the new version using

dnf install -y https://dl.grafana.com/oss/release/grafana-10.2.0-1.x86_64.rpm

The installation run smoothly until it tried to start the service

grafana-10.2.0-1.x86_64.rpm 50 MB/s | 98 MB 00:01
Dependencies resolved.

Package Arch Version Repository Size

Upgrading:
grafana x86_64 10.2.0-1 @commandline 98 M

Transaction Summary

Upgrade 1 Package

Total size: 98 M
Downloading Packages:
Running transaction check
Transaction check succeeded.
Running transaction test
Transaction test succeeded.
Running transaction
Preparing : 1/1
Upgrading : grafana-10.2.0-1.x86_64 1/2
Running scriptlet: grafana-10.2.0-1.x86_64 1/2
Job for grafana-server.service failed because the control process exited with error code. See “systemctl status grafana-server.service” and “journalctl -xe” for details.
warning: %post(grafana-10.2.0-1.x86_64) scriptlet failed, exit status 1

Error in POSTIN scriptlet in rpm package grafana
Cleanup : grafana-9.5.9-1.x86_64 2/2
Running scriptlet: grafana-10.2.0-1.x86_64 2/2
POSTTRANS: Running script

Verifying : grafana-10.2.0-1.x86_64 1/2
Verifying : grafana-9.5.9-1.x86_64 2/2

Upgraded:
grafana-10.2.0-1.x86_64

Complete!

4

Hello,
am I the only one that faced this issue?

besides the admin password, what else might miss importing the dump of the previous version of the DB in a newly created one?