Atlas App Services
23 results found
-
Add retry policy to Atlas Trigger
A failed Atlas trigger gets suspended and will not retry without manual intervention. Some transient conditions can be retried and it would be useful to configure a retry policy with a trigger (at the Realm Function level). Retry X times with Y back-off seconds between retries, then fail (plus notification policy).
31 votes -
Notification on function failure
I would like a notification if a trigger fires a function, and the function fails.
I have a scheduled trigger that calls a function every day. I have no idea if the function fails unless I manually log in and check the status. I would like to be alerted in the event of a failure. I've gotten around this by manually setting up Slack notifications.
20 votes -
Triggers Integration with Azure Event Grid
We have “Event bridge” integration with Atlas Triggers today. Adding Azure Event Grid support will be very helpful in building no-code solutions like real time sync for ODS/ ODL etc for Azure.
8 votes -
Allow manually disabled trigger to resume without resume token
I am filing a feature request for the option to: Disable a trigger manually and be able to restart it without using a resume token (i.e. such that DB events that occurred while it was disabled will be ignored by the trigger, and it will simply start listening for new events upon restart).
8 votes -
Allow to pass user on Realm's Database Triggers
Currently I can create a Database Trigger, that when calling its Function I can restrict based on "%%user" - through "Can Evaluate". Though since the Function is called by a Trigger and therefore runs as a "system user" I cannot access the User object (or its details) within the Function - I (personally) mainly want the userID & email - though whole object would be better.
Since a Function is called by a Trigger and therefore it runs as "System User" - Please allows us to pass an array of arguments for the Function. This could be a field after…
7 votes -
Realm Trigger Event Projection should support exclusion
Realm triggers with a projections must include "operationType": 1. This makes excluding fields impossible because, according to the documentation:
if you exclude fields, you cannot also specify the inclusion of fields
The only work around we have found is to list every single field in the database and this would require an update anytime we add another field to the model. This is quite cumbersome and should be easily resolved with a simple exclusion.
7 votes -
Trigger event queue metrics
We recently ran into the problem that our trigger was suspended because the resume token was no longer present in the oplog.
The problem was a combination of queued up trigger events and a short oplog window and the suggested solution was to increase the triggers performance by parallelizing it and by increasing the oplog size to get a bigger oplog window.Unfortunately, we're not able to verify the real success of these suggestions, because even though we can monitor the oplog window in the metrics, there are no metrics for the trigger queue itself. As a result, we can't…
5 votes -
extend the requirements applied to resource tags.
We want to extend the requirements applied to resource tags. Right now we have a specific project whose requirements do not allow us to use tag values
4 votes -
Configure Realm notifications
Currently only all project owners of the Atlas project in where your Realm app situates will receive notification of suspended trigger. This currently cannot be customized. We want to have the possibility to configure the notifications to include additional emails, not only for project owners.
4 votes -
Add Trigger call for failed/unsuccessful login
We can have a trigger that fires based on a wrong password. For examples, as an user If I enter wrong password then it should be handled using some trigger.
I have already opened a case on mongodb support :
3 votes -
Add the ability to use a value to reference a database in a trigger
It would be great if we could use values for the database reference in a trigger, instead of having to select a database from a drop-down.
An actual use case:
To be able to use Realm for our various staging areas, we have created multiple Realm apps. One for production, one for UAT, one for test, etc. Each of them pointing to the databases for their respective staging environment.
We are using Github deployments, and we use a separate branch for each of the staging environments. So when something gets committed to the "deploy-test" branch, it gets to be deployed…
3 votes -
Support Multiple CRON Schedules Per Trigger
It's not uncommon to want different trigger schedules to run on different days of the week, notably weekdays and weekends. The current solution is to create separate triggers that call the same function, but ideally, a single trigger could support multiple CRON schedules.
2 votes -
Triggers on DB level rather than single collections only
Change streams support events on DB level or single collections, but triggers are limited to single collections. Which means that if we wanted to listen on changes in a generic way, we would have to setup a trigger for each collection and set them up with AWS EventBridge (and think about new triggers when new collections are generated).
Given that the underlying tech (change streams) supports DB-level subscriptions, we would really like to see that on the UI, too.
2 votes -
Triggers - resume via CLI
Currently the only way to resume a suspended Trigger is via the Admin API:
https://realm.mongodb.com/api/admin/v3.0/groups/{groupId}/apps/{appId}/triggers/{triggerId}/resumeIt would be more convenient via the Atlas CLI
1 vote -
Show the last modified date or allow to add a trigger version in a text field
It would be very usful to show the date of the last trigger code edition or better to save a trigger version field.
1 vote -
Migration from Postman
I have a scheduled run in Postman that includes calling an external API and inserting a response into MongoDB Atlas using Data API. It would be nice if I could import my Postman collection of requests, and it would generate a function for a Trigger.
1 vote -
Failover Support
Per Support, Triggers may not process if the App Services app's deployment region is experiencing wide-scale issues. Currently triggers will execute in the single region of their deployment.
For multi-region clusters, triggers should be able to failover to another region and continue processing.
1 vote -
Disable/Enable all triggers at once
It would be very useful to be able to disable/enable all triggers of realm application at once. It useful for instance when you restore data from backup or make some big changes, and it is not needed to run triggers on every change
1 vote -
changeEvent cluster name
I want to be able to programmatically access the clusterName from change event or global variable for example:
changeEvent.ns.cluster = <CLUSTER/SERVICE name>
global.CLUSTER_NAMEI need this when I want to perform an update on same collection/document after the trigger, but i don't want to hardcode the clusterName - because i have multiple environments - each has its own cluster (dev/sandbox/prod etc...)
1 vote -
One-time scheduled triggers
At the moment, MongoDb offers the ability to schedule triggers periodically using Cron Jobs. It would be really handy if a user could schedule a trigger to run only once and execute a function as my use-case requires.
1 vote
- Don't see your idea?