CountDivergingCommits: exit status 128 - fatal: ambiguous argument 'master..5297-allows-unicode': unknown revision or path not in the working tree.

[Alir3z4]

• Gitea version (or commit ref): 1.9.0+rc2-16-gd789170e3 built with GNU Make 4.1, go1.12.7 : bindata, sqlite, sqlite_unlock_notify
• Git version: 2.7.4
• Operating system: Ubuntu 16.04 LTS
• 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:

2019/07/29 14:14:12 .../xorm/session_get.go:99:nocacheGet() [I] [SQL] SELECT "id", "user_id", "repo_id" FROM "watch" WHERE "user_id"=$1 AND "repo_id"=$2 LIMIT 1 []interface {}{1, 90}
2019/07/29 14:14:12 .../xorm/session_get.go:99:nocacheGet() [I] [SQL] SELECT "id", "uid", "repo_id" FROM "star" WHERE "uid"=$1 AND "repo_id"=$2 LIMIT 1 []interface {}{1, 90}
2019/07/29 14:14:12 ...orm/session_exist.go:89:Exist() [I] [SQL] SELECT "id", "repo_id", "branch_name", "can_push", "enable_whitelist", "whitelist_user_i_ds", "whitelist_team_i_ds", "enable_merge_whitelist", "merge_whitelist_user_i_ds", "merge_whitelist_team_i_ds", "approvals_whitelist_user_i_ds", "approvals_whitelist_team_i_ds", "required_approvals", "created_unix", "updated_unix" FROM "protected_branch" WHERE "repo_id"=$1 AND "branch_name"=$2 LIMIT 1 []interface {}{90, "5297-allows-unicode"}
2019/07/29 14:14:12 ...uters/repo/branch.go:182:loadBranches() [E] CountDivergingCommits: exit status 128 - fatal: ambiguous argument 'master..5297-allows-unicode': unknown revision or path not in the working tree.
        Use '--' to separate paths from revisions, like this:
        'git <command> [<revision>...] -- [<file>...]'

2019/07/29 14:14:12 ...c/net/http/server.go:3012:logf() [I] http: superfluous response.WriteHeader call from code.gitea.io/gitea/vendor/gopkg.in/macaron%2ev1.(*responseWriter).WriteHeader (response_writer.go:59)

Description

Tried to open a repository branches page and faced the 500 error.

You can visit the URL as see the issue https://gitfoo.com/Alir3z4/django-cms/branches/

...

Screenshots

[zeripath]

This is very likely the same issue as your other report. The master branch is not in the correct place therefore it can't be found

[zeripath]

This is very likely the same issue as your other report. The master branch is not in the correct place therefore it can't be found

[Alir3z4]

Correct, but how can be prevented from happening again ?
The repo has been migrated via migration interface.

[lunny]

Have you checked the filesystem permissions?

[Alir3z4]

Yep, they're all correctly set as the user that runs the gitea git:git

ls -la .../gitea/gitea-repositories/alir3z4/django-cms.git/
total 52
drwxr-xr-x  7 git git 4096 Jul 29 14:03 .
drwxr-x--- 74 git git 4096 Jul 29 13:22 ..
drwxr-xr-x  2 git git 4096 Jul 28 17:24 branches
-rw-r--r--  1 git git   66 Jul 28 17:25 config
-rw-r--r--  1 git git   73 Jul 28 17:24 description
-rw-r--r--  1 git git   30 Jul 28 17:25 HEAD
drwxr-xr-x  5 git git 4096 Jul 28 17:25 hooks
drwxr-xr-x  2 git git 4096 Jul 29 14:03 info
drwxr-xr-x  4 git git 4096 Jul 28 17:24 objects
-rw-r--r--  1 git git 8345 Jul 29 14:03 packed-refs
drwxr-xr-x  4 git git 4096 Jul 28 17:24 refs

[zeripath]

I suspect you're pushing to that repository without going through Gitea. How are you pushing?

[Alir3z4]

I have not pushed to this repo in anyway, I've just migrated it over from github.

I cloned, pushed, deleted etc other repos on Gitea and all works fine.

[lunny]

So maybe some migrated branches broken the page?

[lunny]

The branch 5297-allows-unicode has no commit base with master.

[zeripath]

Might this be an issue similar to #7610 whereby in 1.9 it appears that the default branch isn't updated properly - however I think on try this doesn't happen?

[Alir3z4]

IMO what is more important here, is how and why such thing has happened and how it can be prevented from happening again, because the problem has not happened via any custom or unofficial way, the repo has been migrated via the Gitea migration interface.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs during the next 2 weeks. Thank you for your contributions.