SQLite - Unknown col &#34;is_empty&#34; in index is_bare in all Versions after 18.01.2019 #2791

Originally created by @wofnull on GitHub (Jan 22, 2019).

Gitea version (or commit ref): 2d94561
Git version: 2.11
Operating system: Debian
Database (use [x]):
- PostgreSQL
- MySQL
- MSSQL
- SQLite
Can you reproduce the bug at https://try.gitea.io:
- Yes (provide example URL)
- No
- Not relevant
Log gist:
2019/01/22 07:10:25 [I] Log Mode: File(Info)
2019/01/22 07:10:25 [I] XORM Log Mode: File(Info)
2019/01/22 07:10:25 [I] Cache Service Enabled
2019/01/22 07:10:25 [I] Session Service Enabled
2019/01/22 07:10:25 [I] Beginning ORM engine initialization.
2019/01/22 07:10:25 [I] ORM engine initialization attempt #1/10...
2019/01/22 07:10:25 [I] Backing off for 3 seconds
2019/01/22 07:10:28 [I] ORM engine initialization attempt #2/10...
2019/01/22 07:10:28 [I] Backing off for 3 seconds
2019/01/22 07:10:31 [I] ORM engine initialization attempt #3/10...
2019/01/22 07:10:31 [I] Backing off for 3 seconds
2019/01/22 07:10:34 [I] ORM engine initialization attempt #4/10...
2019/01/22 07:10:34 [I] Backing off for 3 seconds
2019/01/22 07:10:37 [I] ORM engine initialization attempt #5/10...
2019/01/22 07:10:37 [I] Backing off for 3 seconds
2019/01/22 07:10:40 [I] ORM engine initialization attempt #6/10...
2019/01/22 07:10:40 [I] Backing off for 3 seconds
2019/01/22 07:10:43 [I] ORM engine initialization attempt #7/10...
2019/01/22 07:10:43 [I] Backing off for 3 seconds
2019/01/22 07:10:46 [I] ORM engine initialization attempt #8/10...
2019/01/22 07:10:46 [I] Backing off for 3 seconds
2019/01/22 07:10:49 [I] ORM engine initialization attempt #9/10...
2019/01/22 07:10:49 [I] Backing off for 3 seconds
2019/01/22 07:10:52 [I] ORM engine initialization attempt #10/10...
2019/01/22 07:10:52 [...itea/routers/init.go:81 GlobalInit()] [E] ORM engine initialization failed: sync database struct error: Unknown col "is_empty" in index is_bare of table repository, columns [i
d owner_id lower_name name description website default_branch num_watches num_stars num_forks num_issues num_closed_issues num_pulls num_closed_pulls num_milestones num_closed_milestones is_private i
s_empty is_mirror is_fork fork_id size created_unix updated_unix is_fsck_enabled topics]

Description:

Since the weekend ( last sucessfull DB change Jan 18 01:58 gitea.db ) the gitea instance on my server failed to start after an automatic update to restart, it looks like a db change was not successfully made causing the server to fail on startup.

Switiching back to a version prior to the one from the 18th seems to work for the moment.

Originally created by @wofnull on GitHub (Jan 22, 2019). - Gitea version (or commit ref): 2d94561 - Git version: 2.11 - Operating system: Debian - Database (use `[x]`): - [ ] PostgreSQL - [ ] MySQL - [ ] MSSQL - [x] SQLite - Can you reproduce the bug at https://try.gitea.io: - [ ] Yes (provide example URL) - [ ] No - [x] Not relevant - Log gist: 2019/01/22 07:10:25 [I] Log Mode: File(Info) 2019/01/22 07:10:25 [I] XORM Log Mode: File(Info) 2019/01/22 07:10:25 [I] Cache Service Enabled 2019/01/22 07:10:25 [I] Session Service Enabled 2019/01/22 07:10:25 [I] Beginning ORM engine initialization. 2019/01/22 07:10:25 [I] ORM engine initialization attempt #1/10... 2019/01/22 07:10:25 [I] Backing off for 3 seconds 2019/01/22 07:10:28 [I] ORM engine initialization attempt #2/10... 2019/01/22 07:10:28 [I] Backing off for 3 seconds 2019/01/22 07:10:31 [I] ORM engine initialization attempt #3/10... 2019/01/22 07:10:31 [I] Backing off for 3 seconds 2019/01/22 07:10:34 [I] ORM engine initialization attempt #4/10... 2019/01/22 07:10:34 [I] Backing off for 3 seconds 2019/01/22 07:10:37 [I] ORM engine initialization attempt #5/10... 2019/01/22 07:10:37 [I] Backing off for 3 seconds 2019/01/22 07:10:40 [I] ORM engine initialization attempt #6/10... 2019/01/22 07:10:40 [I] Backing off for 3 seconds 2019/01/22 07:10:43 [I] ORM engine initialization attempt #7/10... 2019/01/22 07:10:43 [I] Backing off for 3 seconds 2019/01/22 07:10:46 [I] ORM engine initialization attempt #8/10... 2019/01/22 07:10:46 [I] Backing off for 3 seconds 2019/01/22 07:10:49 [I] ORM engine initialization attempt #9/10... 2019/01/22 07:10:49 [I] Backing off for 3 seconds 2019/01/22 07:10:52 [I] ORM engine initialization attempt #10/10... 2019/01/22 07:10:52 [...itea/routers/init.go:81 GlobalInit()] [E] ORM engine initialization failed: sync database struct error: Unknown col "is_empty" in index is_bare of table repository, columns [i d owner_id lower_name name description website default_branch num_watches num_stars num_forks num_issues num_closed_issues num_pulls num_closed_pulls num_milestones num_closed_milestones is_private i s_empty is_mirror is_fork fork_id size created_unix updated_unix is_fsck_enabled topics] ## Description: Since the weekend ( last sucessfull DB change Jan 18 01:58 gitea.db ) the gitea instance on my server failed to start after an automatic update to restart, it looks like a db change was not successfully made causing the server to fail on startup. Switiching back to a version prior to the one from the 18th seems to work for the moment.

GiteaMirror closed this issue

GiteaMirror commented

@zeripath commented on GitHub (Jan 22, 2019):

Please could you check the value of the version table in sqlite.

SELECT * FROM version

If your version is 79 you have two options:

Drop it back to 78 and restart - the migration script should work now. Please get back to us if not.
DROP INDEX IF EXISTS IDX_repository_is_bare

If your version is 78 the DROP command above should work but, shouldn't be necessary, and it was definitely working until the last change.

@zeripath commented on GitHub (Jan 22, 2019): Please could you check the value of the version table in sqlite. ```sql SELECT * FROM version ``` If your version is 79 you have two options: * Drop it back to 78 and restart - the migration script should work now. Please get back to us if not. * `DROP INDEX IF EXISTS IDX_repository_is_bare` If your version is 78 the DROP command above should work but, shouldn't be necessary, and it was definitely working until the last change.

GiteaMirror commented

@zeripath commented on GitHub (Jan 22, 2019):

Please get back to us to tell us what version you have

@zeripath commented on GitHub (Jan 22, 2019): Please get back to us to tell us what version you have

GiteaMirror commented

@wofnull commented on GitHub (Jan 22, 2019):

DB Version is 79 as suspected

Set it back to 78 , Update happened upon next restart of GITEA -> Access is now possible again.

@wofnull commented on GitHub (Jan 22, 2019): DB Version is 79 as suspected > Set it back to 78 , Update happened upon next restart of GITEA -> Access is now possible again.

GiteaMirror commented

@zeripath commented on GitHub (Jan 22, 2019):

Cool I'll close this issue.

@zeripath commented on GitHub (Jan 22, 2019): Cool I'll close this issue.

GiteaMirror commented

@xeals commented on GitHub (Jan 23, 2019):

I'm getting the same issue on DB version 74. Not sure how practical the above advice is due to skipping migrations. Any advice?

@xeals commented on GitHub (Jan 23, 2019): I'm getting the same issue on DB version 74. Not sure how practical the above advice is due to skipping migrations. Any advice?

GiteaMirror commented

@zeripath commented on GitHub (Jan 23, 2019):

Have you tried gitea migrate?

@zeripath commented on GitHub (Jan 23, 2019): Have you tried `gitea migrate`?

GiteaMirror commented

@xeals commented on GitHub (Jan 23, 2019):

I thought I had and was getting a non-working result, but it's now working. Sorry for the hassle.

@xeals commented on GitHub (Jan 23, 2019): I thought I had and was getting a non-working result, but it's now working. Sorry for the hassle.

GiteaMirror commented

@MNickelson commented on GitHub (Jan 25, 2019):

My database says it's version 79. I will try resetting it to 78 or dropping it tonight.

@MNickelson commented on GitHub (Jan 25, 2019): My database says it's version 79. I will try resetting it to 78 or dropping it tonight.

GiteaMirror commented

@andrew-from-toronto commented on GitHub (Jul 14, 2019):

I've just recently hit this as well - dropping index resolved (changing version to 78 broke things more), but seems like bad migration if it's happening to multiple people.

@andrew-from-toronto commented on GitHub (Jul 14, 2019): I've just recently hit this as well - dropping index resolved (changing version to 78 broke things more), but seems like bad migration if it's happening to multiple people.

GiteaMirror commented