I have a MongoDB database with 2 collections:
groups: { group_slug, members }
users: { id, display name, groups }
All changes to groups are done by changing the members array of the group to include the users ids.
I want to sync these changes across to the users collection by using map/reduce. How can I output the results of map/reduce into an existing collection (but not merging or reducing).
My existing code is here: https://gist.github.com/morgante/5430907
How can I output the results of map/reduce into an existing collection
You really can't do it this way. Nor is this really suggested behaviour. There are other solutions:
Solution #1:
Output the map / reduce into a temporary collection
Run a follow-up task that updates the primary data store from the temporary collection
Clean-up the temporary collection
Honestly, this is a safe way to do this. You can implement some basic retry logic in the whole loop.
Solution #2:
Put the change on a Queue. (i.e. "user subscribes to group")
Update both tables from separates workers that are listening for such events on the queue.
This solution may require a separate piece (the queue), but any large system is going to have such denormalization problems. So this will not be the only place you see this.
Related
I need to either add new document with children or add child to already existing parent.
The only way I know how to do it is ugly:
public async Task AddOrUpdateCase(params)
{
try
{
await UpdateCase(params);
}
catch (RequestFailedException ex)
{
if (ex.Status != (int)HttpStatusCode.NotFound)
throw;
await AddCase(params);
}
}
private async Task UpdateCase(params)
{
// this line throws when document is not found
var caseResponse = await searchClient.GetDocumentAsync<Case>(params.CaseId)
// need to add to existing collection
caseResponse.Value.Children.Add(params.child);
}
I think there wouldn't be any problem if this document didn't contain collection. You cannot use MergeOrUpload if there are child collections. You need to load them from index and add element.
Is there better way to do it?
Azure Cognitive Search doesn't support partial updates to collection fields, so retrieving the entire document, modifying the relevant collection field, and sending the document back to the index is the only way to accomplish this.
The only improvement I would suggest to the code you've shown is to search for the documents you want to update instead of retrieving them one-by-one. That way, you can update them in batches. Index updates are much more expensive than queries, so to reduce overhead you should batch updates together wherever possible.
Note that if you have all the data needed to re-construct the entire document at indexing time, you can skip the step of retrieving the document first, which would be a big improvement. Azure Cognitive Search doesn't yet support concurrency control for updating documents in the index, so you're better off having a single process writing to the index anyway. This should hopefully eliminate the need to read the documents before updating and writing them back. This is assuming you're not using the search index as your primary store, which you really should avoid.
If you need to add or update items in complex collections often, it's probably a sign that you need a different data model for your index. Complex collections have limitations (see "Maximum elements across all complex collections per document") that make them impractical for scenarios where the cardinality of the parent-to-child relationship is high. For situations like this, it's better to have a secondary index that includes the "child" entities as top-level documents instead of elements of a complex collection. That has benefits for incremental updates, but also for storage utilization and some types of queries.
I have an application that most of the collections in it are heavily read then write, so I demoralized the data in them, and now I need to handle the normalization of the data, for some collections I used jobs in order to sync the data but that not good enough as for some cases I need the data to be normalized in real-time,
for example:
let's say I have orders collections and users collection.
orders have the user email(for search)
{
_id:ObjectId(),
user_email:'test#email.email'
....
}
now whenever I am changing the user email in users I want to change it in orders as well.
so I find that MongoDB has change stream which looks pretty awesome feature, I have played with it a bit and it gives me the results I need to update my other collections, my question is does anyone use it in production? can I trust on this stream to be always set the update data to update the other collections? how does it affect the DB performance if I have many streams open? also, I use the nodejs MongoDB driver does it has any effect
I've not worked yet with change stream but these cases are very common and can be easily solved by building more normalized schema
Normalization form 1 says among the others "don't repeat data" - so you will save the email in the users collection only
orders collection won't have the email field but will have user_id for joining with users collection with lookup command for joining collections
https://docs.mongodb.com/manual/reference/operator/aggregation/lookup/
I'm using a DDD/CQRS/ES approach and I have some questions about modeling my aggregate(s) and queries. As an example consider the following scenario:
A User can create a WorkItem, change its title and associate other users to it. A WorkItem has participants (associated users) and a participant can add Actions to a WorkItem. Participants can execute Actions.
Let's just assume that Users are already created and I only need userIds.
I have the following WorkItem commands:
CreateWorkItem
ChangeTitle
AddParticipant
AddAction
ExecuteAction
These commands must be idempotent, so I cant add twice the same user or action.
And the following query:
WorkItemDetails (all info for a work item)
Queries are updated by handlers that handle domain events raised by WorkItem aggregate(s) (after they're persisted in the EventStore). All these events contain the WorkItemId. I would like to be able to rebuild the queries on the fly, if needed, by loading all the relevant events and processing them in sequence. This is because my users usually won't access WorkItems created one year ago, so I don't need to have these queries processed. So when I fetch a query that doesn't exist, I could rebuild it and store it in a key/value store with a TTL.
Domain events have an aggregateId (used as the event streamId and shard key) and a sequenceId (used as the eventId within an event stream).
So my first attempt was to create a large Aggregate called WorkItem that had a collection of participants and a collection of actions. Participant and Actions are entities that live only within a WorkItem. A participant references a userId and an action references a participantId. They can have more information, but it's not relevant for this exercise. With this solution my large WorkItem aggregate can ensure that the commands are idempotent because I can validate that I don't add duplicate participants or actions, and if I want to rebuild the WorkItemDetails query, I just load/process all the events for a given WorkItemId.
This works fine because since I only have one aggregate, the WorkItemId can be the aggregateId, so when I rebuild the query I just load all events for a given WorkItemId.
However, this solution has the performance issues of a large Aggregate (why load all participants and actions to process a ChangeTitle command?).
So my next attempt is to have different aggregates, all with the same WorkItemId as a property but only the WorkItem aggregate has it as an aggregateId. This fixes the performance issues, I can update the query because all events contain the WorkItemId but now my problem is that I can't rebuild it from scratch because I don't know the aggregateIds for the other aggregates, so I can't load their event streams and process them. They have a WorkItemId property but that's not their real aggregateId. Also I can't guarantee that I process events sequentially, because each aggregate will have its own event stream, but I'm not sure if that's a real problem.
Another solution I can think of is to have a dedicated event stream to consolidate all WorkItem events raised by the multiple aggregates. So I could have event handlers that simply append the events fired by the Participant and Actions to an event stream whose id would be something like "{workItemId}:allevents". This would be used only to rebuild the WorkItemDetails query. This sounds like an hack.. basically I'm creating an "aggregate" that has no business operations.
What other solutions do I have? Is it uncommon to rebuild queries on the fly? Can it be done when events for multiple aggregates (multiple event streams) are used to build the same query? I've searched for this scenario and haven't found anything useful. I feel like I'm missing something that should be very obvious, but I haven't figured what.
Any help on this is very much appreciated.
Thanks
I don't think you should design your aggregates with querying concerns in mind. The Read side is here for that.
On the domain side, focus on consistency concerns (how small can the aggregate be and the domain still remain consistent in a single transaction), concurrency (how big can it be and not suffer concurrent access problems / race conditions ?) and performance (would we load thousands of objects in memory just to perform a simple command ? -- exactly what you were asking).
I don't see anything wrong with on-demand read models. It's basically the same as reading from a live stream, except you re-create the stream when you need it. However, this might be quite a lot of work for not an extraordinary gain, because most of the time, entities are queried just after they are modified. If on-demand becomes "basically every time the entity changes", you might as well subscribe to live changes. As for "old" views, the definition of "old" is that they are not modified any more, so they don't need to be recalculated anyways, regardless of if you have an on-demand or continuous system.
If you go the multiple small aggregates route and your Read Model needs information from several sources to update itself, you have a couple of options :
Enrich emitted events with additional data
Read from multiple event streams and consolidate their data to build the read model. No magic here, the Read side needs to know which aggregates are involved in a particular projection. You could also query other Read Models if you know they are up-to-date and will give you just the data you need.
See CQRS events do not contain details needed for updating read model
So I've been trying to wrap my head around this one for weeks, but I just can't seem to figure it out. So MongoDB isn't equipped to deal with rollbacks as we typically understand them (i.e. when a client adds information to the database, like a username for example, but quits in the middle of the registration process. Now the DB is left with some "hanging" information that isn't assocaited with anything. How can MongoDb handle that? Or if no one can answer that question, maybe they can point me to a source/example that can? Thanks.
MongoDB does not support transactions, you can't perform atomic multistatement transactions to ensure consistency. You can only perform an atomic operation on a single collection at a time. When dealing with NoSQL databases you need to validate your data as much as you can, they seldom complain about something. There are some workarounds or patterns to achieve SQL like transactions. For example, in your case, you can store user's information in a temporary collection, check data validity, and store it to user's collection afterwards.
This should be straight forwards, but things get more complicated when we deal with multiple documents. In this case, you need create a designated collection for transactions. For instance,
transaction collection
{
id: ..,
state : "new_transaction",
value1 : values From document_1 before updating document_1,
value2 : values From document_2 before updating document_2
}
// update document 1
// update document 2
Ooohh!! something went wrong while updating document 1 or 2? No worries, we can still restore the old values from the transaction collection.
This pattern is known as compensation to mimic the transactional behavior of SQL.
I'd like to swap out all documents for a specific index's type. I'm thinking about this like a database transaction, where I'd:
Delete all documents inside of the type
Create new documents
Commit
It appears that this is possible with ElasticSearch's bulk API, but is there a more direct way?
Based on the following statement, from the elasticsearch Delete by Query API Documentation:
Note, delete by query bypasses versioning support. Also, it is not recommended to delete "large chunks of the data in an index", many times, it’s better to simply reindex into a new index.
You might want to reconsider removing entire types and recreating them from the same index. As this statement suggests, it is better to simply reindex. In fact I have a scenario where we have an index of manufacturer products and when a manufacturer sends an updated list of products, we load the new data into our persistent store and then completely rebuild the entire index. I have implemented the use of Index Aliases to allow for masking the actual index being used. When products changes occur a process is started to rebuild the new index in the background (a process that currently takes about 15 minutes) and then switch the alias to the new index once the data load is complete and delete the old index. So this is completely seamless and does not cause any downtime for our users.