1. Laravel Horizon - Monitor Redis Queue Jobs

[00:00] welcome to bed films and let me

[00:01] introduce you with an official package

[00:04] of level which is called larval horizon

[00:07] this is a great package and let's first

[00:10] quickly get started with this package by

[00:12] using this composer command to have this

[00:16] inside our level application so he'll is

[00:19] a fresh installation of lateral and I

[00:21] will say composer require level /

[00:23] horizon-- so this package is very useful

[00:27] when you are using cues with cue driver

[00:31] as readies so it provides a beautiful

[00:34] dashboard and some configuration for

[00:36] your level power red excu so very

[00:39] important thing is you need to use cue

[00:42] driver as reddit's this package provides

[00:44] this kind of cool dashboard where you

[00:47] can have number of jobs number of fill

[00:49] job number of processes and you can also

[00:51] retry some field job but let's first see

[00:55] why we need this kind of package so yeah

[01:00] it is installed so here I am and I will

[01:02] create a simple notification so make

[01:05] notification called user notification

[01:08] and this notification is just a simple

[01:10] notification which is sending an email

[01:12] to a user so I will open web dot PHP

[01:16] file and here I will go and send this

[01:20] notification to our user so find with

[01:24] the idea of one and notify user notify

[01:28] with new notify user notification okay

[01:34] so we need to grab this user also and at

[01:37] last we are going to return simply done

[01:40] so I have already configured my dot e

[01:43] and we file so with the test database

[01:46] and inside our test database we also

[01:49] have a one user with the ID of 1 this

[01:51] means if I start the server and if I go

[01:55] to localhost 8000

[01:58] you will see it is taking long time and

[02:02] by the way I also have configured this

[02:05] mail trap on our dot env file so here

[02:08] you can see I

[02:10] have configured so it always take time

[02:13] so even if I have one small refresh it

[02:15] take time to send an email because it's

[02:18] processing the sending part email

[02:20] sending part and now we have the to

[02:22] email we can make this sending of email

[02:26] or sending notification in a queue so I

[02:29] will open notify user and here we just

[02:33] need to say simply implementing so like

[02:36] implements should queue just by using

[02:41] this we have now moved this notification

[02:44] into queue

[02:45] okay so if i refresh this page still it

[02:49] is taking time because now if I go and

[02:53] on this dot Ian we file I said that

[02:56] queue connection is sync so this means

[03:01] it is thinking the queue we need to

[03:04] change it to red is because now we want

[03:06] the queue to be used by Redis once more

[03:11] restart the server and refresh this page

[03:14] and you can see instantly it's giving

[03:17] error because level by default uses PHP

[03:21] Redis

[03:22] so either you can install PHP Redis or P

[03:26] Redis so once more go to level

[03:28] documentation and on this documentation

[03:31] you can search for Redis and on this

[03:36] Redis part you can see we can you either

[03:38] use red Pierre IDs or PHP IDs but I

[03:41] personally like P Redis and it is very

[03:43] easy because it is also given here so

[03:45] you need to run this command composer

[03:48] command so let's stop the server run

[03:51] this composer command finally our peer

[03:54] Eddy's is installed so now once more we

[03:57] can start the server and go to localhost

[04:01] 8080 in this error but why we are

[04:06] getting this error because we need to

[04:08] tell to level that level don't use PHP

[04:11] Redis you need to use the period is how

[04:14] we can do that for that we need to go to

[04:18] the configuration of Redis so let's go

[04:20] and here inside config we have a

[04:24] spot inside this database at the last we

[04:28] can see we have this red spot and ready

[04:31] is using PHP Redis you can simply define

[04:34] it on env files so instead of doing this

[04:37] from here I will go here and define

[04:40] Redis client so like here right if

[04:44] client is P Redis very nice we have made

[04:49] changes in dot E and V file so we need

[04:51] to restart the server and one more try

[04:54] and this time with no error and super

[04:58] quickly no delay we have moved our email

[05:01] into this queues but once more if I

[05:06] delete everything from here refresh this

[05:08] page you can see we don't have any email

[05:11] this means our queues are not processed

[05:14] so how we can generally process so we

[05:17] can say PHP artisan queue work and then

[05:21] if I do this yes now I'm having the

[05:25] email because now these notification are

[05:28] processed that good but now comes our

[05:32] level horizon so let's now continue our

[05:35] installation of level horizon so we have

[05:38] installed it via this composer command

[05:41] but we need to run this command to

[05:43] install its effects so I will stop this

[05:46] queue worker and then I will say PHP

[05:49] artisan horizon : installed so it has

[05:52] done three things published the service

[05:54] provider some assets and the

[05:56] configuration so let's start with the

[05:58] configuration on this config folder we

[06:00] have this horizontal Asian and in this

[06:03] horizontal figuration the unique thing

[06:06] you can see we have this path which is

[06:08] horizon this means if you have server

[06:12] running you can simply go on localhost

[06:16] 8004 as on and then you can see you have

[06:20] a cool dashboard and instantly you can

[06:23] see it has picked two job because

[06:26] previously we have ran two jobs so

[06:29] without using queue worker and this

[06:33] means so we can simply see the recent

[06:35] job and other

[06:37] things like monitor and matrix but from

[06:41] where this dashboard came from the UI

[06:44] came from so you can see we have

[06:46] published the effects so we have the

[06:49] effects inside public vendor assets this

[06:52] is just for all the UI it is made

[06:55] overview GS and this is because of that

[06:58] one more thing we have published is

[07:01] service provider which is inside app

[07:04] provider and horizontals provider so in

[07:08] this horizontal provider we have a boot

[07:10] method and a gate method so you can see

[07:13] generally we can access this horizon--

[07:16] route even if I am not logged in but

[07:20] when you want to give access to this

[07:21] horizon dashboard just to one or two

[07:24] people you can just define the email

[07:27] here like this and then the user with

[07:31] this email ID when they are logged in

[07:34] then only they can able to access this

[07:38] route otherwise they will get 404 that's

[07:41] a simple configuration kind of things

[07:43] are related to resin but more is inside

[07:47] this horizon-- configuration file so we

[07:51] have seen that part next is it is going

[07:54] to use the connection so default

[07:56] connection of Redis and this by default

[07:59] prefix everything related to your

[08:03] application with horizontal in' if you

[08:06] want to make changes in this you need to

[08:09] give this information inside your dot

[08:12] env file and we can prove this also so I

[08:15] will go here and I will say red is CLI

[08:18] and if you don't have ready you can

[08:21] simply say brew install Redis if you are

[08:25] using Mac and for window you can I think

[08:27] you can simply find it so ready CLI and

[08:30] in this you can simply say keys and X

[08:35] tricks you can see we have this horizont

[08:38] notifier user and last snapshot this is

[08:42] just because of this horizon and this is

[08:45] actually the prefix we have we are

[08:47] talking about here that's cool there

[08:50] middleware and wait time then the most

[08:52] important thing here is this environment

[08:56] so we have this production environment

[08:58] and local end moment so let's talk about

[09:01] these things first so what about the

[09:05] previous two jobs we have put inside the

[09:08] queue so even if I go on localhost 8080

[09:13] processed it processed the notification

[09:17] but we are not getting any email inside

[09:20] our mail trap so you can see we don't

[09:22] have any email say it was eight minutes

[09:25] ago okay so to process the queue on this

[09:29] Redis with the help of horizon what you

[09:32] need to do you need to run the command

[09:34] which is PHP artisan horizon-- so it's

[09:38] just like you run command PHP artisan

[09:40] queue work similar to that we are using

[09:43] PHP artisan horizont so horizont started

[09:46] successfully and this means now we can

[09:49] see it is now converted to active so our

[09:52] dashboard is active cool so now because

[09:57] it is active you can see now but one

[10:01] previous notification is processed and

[10:03] once more if I clear this and refresh

[10:07] this page you can see now it says the

[10:10] for in the past one in the minute and it

[10:14] is now processed so it is very very cool

[10:18] and you can actually check the failed

[10:21] job so how we can check the failed job

[10:23] how we can fail a job one crazy thing I

[10:26] am going to do here is I'm going to say

[10:29] firstly inside tinker so PHP artisan

[10:32] tinker and I will say hey tinker create

[10:36] 50 user for me ok so now we have a 50

[10:40] user inside our database and this means

[10:43] now I can simply say grab me every user

[10:47] and for each user notify user very nice

[10:51] so let's see what's going to happen this

[10:53] is going to be very very interesting so

[10:56] now if I say refresh this page instantly

[10:59] we are getting done but now we have 50

[11:03] five jobs 54 now and five for the

[11:07] previous one so you can see it says 45

[11:10] jobs are in queue it's currently

[11:13] actually working four and then eight are

[11:16] failed because we have too much of jobs

[11:20] and we have just three processes so

[11:24] while this is processing let's go once

[11:26] again to the horizon-- configuration and

[11:29] you can see we can define the number of

[11:31] process so currently we have three

[11:34] processes and we have one retry

[11:36] that's why failed job comes here because

[11:39] they are just retried once so if I go to

[11:44] the failed jobs you can see we have a

[11:46] retry button so if I click here it's now

[11:49] read trying this so that's cool and now

[11:53] totally we have 35 and still four are

[11:57] inside the processes now total zero and

[12:00] if i refresh out of 50 40 jobs are

[12:04] failed this is not good so you can see

[12:08] we just have 10 email here so once again

[12:10] clear this and now what I'm going to do

[12:13] I'm going to make processes as 10 and

[12:16] retry as 5 and whenever you made change

[12:21] in horizontal you need to restart the

[12:24] horizon this is very important whenever

[12:26] you made changes in horizon-- or even in

[12:30] your notification you need to make

[12:33] restart this horizont process ok

[12:37] so for the sake of safety nets I'm also

[12:41] going to restart the server okay so we

[12:43] have 56 jobs here 40 are failed and now

[12:47] because we have restarted it's picked up

[12:49] that we are working with 10 processes

[12:53] cool refresh this page and now we have

[12:57] form 0 7 tasks and after completion you

[13:00] can see we have total 75 fail previously

[13:03] it was 40 and now it is 75 that means 35

[13:06] still feeling okay one more

[13:09] configuration you can do is like here it

[13:12] is defined that yeah so you can give the

[13:18] balance as simple it is given balance as

[13:21] simple so you can also say Auto so

[13:25] simple is very simple which configure it

[13:28] splits incoming job evenly into the

[13:31] process but if you say auto auto

[13:34] strategy adjust the number of vocal

[13:37] processors per cube based on the current

[13:40] workload ok so I am going to make it

[13:43] auto and then because we have made

[13:46] changes let's restart the horizon-- and

[13:49] now I'm going to say refresh and

[13:54] refreshing with that and now we have 168

[13:58] jobs so 50 more is now added so now we

[14:02] have 1 1 10 field job is still having 35

[14:05] failing but the processes took less time

[14:08] to processing all these queues so in

[14:11] that way you can actually define the

[14:14] last one which is false which is just a

[14:16] simple first-come first out kind of

[14:19] things on notifications or in the queues

[14:22] so this is a simple demonstration of how

[14:26] you can use level horizon and how you

[14:28] can configure with different

[14:30] configurations there are some more

[14:32] things on this horizontal continue or

[14:37] even terminate the horizon simply but in

[14:40] the next episode we are going to see how

[14:42] we can actually deploy this horizon

[14:44] error actually use this horizon on a

[14:47] server are you going to server with the

[14:50] supervisor as our client to run the PHP

[14:53] artisan horizon command ok if you liked

[14:56] this episode please give it a thumbs up

[14:58] and comment your views about level

[15:00] horizont is it a good when you have used

[15:03] and what makes you to use this package

[15:08] and don't forget to subscribe to this

[15:10] channel and like which films on Facebook

[15:12] Twitter Instagram LinkedIn and feeble

[15:14] meet in the next episode till then good

[15:15] bye