I've configured an active Geo replication for Azure SQL DB. I have primary and secondary database without failover policy. Also, I have an App Service which fetches data by using primary database connection string. After doing forced failover, reads are working fine most of time but my inserts/updates were failed.
My question is do I need to update my connection string(pointing to secondary database which has become primary database now due to failover) in the App Service to make insert/update work or is there any other way to make my application work without changing connection string in my app service?
Thanks In Advance!!!
P.S - I am new to Azure.
There are two types of authentication level:
server level
database level
When you are using SQL admin server user(server-level authentication) then you must update connection string manually in the application.
If you do not want to update connection string then follow this to handle this situation by creating contained DB users in the primary database, wait until the next replication takes place or manual failover, and then try, here the authentication work on database level.
For more information:https://learn.microsoft.com/en-us/sql/relational-databases/security/contained-database-users-making-your-database-portable?view=sql-server-2017
Related
We can have passive read-only asynchronous real-time sync-up for Azure SQL database, for disaster recovery.
But our requirement is to have real-time sync-up between both active read-write databases to provide low latency to customers in different locations of the world.
for example:
I'm providing e-commerce website, I will update data in one of the
database server and other connected databases in sync with this
database should get updates.
Users from different servers of the world will get connected to their
nearest data center for low latency. If someone buys something or puts
some review, it should get updated in all other databases. In this
way we need active-active database sync.
We explored multiple items on this, but did not find anything relative.
Can anyone please guide me on how to achieve this.
SQL Server has Peer-to-Peer Transactional Replication, but you need to ensure in the application that conflicting changes are not introduced on multiple nodes.
SQL Server also has Merge Replication, which allows updates at any subscriber, and supports custom conflict resolution.
These are both available on SQL Server VMs. Limited replication options are available on Azure SQL Database Managed Instance. Azure SQL Database also has Data Sync.
Azure Cosmos DB also supports Multi-Master.
In either case multi-master introduces significant cost/complexity. Often it's better to just have a single writable master with regional readable replicas. In that configuration the application needs to connect to the global master for writing, but can read from a local replica. For this pattern you can simply use Failover Groups.
Based on the Microsoft Azure Elastic Scale sample apps online I have been able to create my Shard Map Manager (SMM) and elastic pool databases in Azure. My architecture is separate database per tenant. I am using Entity Framework in my web application. I am using a byte[] hash as my Shard Key based on an alphanumeric customer name. The customer name is entered as part of customer login so I can determine the unique shard key at the time of login to be passed to the SMM.
My questions are:
1.) Since each tenant has its own database, do I still need to include the hashed customer name/shard Key in each row of the customer tables?
2.) I don't understand where the shard key information gets passed to the SMM during a call to the server. Is it within the context of the entity or does it need to be a part of the query itself? Any sample of this would be greatly appreciated!
You access the Shard Map Manager database when finding the connection string for a particular tenant. Once you have the connection string, you connect to a tenant-specific database. Inside the database you don't need to use the shard key at all.
The Elastic Database Tools library has an implementation of data dependent routing (DDR). But you might find it overkill for when you have a simple single tenant sharding pattern implementation. You can always just query the shard map database (or custom configuration store) at startup and load a Dictonary<string,string> to store the CustomerName -> ConnectionString lookup.
We have one webapp running on Azure, which pushes data to Azure Redis, then we have an on-prem component which reads that data from Azure redis and processes that.
Recently due to Azure region failure that Azure Redis went down. Web app and my on-prem component was not able to contact Azure redis.
How can I make sure zero down time for my web app to access Azure redis ?
Redis-GeoRelication doesn't solves my problem as it is unideirectional, and Manual failover. Also my web app and on-prem component need to know both redis endpoint, and contact accrondignly. which is not seemless.
Azure redis doesn't support cluster having shards in multiple region.
So my requirement is, Web-app and on-prem component both need to contain one cache/database endpoint ( without having any knowledge about the replication of the cache/database). if primary cache/db fails then, that endpoint should automatically goes to replicated cache or DB.
As per Documentation from Azure, it doesn't seem Azure Redis is correct fit for this requirment, is there any other Azure component which fits this requiremnet.
Had a look to Azure sql with failover group. As per documentation, "you can configure a grace period that controls the time between the detection of the outage and the failover itself. It is possible that traffic manager initiates the endpoint failover before the failover group triggers the failover of the database. In that case the web application cannot immediately reconnect to the database. But the reconnections will automatically succeed as soon as the database failover completes." . We can set that grace period to 1 hour (minimum) .
So it means with Azure sql also. In case of failure of one db server, my web application will not be able to write to db for atleast 1 hour, Is my understanding correct ?
Azure SQL and Azure Cosmos DB both support single endpoint and HA across regions, you might want to look into those.
Those are not caches, but they do allow for a single endpoint and failover
We have an azure web app & a db we want to replicate all over the world.
So, we use Traffic manager to redirect the User to the closest hosted Web app , and with a location setting in the web app, It knows to which database it should go against.
Now, my question is , as the mode is One database Writeable (Primary) and the replicas being read only , how do me or azure handle that at the moment of calling the database?
For example, if from my app I am going to Add a record to database, I cant use the nearest DB connection string, I need to go against the Primary one.
Should I handle this? or I will go always against the nearest one even if its read-only an azure will handle the write transferring it to the primary db ?
In the case I am the one that should manage that, then I should handle 2 connection strings, one for the primary DB writeable, and one for the closest db readable, and I should split my services , categorized by write/read actions
and following this scenario, if I have a Store procedure which WIRTES AND READS, how would I handle that?
This is a common issue when it comes to using Azure SQL in geo-replication mode. You cannot use traditional LB techniques such as Azure Traffic Manager. In this case, you should be using the retry pattern on your database connections, working from the primary down to the alternate names as required.
AFAIK, there is no easy way to tell, after connected to a database, if you are on a primary or a read-only secondary. As per this link there are some stored procs you can call to understand the topology. You can understand this using Azure PS/API, but then you would have to build that logic in to your application.
In short:
You need to handle your database connections and employ retry
patterns,etc
You should implement CQRS to separate read/write workloads from
each other if you want to take advantage of read-only secondaries
Hope that helps.
One of my customer is developing multi-tenant solution. And I'm working as developer for the automation of resource provisioning part. The solution is developed such that each tenant have their resources separate from each other.
So for example, a single tenant will require a SQL database (PAAS), A Storage Account, and also many other resources.
One of the requirement that, customer set is, he wants to have X number of databases to be hosted on a SQL server (a logical server not VM). Which I don't think is valid having been using SQL as PAAS.
So My question is, Should we create SQL Server and SQL database for each tenant?
Or
Should we create a SQL server then host X number of databases on that server. when server reaches limits (X databases), create another server and execute same logic.
In either scenario, what difference does it make from Database Performace, Pricing and Database security point of view?
FYI, My thinking is that, If I host 'X' database on a single SQL Logical Server or If I create 'X' SQL Logical Server for 'X' SQL database hosting, It won't make any difference from Pricing and Database Performace point of view.
Few differences i could think of, if you go with single server for all clients..
1.Administrator Password is per Server and using this,one client can have access to other databases as well..
2.Azure has a limit of how many DTU's can be capped under one server,so if you have many databases under one server..This may lead to few issues like
a.)frequent DTU increase requests
b.)some times automated backup may fail,if there are no DTU's available(Backup needs to copy the whole database,so in this process ,DTU's needed will be equal to database which is backed up)
Your question is too broad, as there are many opinions and approaches to your question.
But in any way you should take a look at elastic database pools: https://azure.microsoft.com/en-us/documentation/articles/sql-database-elastic-pool/ which is a feature exactly designed for multi-tenant SaaS solutions.
Your end solution may be a combination of both - you may want to use a single server to "bigger" tenants, while you can host multiple small tenants together in a single server.
Security shall not be a factor with big weight because, when you use database contained credentials for application access, it does not really matter whether the databases are allocated in single logical server or not.