Using only EXTERNAL_HOST results in:
```
CRITICAL: Error subscribing to Zulips: Account is not associated with this subdomain
```
Since system bots like the nagios bot are created in
`SYSTEM_BOT_REALM`, default to that subdomain of `EXTERNAL_HOST`.
In #23380, we are changing all occurrences of uri with url in order to
follow the latest URL standard. Previous PRs #25038 and #25045 has
replaced the occurences of uri that has no direct relation with realm.
This commit changes just the model property, which has no API
compatibility concerns.
This commit updates code to access name from named_user_group
field which points to the "NamedUserGroup" instead of directly
accessing name from "UserGroup", since name field will only
be present on NamedUserGroup objects in further commits.
Earlier, we didn't soft-reactivate users for group mentions
at all because it wasn't easy to calculate group size.
Now, we will soft reactivate if the user group mentions has
less than 12 members.
We don't reactivate all users because a user group can have a
very large size, which can lead to large backlogs in the
deferred-work queue.
Fixes part of #27586.
Factor out the repeated pattern of taking a lock, or immediately
aborting with a message if it cannot be acquired. The exit code in
that situation is changed to be exit code 1, rather than the successful
0; we are likely missing new work since that process started.
We move the lockfiles to a common directory under `/srv/zulip-locks`
rather than muddy up `/home/zulip/deployments`.
Fixes an import cycle that breaks mypy inference with django-stubs:
zproject.settings → zproject.configured_settings →
zproject.default_settings → zerver.models.users → django.conf →
zproject.settings
(The Django settings system is really not an appropriate place to put
functions.)
Signed-off-by: Anders Kaseorg <anders@zulip.com>
This middleware was highly-specific to a set of URLs, and pulled in a
beautifulsoup dependency for Tornado. Move it closer to where it is
used, minimizing action at a distance, as well as trimming out a
dependency.
Fixes#28621
Till now, this was actually a flat 50GB despite what the /plans/ page
says and was adjusted flexibly when somebody asked for a higher limit.
This actually implements the advertised formula, but changing it to
5GB/user since that's a more reasonable limit.
Keeps the 50GB limit for sponsored Standard Free organizations and also
places it as the floor for the quota for paid orgs, to not lower this
for tiny orgs with less than 5 users.
When an organization (without open ability for anyone to join) invites a
guest user, the invitation prompts allows them to choose whether the
guest should be added to default streams or not. This is useful, because
since we don't have per-role default streams configs, they may want
default streams to be for full Members.
SCIM provisioning doesn't have this control, since a newly provisioned
user gets created via a direct do_create_user call, thus adding them to
the organization's default streams, with no workaround possible aside of
just getting rid of default streams in the organization.
To make provisioning guests in such an organization usable, we add a
simple config option to create them with no streams. It's configured by
adding
```
"create_guests_without_streams": True
```
to the config dict in settings.SCIM_CONFIG.
SHA1PasswordHasher will be removed in Django 5.1. MD5PasswordHasher
will remain for exactly this purpose of speeding up tests.
Use MD5PasswordHasher by default, but leave SHA1PasswordHasher in the
list for compatibility with test databases that have already been
generated. Once some other change forces test databases to be
rebuilt, we can remove SHA1PasswordHasher.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
Previously, users were allowed to signup or change their names to
those which already existed in the realm.
This commit adds an Organization Permission, that shall enforce
users to use unique names while signing up or changing their
names. If a same or normalized full name is found in realm,
then a validation error is thrown.
Fixes#7830.
This was only used in the undocumented narrow_stream mode, and relied
on a deprecated synchronous XHR request.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
Replace a separate call to subprocess, starting `node` from scratch,
with an optional standalone node Express service which performs the
rendering. In benchmarking, this reduces the overhead of a KaTeX call
from 120ms to 2.8ms. This is notable because enough calls to KaTeX in
a single message would previously time out the whole message
rendering.
The service is optional because he majority of deployments do not use
enough LaTeX to merit the additional memory usage (60Mb).
Fixes: #17425.
Just shows a config error page if the bouncer is not enabled. Uses a new
endpoint for this so that it can work nicely for both browser and
desktop app clients.
It's necessary, because the desktop app expects to get a json response
with either an error or billing_access_url to redirect to. Showing a
nice config error page can't be done via the json error mechanism, so
instead we just serve a redirect to the new error page, which the app
will open in the browser in a new window or tab.
Only affects zulipchat, by being based on the BILLING_ENABLED setting.
The restricted backends in this commit are
- AzureAD - restricted to Standard plan
- SAML - restricted to Plus plan, although it was already practically
restricted due to requiring server-side configuration to be done by us
This restriction is placed upon **enabling** a backend - so
organizations that already have a backend enabled, will continue to be
able to use it. This allows us to make exceptions and enable a backend
for an org manually via the shell, and to grandfather organizations into
keeping the backend they have been relying on.
By convention, all our backends inherit form ZulipAuthMixin. This is
helpful to have a single place if we want to add some class attributes
that all the backends should have.
Adds a re-usable lockfile_nonblocking helper to context_managers.
Relying on naive `os.mkdir` is not enough especially now that the
successful operation of this command is necessary for push notifications
to work for many servers.
We can't use `lockfile` context manager from
`zerver.lib.context_managers`, because we want the custom behavior of
failing if the lock can't be acquired, instead of waiting.
That's because if an instance of this gets stuck, we don't want to start
queueing up more processes waiting forever whenever the cronjob runs
again and fail->exit is preferrable instead.
When you click "Plan management", the desktop app opens
/self-hosted-billing/ in your browser immediately. So that works badly
if you're already logged into another account in the browser, since that
session will be used and it may be for a different user account than in
the desktop app, causing unintended behavior.
The solution is to replace the on click behavior for "Plan management"
in the desktop app case, to instead make a request to a new endpoint
/json/self-hosted-billing, which provides the billing access url in a
json response. The desktop app takes that URL and window.open()s it (in
the browser). And so a remote billing session for the intended user will
be obtained.
We already override PUSH_NOTIFICATION_BOUNCER_URL in
test_extra_settings.py, so making this change should have as its only
impact making it a bit easier to test the push notifications bouncer
manually in a development environment.
I submitted a related PR to the mobile app documentation for testing
the push notifications software against a production server motivated
by this.
See https://chat.zulip.org/#narrow/stream/6-frontend/topic/typing.20notifications.20efficiency/near/1664991.
As detailed in that discussion,
`TYPING_STARTED_EXPIRY_PERIOD_MILLISECONDS` and
`TYPING_STARTED_WAIT_PERIOD_MILLISECONDS` are coupled constants, and
the impact of them being large is mainly that if a user closes their
computer or loses network in the middle of typing something (not
exactly a common occasion), then the client will suggest they kept on
typing longer than they in fact did.
There's a substantial decrease in resources consumed by this feature
associated with raising `TYPING_STARTED_WAIT_PERIOD_MILLISECONDS`, so
that at least seems worth doing.
Meanwhile, because TYPING_STOPPED_WAIT_PERIOD_MILLISECONDS measures
how long we should wait before deciding to stop suggesting a user is
still typing if they were previously typing a message but paused doing
so without closing the compose box (example causes being stepping away
from the computer, tabbing to go look something up, or just thinking
for a bit).
On the one hand, even the original 5 seconds is a fairly long time to
pause to think without touching the keyboard; on the other hand,
sitting with text you've written in the compose box is likely still a
quite high intent-to-send-soon state. Increasing this to 12 seconds
seems like a reasonable balance between being too trigger-happy here
here and avoiding someone who left their computer appearing like they
are still typing for a long time afterwards.
This commit adds a new endpoint 'users/me/onboarding_steps'
deprecating the older 'users/me/hotspots' to mark hotspot as read.
We also renamed the view `mark_hotspot_as_read` to
`mark_onboarding_step_as_read`.
Reason: Our plan is to make this endpoint flexible to support
other types of UI elements not just restricted to hotspots.
Currently, the sender names for outgoing emails sent by Zulip
are hardcoded. It should be configurable for self-hosted systems.
This commit makes the 'Zulip' part a variable in the following
email sender names: 'Zulip Account Security', 'Zulip Digest',
and 'Zulip Notifications' by introducing a settings variable
'SERVICE_NAME' with the default value as f"{EXTERNAL_HOST} Zulip".
Fixes: #23857
This makes it possible for a self-hosted realm administrator to
directly access a logged-page on the push notifications bouncer
service, enabling billing, support contacts, and other administrator
for enterprise customers to be managed without manual setup.
The presence and user status update events are only sent to accessible
users, i.e. guests do not receive presence and user status updates for
users they cannot access.
The original behavior of this setting was to disable LDAP
authentication for any realms not configured to use it. This was an
arbitrary choice, and its only value was to potentially help catch
typos for users who are lazy about testing their configuration.
Since it makes it a very inconvenient to potentially host multiple
organizations with different LDAP configurations, remove that
behavior.
This makes it possible to send notifications to more than one app ID
from the same server: for example, the main Zulip mobile app and the
new Flutter-based app, which has a separate app ID for use through its
beta period so that it can be installed alongside the existing app.
This commit adds new API endpoint to get stream email which is
used by the web-app as well to get the email when a user tries
to open the stream email modal.
The stream email is returned only to the users who have access
to it. Specifically for private streams only subscribed users
have access to its email. And for public streams, all non-guest
users and only subscribed guests have access to its email.
All users can access email of web-public streams.
In 0b3f7a5a6 we started automatically setting this in dev if we
found a cert file in an appropriate place. But for running tests,
we don't want such variability. Set it back to None there.
Originally, this was how the notification emails worked, but that was changed
in 797a7ef97b, with this old behavior
available as an option.
The footer and from address of emails that are sent when this
setting is set to True are confusing, especially when more people
are involved in a stream and since we have changed the way we send
emails, it should be removed. It’s also not widely used.
Fixes#26609.
This is an exception that we should be generally catching like the
others, which will give our standard /login/ redirect and proper logging
- as opposed to a 500 if we don't catch.
Addresses directly a bug we occurred in the wild, where a SAMLResponse
was submitted without issuers specified in a valid way, causing this
exception. The added test tests this specific type of scenario.
This endpoint verifies that the services that Zulip needs to function
are running, and Django can talk to them. It is designed to be used
as a readiness probe[^1] for Zulip, either by Kubernetes, or some other
reverse-proxy load-balancer in front of Zulip. Because of this, it
limits access to only localhost and the IP addresses of configured
reverse proxies.
Tests are limited because we cannot stop running services (which would
impact other concurrent tests) and there would be extremely limited
utility to mocking the very specific methods we're calling to raising
the exceptions that we're looking for.
[^1]: https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/
Rewrite the test so that we don't have a dedicated URL for testing.
dev_update_subgroups is called directly from the tests without using the
test client.
**Background**
User groups are expected to comply with the DAG constraint for the
many-to-many inter-group membership. The check for this constraint has
to be performed recursively so that we can find all direct and indirect
subgroups of the user group to be added.
This kind of check is vulnerable to phantom reads which is possible at
the default read committed isolation level because we cannot guarantee
that the check is still valid when we are adding the subgroups to the
user group.
**Solution**
To avoid having another transaction concurrently update one of the
to-be-subgroup after the recursive check is done, and before the subgroup
is added, we use SELECT FOR UPDATE to lock the user group rows.
The lock needs to be acquired before a group membership change is about
to occur before any check has been conducted.
Suppose that we are adding subgroup B to supergroup A, the locking protocol
is specified as follows:
1. Acquire a lock for B and all its direct and indirect subgroups.
2. Acquire a lock for A.
For the removal of user groups, we acquire a lock for the user group to
be removed with all its direct and indirect subgroups. This is the special
case A=B, which is still complaint with the protocol.
**Error handling**
We currently rely on Postgres' deadlock detection to abort transactions
and show an error for the users. In the future, we might need some
recovery mechanism or at least better error handling.
**Notes**
An important note is that we need to reuse the recursive CTE query that
finds the direct and indirect subgroups when applying the lock on the
rows. And the lock needs to be acquired the same way for the addition and
removal of direct subgroups.
User membership change (as opposed to user group membership) is not
affected. Read-only queries aren't either. The locks only protect
critical regions where the user group dependency graph might violate
the DAG constraint, where users are not participating.
**Testing**
We implement a transaction test case targeting some typical scenarios
when an internal server error is expected to happen (this means that the
user group view makes the correct decision to abort the transaction when
something goes wrong with locks).
To achieve this, we add a development view intended only for unit tests.
It has a global BARRIER that can be shared across threads, so that we
can synchronize them to consistently reproduce certain potential race
conditions prevented by the database locks.
The transaction test case lanuches pairs of threads initiating possibly
conflicting requests at the same time. The tests are set up such that exactly N
of them are expected to succeed with a certain error message (while we don't
know each one).
**Security notes**
get_recursive_subgroups_for_groups will no longer fetch user groups from
other realms. As a result, trying to add/remove a subgroup from another
realm results in a UserGroup not found error response.
We also implement subgroup-specific checks in has_user_group_access to
keep permission managing in a single place. Do note that the API
currently don't have a way to violate that check because we are only
checking the realm ID now.
Adds typing notification constants to the response given by
`POST /register`. Until now, these were hardcoded by clients
based on the documentation for implementing typing notifications
in the main endpoint description for `api/set-typing-status`.
This change also reflects updating the web-app frontend code
to use the new constants from the register response.
Co-authored-by: Samuel Kabuya <samuel.mwangikabuya@kibo.school>
Co-authored-by: Wilhelmina Asante <wilhelmina.asante@kibo.school>
This adds API support to reorder linkifiers and makes sure that the
returned lists of linkifiers from `GET /events`, `POST /register`, and
`GET /realm/linkifiers` are always sorted with the order that they
should processed when rendering linkifiers.
We set the new `order` field to the ID with the migration. This
preserves the order of the existing linkifiers.
New linkifiers added will always be ordered the last. When reordering,
the `order` field of all linkifiers in the same realm is updated, in
a manner similar to how we implement ordering for
`custom_profile_fields`.
Restore the default django.utils.log.AdminEmailHandler when
ERROR_REPORTING is enabled. Those with more sophisticated needs can
turn it off and use Sentry or a Sentry-compatible system.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
Uploads are well-positioned to use S3's "intelligent tiering" storage
class. Add a setting to let uploaded files to declare their desired
storage class at upload time, and document how to move existing files
to the same storage class.
Pass the HttpRequest explicitly through the two webhooks that log to
the webhook loggers.
get_current_request is now unused, so remove it (in the same commit
for test coverage reasons).
Signed-off-by: Anders Kaseorg <anders@zulip.com>
Combine nginx and Django middlware to stop putting misleading warnings
about `CSRF_TRUSTED_ORIGINS` when the issue is untrusted proxies.
This attempts to, in the error logs, diagnose and suggest next steps
to fix common proxy misconfigurations.
See also #24599 and zulip/docker-zulip#403.
Having exactly 17 or 18 middlewares, on Python 3.11.0 and above,
causes python to segfault when running tests with coverage; see
https://github.com/python/cpython/issues/106092
Work around this by adding one or two no-op middlewares if we would
hit those unlucky numbers. We only add them in testing, since
coverage is a requirement to trigger it, and there is no reason to
burden production with additional wrapping.
Alex Vandiver
Vector73
Tim Abbott
Sahil Batra
Prakhar Pratyush
Anders Kaseorg
Mateusz Mandera
roanster007
Aman Agrawal
Prakhar Pratyush
Greg Price
Esther Anierobi
Tam Le
Simon Michalke
Zixuan James Li
Samuel
Lauryn Menard