Close#34022 , #33550
This error message always appears when using the `check-attr` command,
even though it works correctly.
The issue occurs when the stdin writer is closed, so I added a special
case to handle and check the error message when the exit code is 1.
* reuse recoverable error checks across mirror_pull
* add new cases for 'cannot lock ref/not our ref' (race condition in
fetch) and 'Unable to create/lock"
* move lfs sync right after commit graph write, and before other
maintenance which may fail
* try a prune for 'broken reference' as well as 'not our ref'
* always sync LFS right after commit graph write, and before other
maintenance which may fail
This handles a few cases where our very large and very active
repositories could serve mirrored git refs, but be missing lfs files:
## Case 1 (multiple variants): Race condition in git fetch
There was already a check for 'unable to resolve reference' on a failed
git fetch, after which a git prune and then subsequent fetch are
performed. This is to work around a race condition where the git remote
tells Gitea about a ref for some HEAD of a branch, then fails a few
seconds later because the remote branch was deleted, or the ref was
updated (force push).
There are two more variants to the error message you can get, but for
the same kind of race condition. These *may* be related to the git
binary version Gitea has access to (in my case, it was 2.48.1).
## Case 2: githttp.go can serve updated git refs before it's synced lfs
oids
There is probably a more aggressive refactor we could do here to have
the cat-file loop use FETCH_HEAD instead of relying on the commit graphs
to be committed locally (and thus serveable to clients of Gitea), but a
simple reduction in the occurrences of this for me was to move the lfs
sync block immediately after the commit-graph write and before any other
time-consuming (or potentially erroring/exiting) blocks.
---------
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
The old logic is incomplete. See the comment for the improved logic.
Fix#34011
And more fixes:
1. use empty "alt" for images, otherwise the width is not right when the
image fails to load
2. remove the "dropdown icon" patch, because it has been clearly done in
"dropdown.js" now
3. remove the "dropdown filtered item" patch, added a clear callback,
and improve the logic
4. fix global init when a node is removed and added back gain (eg: the
"cherry pick" dialog with a dropdown)
blevesearch is skipped because it causes errors
---------
Co-authored-by: Lunny Xiao <xiaolunwen@gmail.com>
Co-authored-by: silverwind <me@silverwind.io>
1. Ignore empty inputs in `UnmarshalHandleDoubleEncode`
2. Ignore non-existing `stateEvent.User` in gitlab migration
3. Enable `release` and `wiki` units when they are selected in migration
4. Sanitize repo name for migration and new repo
The pagination on the user dashboard sounds unnecessary, this will
change it to a prev/next buttons. For instances with around `10 million`
records in the action table, this option affects how the user dashboard
is loaded on first visit.
---------
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Co-authored-by: Giteabot <teabot@gitea.io>
This change targets https://github.com/go-gitea/gitea/issues/32663
We drop the hardcoded timeout of 60 seconds for requests to the internal
hook api. With this change the timeout is completly removed.
---------
Co-authored-by: wxiaoguang <wxiaoguang@gmail.com>
Two SQLs are very slow when `action` table have over 5M records.
```
database duration=1.8881s db.sql="SELECT created_unix DIV 900 * 900 AS timestamp, count(user_id) as contributions FROM `action` WHERE user_id=? AND act_user_id=? AND (created_unix > ?) GROUP BY timestamp ORDER BY timestamp"
database duration=1.5408s db.sql="SELECT count(*) FROM `action` WHERE (user_id = ?) AND (is_deleted = ?)"
```
This will cache the count for the first loading or when the activities
changed.
When there are over 5M records on `action` table, the heatmap on
dashboard is very slow as below SQL.
```
database duration=1.8881s db.sql="SELECT created_unix DIV 900 * 900 AS timestamp, count(user_id) as contributions FROM `action` WHERE user_id=? AND act_user_id=? AND (created_unix > ?) GROUP BY timestamp ORDER BY timestamp"
```
This PR add a new index for `action` table with columns `user_id`,
`act_user_id` and `created_unix` so that this query will become about 6
times faster than before.
