Context: https://docs.joinmastodon.org/user/network/#search
Vanilla Mastodon intentionally refuses to search outside a user's
own toots, favs, bookmarks, and mentions. This flag makes that
restricted search behavior a per-instance choice, defaulting
to the same behavior as vanilla Mastodon if the flag is absent.
As pointed out by https://github.com/mastodon/mastodon/pull/21297#discussion_r1028372193
at least one of our dependencies already dropped support for Ruby 2.6, and we
had removed Ruby 2.6 tests from the CI over a year ago (#16861).
So stop advertising Ruby 2.6 support, bump targeted version, and drop some
compatibility code.
* refactor(statuses_vacuum): remove dead code - unused
Method is not called inside class and private.
Clean up dead code.
* refactor(statuses_vacuum): make retention_period present test explicit
This private method only hides functionality.
It is best practice to be as explicit as possible.
* refactor(statuses_vacuum): improve query performance
- fix statuses_scope having sub-select for Account.remote scope by
`joins(:account).merge(Account.remote)`
- fix statuses_scope unnecessary use of `Status.arel_table[:id].lt`
because it is inexplicit, bad practice and even slower than normal
`.where('statuses.id < ?'`
- fix statuses_scope remove select(:id, :visibility) for having reusable
active record query batches (no re queries)
- fix vacuum_statuses! to use in_batches instead of find_in_batches,
because in_batches delivers a full blown active record query result,
in stead of an array - no requeries necessary
- send(:unlink_from_conversations) not to perform another db query, but
reuse the in_batches result instead.
- remove now obsolete remove_from_account_conversations method
- remove_from_search_index uses array of ids, instead of mapping
the ids from an array - this should be more efficient
- use the in_batches scope to call delete_all, instead of running
another db query for this - because it is again more efficient
- add TODO comment for calling models private method with send
* refactor(status): simplify unlink_from_conversations
- add `has_many through:` relation mentioned_accounts
- use model scope local instead of method call `Status#local?`
- more readable add account to inbox_owners when account.local?
* refactor(status): searchable_by way less sub selects
These queries all included a sub-select. Doing the same with a joins
should be more efficient.
Since this method does 5 such queries, this should be significant,
since it technically halves the query count.
This is how it was:
```ruby
[3] pry(main)> Status.first.mentions.where(account: Account.local, silent: false).explain
Status Load (1.6ms) SELECT "statuses".* FROM "statuses" WHERE "statuses"."deleted_at" IS NULL ORDER BY "statuses"."id" DESC LIMIT $1 [["LIMIT", 1]]
Mention Load (1.5ms) SELECT "mentions".* FROM "mentions" WHERE "mentions"."status_id" = $1 AND "mentions"."account_id" IN (SELECT "accounts"."id" FROM "accounts" WHERE "accounts"."domain" IS NULL) AND "mentions"."silent" = $2 [["status_id", 109382923142288414], ["silent", false]]
=> EXPLAIN for: SELECT "mentions".* FROM "mentions" WHERE "mentions"."status_id" = $1 AND "mentions"."account_id" IN (SELECT "accounts"."id" FROM "accounts" WHERE "accounts"."domain" IS NULL) AND "mentions"."silent" = $2 [["status_id", 109382923142288414], ["silent", false]]
QUERY PLAN
------------------------------------------------------------------------------------------------------------------
Nested Loop (cost=0.15..23.08 rows=1 width=41)
-> Seq Scan on accounts (cost=0.00..10.90 rows=1 width=8)
Filter: (domain IS NULL)
-> Index Scan using index_mentions_on_account_id_and_status_id on mentions (cost=0.15..8.17 rows=1 width=41)
Index Cond: ((account_id = accounts.id) AND (status_id = '109382923142288414'::bigint))
Filter: (NOT silent)
(6 rows)
```
This is how it is with this change:
```ruby
[4] pry(main)> Status.first.mentions.joins(:account).merge(Account.local).active.explain
Status Load (1.7ms) SELECT "statuses".* FROM "statuses" WHERE "statuses"."deleted_at" IS NULL ORDER BY "statuses"."id" DESC LIMIT $1 [["LIMIT", 1]]
Mention Load (0.7ms) SELECT "mentions".* FROM "mentions" INNER JOIN "accounts" ON "accounts"."id" = "mentions"."account_id" WHERE "mentions"."status_id" = $1 AND "accounts"."domain" IS NULL AND "mentions"."silent" = $2 [["status_id", 109382923142288414], ["silent", false]]
=> EXPLAIN for: SELECT "mentions".* FROM "mentions" INNER JOIN "accounts" ON "accounts"."id" = "mentions"."account_id" WHERE "mentions"."status_id" = $1 AND "accounts"."domain" IS NULL AND "mentions"."silent" = $2 [["status_id", 109382923142288414], ["silent", false]]
QUERY PLAN
------------------------------------------------------------------------------------------------------------------
Nested Loop (cost=0.15..23.08 rows=1 width=41)
-> Seq Scan on accounts (cost=0.00..10.90 rows=1 width=8)
Filter: (domain IS NULL)
-> Index Scan using index_mentions_on_account_id_and_status_id on mentions (cost=0.15..8.17 rows=1 width=41)
Index Cond: ((account_id = accounts.id) AND (status_id = '109382923142288414'::bigint))
Filter: (NOT silent)
(6 rows)
```
When a poll is edited, we reset the poll and remove all previous
votes. However, prior to this commit, the voter count on the poll
was not reset. This leads to incorrect percentages being shown in
poll results.
Fixes#21696
* fix(chart): Remove non-functional Horizontal Pod Autoscaler
The Horizontal Pod Autoscaler (HPA) refers to a Deployment that
doesn't exist and therefore can not work. As a result it's
pointless to carry it around in this chart and give the wrong
impression it could work. This patch removes it from the helm
chart and drops all references to it.
* refactor(chart): Refactor sidekiq deployments to scale
This patch reworks how the sidekiq deployment is set up, by
splitting it into many sidekiq deployments, but at least one,
which should allow to scale the number of sidekiq jobs as
expected while being friendly to single user instances as well
as larger ones.
Further it introduces per deployment overwrites for the most
relevant pod fields like resources, affinities and processed
queues, number of jobs and the sidekiq security contexts.
The exact implementation was inspired by an upstream issue:
https://github.com/mastodon/mastodon/issues/20453
* fix(chart): Remove linode default values from values
This patch drops the linode defaults from the values.yaml since
these are not obvious and can cause unexpected connections as
well as leaking secrets to linode, when other s3 storage
backends are used and don't explicitly configure these options
by accident.
Mastodon will then try to authenticate to the linode backends
and therefore disclose the authentication secrets.
* refactor(chart): Rework reduce value reference duplication
Since most of the values are simply setup like this:
```
{{- if .Values.someVariable }}
SOME_VARIABLE: {{ .Values.someVariable }}
{{- end }}
```
There is a lot of duplication in the references in order to
full in the variables. There is an equivalent notation, which
reduces the usage of the variable name to just once:
```
{{- with .Values.someVariable }}
SOME_VARIABLE: {{ . }}
{{- end }}
```
What seems like a pointless replacement, will reduce potential
mistakes down the line by possibly only adjusting one of the
two references.
* fix(chart): Switch to new OMNIAUTH_ONLY variable
This patch adjusts the helm chart to use the new `OMNIAUTH_ONLY`
variable, which replaced the former
`OAUTH_REDIRECT_AT_SIGN_IN` variable in the following commit:
https://github.com/mastodon/mastodon/pull/172883c8857917e
* fix(chart): Repair connection test to existing service
Currently the connect test can't work, since it's connecting to
a non-existing service this patch fixes the service name to
make the job connect to the mastodon web service to verify the
connection.
* docs(chart): Adjust values.yaml to support helm-docs
This patch updates most values to prepare an introduction of
helm-docs. This should help to make the chart more user
friendly by explaining the variables and provide a standardised
README file, like many other helm charts do.
References:
https://github.com/norwoodj/helm-docs
* refactor(chart): Allow individual overwrites for streaming and web deployment
This patch works how the streaming and web deployments work by
adding various fields to overwrite values such as affinities,
resources, replica count, and security contexts.
BREAKING CHANGE: This commit removes `.Values.replicaCount` in
favour of `.Values.mastodon.web.replicas` and
`.Values.mastodon.streaming.values`.
* feat(chart): Add option for authorized fetch
Currently the helm chart doesn't support authorized fetch aka.
"Secure Mode" this patch fixes that by adding the needed config
option to the values file and the configmap.
* docs(chart): Improve helm-docs compatiblity
This patch adjust a few more comments in the values.yaml to be
picked up by helm-docs. This way, future adoption is properly
prepared.
* fix(chart): Add automatic detection of scheduler sidekiq queue
This patch adds an automatic switch to the `Recreate` strategy
for the sidekiq Pod in order to prevent accidental concurrency
for the scheduler queue.
* fix(chart): Repair broken DB_POOL variable
