This video is absolutely gold. Thank you, it helped me a lot.
@paolonervi2208
11 ай бұрын
Hi Daniel,thank you very much for the time you have dedicated in making this super useful video..cheers from Italy.
@aaronperl
Жыл бұрын
This was a very interesting demonstration. Thanks!
@rajaparthiban410
4 ай бұрын
thanks for the video, this made some knowledge about Maria DB HA
@clecimarfernandes4992
11 ай бұрын
We used to say: "Aonde está a galera ?" that means: How about the guys ? Where they are ? Something like this.
@djkadakamc1889
7 ай бұрын
❤ Next video: configurate maxscale, haproxy, corosync and pacemaker
@DanielPersson
7 ай бұрын
Hi Thank you for watching my videos. I've added the idea to the list of ideas. Best regards Daniel
@andresj5512
2 ай бұрын
Thanks for this video. I have recently had an outage on my DB server, so now I'm running two MariaDB servers on primary and replicate configuration. My question is: I have lots of services that access the primary DB server, if that one goes down, how should I setup things for the services to automatically switch to the replicate server? At the moment I defined a domain with the local IP of the current primary, so I can change the register of that DNS and every service would resolve the new IP.
@DanielPersson
2 ай бұрын
Hi Thank you for watching my videos. At work we use heartbeat a lot, there is also more complicated solutions that do similar things but you could specify an static IP for your server that the clients connect to and then if one server goes down the other one will realize and heartbeat will assign the IP. This video covers heartbeat in general: kzitem.info/news/bejne/r6dusW2JiXOAd5g I hope this helps. Best regards Daniel
@andrespanella5050
20 күн бұрын
Hi Daniel! Thanks for the video. I need to know how many tps maximum I can expect with this configuration. I saw some videos that talk 1million qps. I think that isn't possible in production environments.
@DanielPersson
7 күн бұрын
Hi Andres. That is pretty impossible to answer, you need to test it yourself. The parameters to handle is speed of ram, drive speeds and general CPU utilization of the machine. If you have a really fast machine, ram and disk you can achieve fast queries and even improve speed if you can do them in parallel. I don't have any tooling to benchmark databases but I might find one in the future. Thank you for watching my videos. Best regards Daniel
@damodarsimkhada9482
6 ай бұрын
Hi Daniel , I'm doing this in ubuntu sysrtem and i have come far but while executing sudo galera_ it says that commant not found
@DanielPersson
6 ай бұрын
Hi Maybe you have an older version of mariadb or perhaps you've installed mysql. Hard to know what is wrong without more information. Thank you for watching my videos. Best regards Daniel
@manit77
11 ай бұрын
great video thanks. resources are hard to find on this subject.
@herrsammy
9 ай бұрын
🤝
@ir5had
11 ай бұрын
Thanks
@SangreNL
5 ай бұрын
Thnx for the video, i had one question though, After setting the global var, wsrep_provider_options='pc.bootstrap=true', would it be wise to set is to false after the cluster is up and running again, to prevent that this node will allways be the bootstrap-one?
@DanielPersson
5 ай бұрын
Hi Sangre Yes, that is a really good point. Especially if you need to upgrade / restart a node. So these variables needs to be maintained. Thank you for watching my videos. Best regards Daniel
@kaustubhmanitripathi6942
11 ай бұрын
please make a detailed video on this. please it would really mean alot. thank youuu
@DanielPersson
11 ай бұрын
Hi What detail in the video are you missing? Thank you for watching my videos. Best regards Daniel
@shrine010
11 ай бұрын
So I in case of 3 nodes, when one node fails I shouldn't restart any of remaining nodes without setting pc.bootstrap=true?
@DanielPersson
10 ай бұрын
Hi Shrine. Yes, if the remaining nodes are in a quorum you should be able to start the third node and it should pick up where it left off. Might take a while before it has catches up. I hope this helps, thank you for watching my videos. Best regards Daniel
@shrine010
10 ай бұрын
@@DanielPersson That's correct but I'm mentioning situation when 1 of 3 node fails (or is down) and 2 left are running. Is it safe to restart any of those 2 remaining nodes?
@DanielPersson
10 ай бұрын
Hi Shrine. Ohh, I see how I misunderstood your question. Well in that case I would not restart it without handling it as a "one node left case". It's a complicated case because if you don't want to loose data or connectivity I would much rather get the third node up and running and then restart the first two. Best regards Daniel
@niloufarpadkan822
7 ай бұрын
Thanks. But you did not show the procedure of installing gelara. And also load balancers
@DanielPersson
7 ай бұрын
Hi Well I did all the steps required to install it on a Debian machine. All packages are included in a normal maria-db installation. Load balancing is a different topic but the servers will be in quorum so you could connect to any of them. In my installation I would put a haproxy in front, I have a video about that but it's not required for the service to be useful. Thank you for watching my videos. Best regards Daniel
Пікірлер: 28