วีดีโอ

definitely yes..

docs.aws.amazon.com/dms/latest/sbs/chap-mongodb2documentdb.02.html

www.geeksforgeeks.org/how-to-install-mongodb-on-aws-ec2-instance/

There are two version, community and enterprise. Yes you can do with either of the version.

Dear, sorry for the speed.. the target was to cover it within 10 min. next time, gonna try to go slow. thanks for the feedback.

The client is what ever you want.. lets say.. you have an web serber might be iis or nginx etc.. where you application log dumped in to the mongodb.. so you client is the web server.. its a database server.. so think of anything that could uaed the db as a container will be the client.. and it could be on vm on docker on physical machine... erc.

noted..and thanks.

Usage of LDAP. Usage. .. would you please pick some specific topic so i can try to make vdo on those..

Yes.. definitely you can. And it is very common. You need to have seperate co fig for each instance. Its like.. you are running three instance of redis on a single server and the only change is in the ip and port..

Can you please let me know more about the scenario.. master to master u have two master.. you just want to sync right..?

The enpoint should be directly towards haproxy ip and specific port for redis listener. The switching will be handled by the proxy server.

@@mroyon hi, is this correct? default redis port used. redis1-192.168.1.2 redis2-192.168.1.3 haproxy-192.168.1.4 my app should connect to 192.168.1.4:6379?

@@driggsalonzo9409 yes.. if you take a look at the haproxy config of my vdo.. frontend front_redis bind 192.168.1.125:6379 name redis The front end listener is the endpoint for your app.. and the backend Server redis-a 192.168.1.121:6379 check inter 1s server redis-b 192.168.1.122:6379 check inter 1s server redis-c 192.168.1.123:6379 check inter 1s From your app the u should point to the proxy with port so it forward it to ur actual redis server.

@@mroyon Thank you!

@@driggsalonzo9409 you are always welcome. tc

The purpose of using AD is to resolve the name service. Hence i have deployed windows server for dns and AD. If you notice that in the haproxy i used full name quantifiers, the AD or the DNS server is helping to resolve the name. Haproxy ent 1.8 has the capability to act as dns and can be used to resolve names. But i prefered Windows server DNS.

@@mroyon Thanks for the clarification

@@fasooly1992 you are most welcone

thank you so much.

you are always welcome.. :)

no dear, I haven't tried that. let me know your experiences. tc

Ok.. ill try.. first is the replication config has been done perfectly.. forget about the master slave at the beginning.. make sure if you have three redis.. are the configured and doing replication? If no make the changes in the config to do so.. this is your first objective.. if u have configured properly.. then this would take care of master slave data sync.

If possible can u share the configs so i can test at my lab to help you more precisely..

@@mroyon Thanks for your reply but I have resolved this particular issue from my end. In order to resolve this I have set the BIND address 0.0.0.0 rather than private IP of all the redis-server and sentinel configs, By using your guide the replication and HA availability working fine on my production. Only this is the video/tutorial which helps me to understand the proper behaviour. Thank you so much.

@@waqasshaikh8722 great job, you are most welcome.

Yes, you r right. The main routing point is one, and if it is down.. then routing is down.. so to make high availability you have to place two node for ha too.. now.. lets take one more step before ha.. user to dns (even this dns can fail or down so... eve if theres 5% chance) need backup dns. From dns to ha, need 2... then.. you continue..

​@@mroyon Thanks for the comment. I am using Jedis in java which has a client support for sentinel, so I guess I don't need to use HA Proxy :D

@@abhinavporwal4766 you are welcome. Nop, you do not need ha proxy.. you can manage without that.

software based: NGINX, ENVOY stackshare.io/haproxy/alternatives

Thank you. I really appreciate your words.

Yes you can.. Just distribute the bandwidth else request might clog the nic.

@@mroyon thank you

@@rezanajafi2140 you are most welcome.

Just few things to ask, is it necessary to use cache of webfarm for your scenario? What kind of data and for how long you want to put it in the cache? Is it just you put the event in the cache and after using that data you remove from the cache. What are you planning to use for cache storage? If you have load balancing for IIS-A and IIS-B, and from the question, Im guessing its a fail over scenario. If the webfarm cache is using shared storage and has been configured in such manner to serve independently, mean to say not depending on the IIS, then yes IIS-B will take over from where IIS-A left off. Again it depends on what and how you have configured. My way of thinking would suggest that, make the cache server completely independent from the Web Servers. Weather IIS-A or II-B fails.. it does not affect the cache server. even the scenario would support you to gain high salability. If one cache server is not enough later you can add more node to it.

:) thanks for the comment. I will try my best.