This commit reimplements how we monitor Sidekiq processes that are
forked from the Unicorn master process. Prior to this change, we rely on
`Jobs::Heartbeat` to enqueue a `Jobs::RunHeartbeat` job every 3 minutes.
The `Jobs::RunHeartbeat` job then sets a Redis key with a timestamp. In
the Unicorn master process, we then fetch the timestamp that has been set
by the job from Redis every 30 minutes. If the timestamp has not been
updated for more than 30 minutes, we restart the Sidekiq process. The
fundamental flaw with this approach is that it fails to consider
deployments with multiple hosts and multiple Sidekiq processes. A
sidekiq process on a host may be in a bad state but the heartbeat check
will not restart the process because the `Jobs::RunHeartbeat` job is
still being executed by the working Sidekiq processes on other hosts.
In order to properly ensure that stuck Sidekiq processs are restarted,
we now rely on the [Sidekiq::ProcessSet](https://github.com/sidekiq/sidekiq/wiki/API#processes)
API that is supported by Sidekiq. The API provides us with "near real-time (updated every 5 sec)
info about the current set of Sidekiq processes running". The API
provides useful information like the hostname, pid and also when Sidekiq
last did its own heartbeat check. With that information, we can easily
determine if a Sidekiq process needs to be restarted from the Unicorn
master process.
* port CanCheckEmails mixin to helper class with explicit dependencies
* move isCurrentUser getter to components
* anonymous users should not be able to see sensitive information
This converts the `<AdminPageHeader />` component and the
`<AdminPageSubheader />` components into new components
that can be used outside of admin, and updates the CSS classes.
Also introduces a `<DPageActionButton />` component and child
components for the header action buttons.
I have to keep the old admin-only components around for
now until plugins are updated, then we can remove it,
and remove the re-exports that are done within
admin-page-action-button.gjs
When freeze_original option is passed to PostMover, and we are moving all posts there is an issue. We attempt to put the small_action right after the last moved post. The issue is when there is an existing small action after the last moved "real" post. We then try to put the moderator post at the same location of the existing small action, which causes an index conflict and the move fails.
This makes sure that we place the moderator post at the verrrrrry end of the topic :)
* UX: Apply admin table classes for consistent mobile styling on the web hooks page
* DEV: Remove icon on the status component; update status classes
* DEV: Update tests for webhook status component
* DEV: add space var with a smaller value
* DEV: Add styling for different status labels
When using FK select, we include a "None" option automatically. However, for required select fields, "None" isn't a valid option, so we exclude it instead.
These controller tests are passing locally and in CI, but are failing the build when run in parallel.
I managed to recreate the failures by running the entire suite with turbo_spec and the right seed locally. After these changes, the parallel suite passes locally as well. 🤞
`new_in_category` was using `first` instead of `limit`
This meant it gets an array and that means that you can not operate on it easily in a modifier.
This ensures we always give the modifier a relation, with the notable exception of suggested topics.
When adding new fonts to our discourse-fonts gem,
you may want to test these changes locally using the
`path` option of the gem line in bundler.
However we set a symlink for the fonts dir for public/fonts,
which does not get updated if the gem path changes. This
fix checks if the current symlink resolved path is equal to
the DiscourseFont gem path, and if not deletes + recreates it.
We're changing the default of hide_email_address_taken to true. This is a trade-off we want to make, as it prevents account enumeration with minimal impact on legitimate users. If you forget you have an account and try to sign up again with the same e-mail you'll receive an e-mail letting you know.
Add flag reason filter and improve handling of deleted content in review queue
This commit enhances the review queue with several key improvements:
1. Adds a new "Reason" filter to allow filtering flags by their score type
2. Improves UI for deleted content by:
- Adding visual indication for deleted posts (red background)
- Properly handling deleted content visibility for staff (category mods can not see deleted content)
3. Refactors reviewable score type handling for better code organization
4. Adds tests for trashed topics/posts visibility
This change will help moderators more efficiently manage the review queue by
being able to focus on specific types of flags and better identify deleted
content.
Using the (array) helper creates an array instance, passes it to applyValueTransformer, which then allows themes/plugins to mutate it.
When the helper is re-computed, the **same array instance** is passed to the transformers again. Any elements added in the last run are still there.
This commit moves applyValueTransformer to a getter, so that it's run with a brand new array each time.
This commit introduces <NotificationsTracking /> which is a wrapper component around <DMenu /> which replaces the select-kit component <TopicNotificationsButton />.
Each tracking case has its dedicated component:
- topic -> `<TopicNotificationsTracking />`
- group -> `<GroupNotificationsTracking />`
- tag -> `<TagNotificationsTracking />`
- category -> `<CategoryNotificationsTracking />`
- chat thread -> `<ThreadNotificationsTracking />`
When fetching DM channels, we would fetch all the user's DM channels,
and limit the number of DM channels we display on the client-side using
the `DIRECT_MESSAGE_CHANNELS_LIMIT`.
This is obviously not scalable as users are having more and more
discussions with other users.
This commit ensures we always limit the number of DM channels we send to the
client.
It adds the `MAX_DM_CHANNEL_RESULTS` in `ChannelFetcher` and use it to
limit the records returned by the server. This value should always be
lower than `DIRECT_MESSAGE_CHANNELS_LIMIT` on the client-side, so that
when the user has more "unread DMs" they still show up on the client as
they read them.
While adding a spec, I also updated the spec that ensures we have a
limit on public channels to use `stub_const` to temporarily lower the number
of fixtures we create during the test.
Previously, theme hbr files were compiled to an IIFE, which would be executed before the app is booted. That is causing silenced deprecations to be printed, because the deprecation-workflow isn't set up when the IIFE is run.
This commit updates the theme compiler so that it matches the ember-cli-based raw-hbs compiler. Templates are output to normal modules, which will then be loaded by the existing `eager-load-raw-templates` initializer. This runs after the app has started booting.
* DEV: add db consistency check for UserEmail
* DEV: add db consistency check for UserAvatar
* DEV: ignore inconsistent data related to user avatars when deciding whether to rebake old posts
Co-authored-by: Alan Guo Xiang Tan <gxtan1990@gmail.com>
---------
Co-authored-by: Alan Guo Xiang Tan <gxtan1990@gmail.com>
This reverts commit 9694dc6cb0.
Some of our previous email styling depended on this 'incorrect' ordering, so the change caused some text to become illegible. Reverting while we work out a better solution
