We are currently using "ioredis" adapter for nodejs.
Some of our logic is calling a none clustered redis, we would like to scale our calls using "scale reads" attribute that sends commands to the "replicas" to reduce load from the primary node.
The problem is that the most common command we use and we would like to scale is "zrevrange", and it's simply not calling the replicas.
ioredis version: 2.5.0,
the redis we are using is on AWS ElastiCache engine version: 3.2.4
IORedis do not support scale reads on non clustered redis.
Related
One of my backend service uses Ignite. The backend service itself is stateless, meaning, the service doesn't have any internal or shared states, and can scale up and down on the need. The backend service is deployed in Kubernetes.
I have used the Ignite, now, in the embedded way, with cache mode set to REPLICATED, and native persistence enabled. Additionally I have also enabled the baselineAutoAdjustEnabled. The reason for using replicated mode is to have the data available to all the backend instances locally & consistent way.
The application seems to work correctly -
as and when data is modifies the data gets replicated across all the instances of the backend
data replication works even if a new instance joins the topology after (long time) the initial baseline is set & activated.
The question now is, is this is the right approach in running the Ignite in embedded (server mode), with full replication, and native persistence?
In general your design looks well. But be careful with using of baselineAutoAdjustEnabled. This feature works only in case of topology changes do not leads to lost partitions. But in your case, as you use only replicated caches, any node stoping should not lead to lost partitions.
Embeded mode in this case does not have any matter.
Can anyone explain to me what is aware client in Redis Enterprise ?
I found this post: Redis Enterprise Clustering Command Error 'CLUSTER'
I try to use Redis Enterprise Cluster with docker.
I create 3 docker redis nodes with two shards for better scalability.
So what exactly is that aware client and what is the difference between non-cluster aware ?
Also, what is regular OSS cluster ?
Thank You..
"Cluster Aware" means a Redis client that supports the OSS Cluster API (https://redis.io/topics/cluster-spec). For example, the Ruby client https://github.com/redis/redis-rb#cluster-support supports it.
A non-aware client is a client that only supports connecting to Redis in single-instance mode (and perhaps Sentinel), such as the Python client https://github.com/andymccurdy/redis-py.
The Enterprise Cluster can be used by both types of clients regardless of how the database is deployed (i.e. clustered or not).
To clear some more of the confusion:
OSS Cluster - a mode of deployment and API (i.e. not single-instance)
Enterprise Cluster - a product
I'm using Hortonworks schema registry with NIFI and things are working fine. I have installed Hortonworks schema registry on a single node and I'm afraid if that machine goes down what will happen to my NIFI flows. I have seen in Hortonworks schema registry architecture that we can use mysql, PostgreSql and In-Memory storage for storing schema. AFAIK none of them are distributed system. Is there any way to achieve cluster mode for high availability?
Sure, you can do active-active or active-passive replication for MySQL and Postgres, but that is left up to you to implement, as Hortonworks will likely forward you to the respective documentation on each tool, and that is the reason why the documentation for these tools doesn't guide you towards these design decisions in itself, as you should be aware of the drawbacks when having a SPoF
The Schema Registry itself is just a web-app, so you could put it behing your favorite reverse proxy, or within a container orchestrator, such as Docker support in HDP 3.x
I was wondering why do we need Redis server for running CKAN.
If needed, why? And How do I configure it with CKAN?
p.s
I am running my ckan instance in RHEL7.
Update: Redis has been a requirement since CKAN 2.7, when a new system for asynchronous background jobs was introduced which relies on Redis. You can configure the Redis connection using the ckan.redis.url option.
Redis is not required for the current version of CKAN (2.6.2 at the time of this writing), it's not even mentioned in the CKAN 2.6.2 documentation.
However, the upcoming 2.7 release will require Redis for its new system of asynchronous background jobs. Redis will be configured using the new ckan.redis.url option.
I just started evaluating Redis. I am using Redis 2.8.19 which the most latest stable release. Redis 2.9 is still unstable and Redis 3.0 is just available for developer's preview (not recommended for production). I was tryin to setus a cluster of Redis and when I changed my redis.conf and appended
cluster-enabled yes
cluster-config-file nodes.conf
cluster-node-timeout 5000
and started my Redis server by
src/redis-server ./redis.conf
it gave me an error as follows
* FATAL CONFIG FILE ERROR *
Reading the configuration file, at line 2
'cluster-enabled yes'
Bad directive or wrong number of arguments
I googled the error and got to know that my version (2.8.19) does not support cluster. I was still unable to fine any such specification in Redis Docs. My question is simple. Does Redis 2.8.19 supports redis cluster configuration? Or I have to upgrade to Redis 2.9 or Redis 3.0. I am evaluating Redis because I need to deploy it in production. Please guide.
Redis Cluster support is only for versions >= 3.0.0. Redis 3.0.0 will be released as a stable version in a matter of days, it's a good idea to use it if you want to use Cluster. The cluster support is considered to be stable, however for it to be considered mature we want to see adoption. Btw there is already at least a very large site using it in production. Currently the most sane thing to do if you need Redis Cluster is to test it for your use case, and if it looks great, use it.
Redis cluster is supported only in Redis 3.0+ (which is now stable). I have written a simple API called "Simple Redis Cluster Client" which can be used in redis's sub 3.0 versions for running in a cluster like mode (Not precisely a cluster, it just distributes keys among redis nodes based on the key's hashcode, You can have a look # https://github.com/prash-mi/simple-redis-cluster-client
Cluster support for Redis is only from v3 - v2.8.19 doesn't do clustering.