Your graph on the installation is already completed and you can use it without making any configuration changes however.
And now I can show you how you can update some of the configurations in order to access the configuration file.
You have to go to slash ATC slash Carafano.
If I do on a less command you will see that there is a graph on that your knife fight and uh configuration are inside that fight.
It is recommended that you don't change that file directly although you can if you want but the recommendation is that you make a copy of that read like pseudo C.P.
and call it cost them eye an eye and make your changes on custom iron eye.
So whichever way you use in order to update the file we can do a pseudo Nano for example graph on the I and I and the configurations that you have one of them is instance name which is by default set to your machine name or machine IP.
We had to dollar hostname and this will be used to identify your own instance when you have multiple instances of graph on.
There are some parts.
The most important one is lugs the luck funds by default or in slash var slash log slash graph 1A.
If you update this path then you come place your luck funds in a different place.
However you have to make sure to remove this semicolon from here.
This semicolon comments these lines and it's a very common mistake that people just change the value but they forget to remove the semicolon.
And if you don't remove this semicolon your update is basically ineffective.
So you need to remove the semicolon to make this effective something that I can show you in here is that by default when you installed graph on r it is installed by Sequoia light which is on the same server.
90 percent of cases.
It's not a problem because you can always export your dashboards and data sources into Jason files and keep them as a backup so in case you lose your machine it's not a problem.
You can always import your dashboards but if you want to put your data in an external database which is very important especially when you use Docker because your containers might die at any time and you might want to put your data in an external database you can install.
For example my sequel or post grass and change this configuration and just change the type to my sequel.
So as you see in here there are.
It says either my sequel progress or sequel light.
So you just remove this sequel like the 3 am type in my sequel instead in here and put the hostname and user name and password and then your graph on R will connect to that database.
Normally if you install only one instance of the graph now you can just leave it on sequel light.
However in two cases it's better to use an external database which is first if you want to make sure that in case you lose your server or your docker container you won't lose data.
The second scenario is that when you have multiple instances of graph fauna and they need to share data so no matter which graph on the instance you access you see the same dashboards and same data sources and everything then you need to have a shared database which has to be either a minus equal external minus equal or external Postgres by external.
I mean that your database must be outside your graph on a server whatever machine you used to install the graph on on.
You cannot put the database on that to put to save these are just a control oh and then control X to exist to make sure that your changes are effective.
You need to do as do system s.t.
El restart graph on the server.
If you don't restart graph on a server then it won't free load the config even though sometimes it's recommended to do sys city el stop graph on our server and then later started a stop start.
Normally is more effective.
That's it no matter whether you install your graph on or on and a one to a machine or you use dark hair as I will explain in the next lecture.
The location of the configuration fight is the same.
You can access the configuration file in the same way and update it in the same way.
0 Comments