This PR adds the back end for API token resource limits.
It adds the limit to the schema and checks the limit in the service.
## Discussion points
The PAT service uses a different service and different store entirely,
so I have not included testing any edge cases where PATs are included.
However, that could be seen as "knowing too much". We could add tests
that check both of the stores in tandem, but I think it's overkill for
now.
This PR updates the Unleash UI to use the new environment limit.
As it turns out, we already had an environment limit in the UI, but it
was hardcoded (luckily, its value is the same as the new default value
🥳).
In addition to the existing places this limit was used, it also disables
the "new environment" button if you've reached the limit. Because this
limit already exists, I don't think we need a flag for it. The only
change is that you can't click a button (that should be a link!) that
takes you to a page you can't do anything on.
This PR adds limits for environments to the resource limit schema. The
actual limiting will have to be done in Enterprise, however, so this is
just laying the groundwork.
This fixes the issue where project names that are 100 characters long
or longer would cause the project creation to fail. This is because
the resulting ID would be longer than the 100 character limit imposed
by the back end.
We solve this by capping the project ID to 90 characters, which leaves
us with 10 characters for the suffix, meaning you can have 1 billion
projects (999,999,999 + 1) that start with the same 90
characters (after slugification) before anything breaks.
It's a little shorter than what it strictly has to be (we could
probably get around with 95 characters), but at this point, you're
reaching into edge case territory anyway, and I'd rather have a little
too much wiggle room here.
This PR removes the last two flags related to the project managament
improvements project, making the new project creation form GA.
In doing so, we can also delete the old project creation form (or at
least the page, the form is still in use in the project settings).
This PR:
- adds a flag to anonymize user emails in the new project cards
- performs the anonymization using the existing `anonymise` function
that we have.
It does not anonymize the system user, nor does it anonymize groups. It
does, however, leave the gravatar url unchanged, as that is already
hashed (but we may want to hide that too).
This PR also does not affect the user's name or username. Considering
the target is the demo instance where the vast majority of users don't
have this (and if they do, they've chosen to set it themselves), this
seems an appropriate mitigation.
With the flag turned off:
With the flag on:
Fix project role assignment for users with `ADMIN` permission, even if
they don't have the Admin root role. This happens when e.g. users
inherit the `ADMIN` permission from a group root role, but are not
Admins themselves.
---------
Co-authored-by: Gastón Fournier <gaston@getunleash.io>
This PR adds metrics tracking for:
- "maxConstraintValues": the highest number of constraint values that
are in use
- "maxConstraintsPerStrategy": the highest number of constraints used on
a strategy
It updates the existing feature strategy read model that returns max
metrics for other strategy-related things.
It also moves one test into a more fitting describe block.
Instead of running exists on every row, we are joining the exists, which
runs the query only once.
This decreased load time on my huge dataset from 2000ms to 200ms.
Also added tests that values still come through as expected.
Instead of running exists on every row, we are joining the exists, which
runs the query only once.
This decreased load time on my huge dataset from 2000ms to 200ms.
Also added tests that values still come through as expected.
**Upgrade to React v18 for Unleash v6. Here's why I think it's a good
time to do it:**
- Command Bar project: We've begun work on the command bar project, and
there's a fantastic library we want to use. However, it requires React
v18 support.
- Straightforward Upgrade: I took a look at the upgrade guide
https://react.dev/blog/2022/03/08/react-18-upgrade-guide and it seems
fairly straightforward. In fact, I was able to get React v18 running
with minimal changes in just 10 minutes!
- Dropping IE Support: React v18 no longer supports Internet Explorer
(IE), which is no longer supported by Microsoft as of June 15, 2022.
Upgrading to v18 in v6 would be a good way to align with this change.
TS updates:
* FC children has to be explicit:
https://stackoverflow.com/questions/71788254/react-18-typescript-children-fc
* forcing version 18 types in resolutions:
https://sentry.io/answers/type-is-not-assignable-to-type-reactnode/
Test updates:
* fixing SWR issue that we have always had but it manifests more in new
React (https://github.com/vercel/swr/issues/2373)
---------
Co-authored-by: kwasniew <kwasniewski.mateusz@gmail.com>
This change adds a test to the tags API to ensure that even if you
can't create tags that are pure whitespace anymore, you'll still
receive pre-existing tags from the API that fit this description.
The test is here to ensure that we don't break this in future versions
of Unleash.
This PR fixes how Unleash handles tag values. Specifically, it does
these things:
1. Trims leading and trailing whitespace from tag values before
inserting them into the database
2. Updates OpenAPI validation to not allow whitespace-only and to ignore
leading and trailing whitespace
Additionally, it moves the tag length constants into the constants file
from the Joi tag schema file. This is because importing the values
previously rendered them as undefined (probably due to a circular
dependency somewhere in the system). This means that the previous values
were also ignored by OpenAPI.
UI updates reflecting this wil follow.
## Background
When you tag a flag, there's nothing stopping you from using an entirely
empty tag or a tag with leading/trailing whitespace.
Empty tags make little sense and leading trailing whitespace differences
are incredibly subtle:
Additionally, leading and trailing whitespace is not shown in the
dropdown list, so you'd have to guess at which is the right one.
Joining might not always be the best solution. If a table contains too
much data, and you later run sorting on top of it, it will be slow.
In this case, we will first reduce the instances table to a minimal
version because instances usually share the same SDK versions. Only
after that, we join.
Based on some customer data, we reduced query time from 3000ms to 60ms.
However, this will vary based on the number of instances the customer
has.
This PR removes the flag for the new project card design, making it GA.
It also removes deprecated components and updates one reference (in the
groups card) to the new components instead.
This PR removes all the feature flags related to the project list split
and updates the snapshot.
Now the project list will always contain "my projects" and "other
projects"
## About the changes
After an internal conversation, we concluded that syncExternalGroups is
an action that Unleash performs as a system, not something triggered by
the user. We keep the method and just write the event log that the
action was performed by the system user.
We have an issue that if you open up Insights, the Time to Production
chart was showing nothing because it was taking the median across all
projects. You might have many new or empty projects where the median was
0 (no data).
For example, the median from [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1.7, 50.3, 140] was 0.
Now, we will remove the 0 values to have a more reasonable median.
## About the changes
Removes the deprecated state endpoint, state-service (despite the
service itself not having been marked as deprecated), and the file
import in server-impl. Leaves a TODO in place of where file import was
as traces for a replacement file import based on the new import/export
functionality
Removes /edge/metrics. This has been superseded by
/api/client/metrics/bulk. Once this is merged, Unleash 6.0 will require
Edge > 17.0.0. (We recommend at least v19.1.3)
We are keeping the UI hidden for mdsol behind kill switch, but I feel
like we can remove the flag completely for backend, so everyone will
keep collecting data.
Co-authored-by: Gitar Bot <noreply@gitar.co>
## About the changes
This aligns us with the requirement of having ip in all events. After
tackling the enterprise part we will be able to make the ip field
mandatory here:
2c66a4ace4/src/lib/types/events.ts (L362)
In preparation for v6, this PR removes usage and references to
`error.description` instead favoring `error.message` (as mentioned
#4380)
I found no references in the front end, so this might be (I believe it
to be) all the required changes.
This PR is part of #4380 - Remove legacy `/api/feature` endpoint.
## About the changes
### Frontend
- Removes the useFeatures hook
- Removes the part of StrategyView that displays features using this
strategy (not been working since v4.4)
- Removes 2 unused features entries from routes
### Backend
- Removes the /api/admin/features endpoint
- Moves a couple of non-feature related tests (auth etc) to use
/admin/projects endpoint instead
- Removes a test that was directly related to the removed endpoint
- Moves a couple of tests to the projects/features endpoint
- Reworks some tests to fetch features from projects features endpoint
and strategies from project strategies
## About the changes
EdgeService is the only place where we use active tokens validation in
bulk. By switching to validating from the cache, we no longer need a
method to return all active tokens from the DB.
We talked with @nunogois that this test is testing migration from 2023,
but time has passed and the migration is working properly, so we think
to cut down on test run time, we can remove it.
## About the changes
We've identified that Bearer token middleware is not working for
/enterprise instance.
Looking at a few lines below:
88e3b1b79e/src/lib/app.ts (L81-L84)
we can see that we were missing the basePath in the use definition.
## About the changes
Current state, when returning the HTML entry point from the server,
there are no headers attached. We encountered an issue with a deployment
and this had an impact for us.
A brief description:
1. We deployed the most recent version. Noticed an unrelated issue.
2. Users tried to use the most recent version and due to their client
cache, requested assets that did not exist in the newest version.
3. Our cache layer cached the assets that were not there with the HTML
response. It had to infer the type based on the filename because there
was no attached `Content-Type` header. This cache was very sticky.
4. After rolling back we saw the HTML response (from the cache) instead
of the appropriate response from the upstream Unleash application.
This PR does a few things.
1. When responding with the HTML entry point, it adds header
(`Content-Type: text/html`).
2. When the client is requesting an asset (a path that ends with an
extension), it also instructs the resource not to be cached
(`Cache-Control: no-cache`) and returns a 404. This will prevent misses
from getting cached.
## Discussion points
To me, there doesn't seem to be a lot of test infra on serving the SPA
application. If that is an error, please indicate where that is and an
appropriate test can be added.
Adds a postgres_version gauge to allow us to see postgres_version in
prometheus and to post it upstream when version checking. Depends on
https://github.com/bricks-software/version-function/pull/20 to be merged
first to ensure our version-function doesn't crash when given the
postgres-version data.
1. Added new schema and tests
2. Controller also accepts the data
3. Also sending fake data from frontend currently
Next steps, implement service/store layer and frontend
This fixes the case when a customer have thousands of strategies causing
the react UI to crash. We still consider it incorrect to use that amount
of strategies and this is more a workaround to help the customer out of
a crashing state.
We put it behind a flag called `manyStrategiesPagination` and plan to
only enable it for the customer in trouble.
Now we are also sending project id to prometheus, also querying from
database. This sets us up for grafana dashboard.
Also put the metrics behind flag, just incase it causes cpu/memory
issues.
This PR updates the project service to automatically create a project id
if it is not provided. The feature is behind a flag. If an ID is
provided, it will still attempt to use that ID instead.
This PR adds a function to automatically generate a project ID on
creation. Using this when the id is missing will be handled in following
PRs.
The function uses the existing `slug` package to create a slug, and then
takes the 12 characters of a uuidv4 string to generate an ID.
The included tests check that the 12 character hash is added and that
the resulting string is url friendly (by checking that
`encodeURIComponent` doesn't change it).
We could also test a lot of edge cases (such as dealing with double
spaces, trimming the string, etc), but I think that's better handled by
the library itself (but you can check out what I removed in
2d9bcb6390
for an idea).
The function doesn't really need to be in the service; it could be moved to a util. But for proximity, I'll create it here first.
Regarding ticket #6892:
I would like to enable the use of a CA certificate without requiring
other certificates. This would be useful for AWS Helm, as AWS only
provides a single PEM file for DB connections.
We are getting questions from engineers, why I do not see lifecycle. The
same will happen with our customers. Now customers will see lifecycle
component unified across features.
Final rank has always been ordering correctly by default. But after 5.12
I see some issues that sometimes it is not ordered. Just to be extra
sure, I am for ordering it.
Add a flag to enable/disable the new UI for project creation.
This flag is separate from the impl on the back end so that we can
enable one without the other (but uses flag dependencies in Unleash, so
that we can never enable the new UI without the new back end).
I have not set the flag to `true` in server startup because the form
doesn't work yet, so it's a manual step for now.
This PR removes the workaround introduced in
https://github.com/Unleash/unleash/pull/6931. After
https://github.com/ivarconr/unleash-enterprise/pull/1268 has been
merged, this should be safe to apply.
Notably, this PR:
- tightens up the type for the enable change request function, so we can
use that to inform the code
- skips trying to do anything with an empty array
The last point is less important than it might seem because both the env
validation and the current implementation of the callback is essentially
a no-op when there are no envs. However, that's hard to enforce. If we
just exit out early, then at least we know nothing happens.
Optionally, we could do something like this instead, but I'm not sure
it's better or worse. Happy to take input.
```ts
const crEnvs = newProject.changeRequestEnvironments ?? []
await this.validateEnvironmentsExist(crEnvs.map((env) => env.name));
const changeRequestEnvironments =
await enableChangeRequestsForSpecifiedEnvironments(crEnvs,);
data.changeRequestEnvironments = changeRequestEnvironments;
```
