WEBVTT

00:01.240 --> 00:05.680
So I had a lot of fun putting this course together, and I hope you found some useful content in here,

00:05.950 --> 00:10.720
and I just want to take this last lecture to point you to a few resources that you might want to investigate

00:10.720 --> 00:12.320
as time goes on.

00:13.300 --> 00:19.150
So, for example, our micro service, if you think about what happens every time we call our micro

00:19.150 --> 00:21.200
service, it works this way.

00:21.580 --> 00:24.760
We actually take some information we already have and go format.

00:25.210 --> 00:29.230
We convert it into effectively a text file in the form of JSON.

00:29.590 --> 00:35.380
We send that off to the remote service or micro service, which then captures that text file, which

00:35.380 --> 00:40.700
is Jason translates it back into something go canoe's and then it moves on and does something.

00:41.230 --> 00:46.960
So there's that whole middle part where we're taking data that's already in a go format, converting

00:46.960 --> 00:50.680
into text, then converting it back and then doing the work that's required.

00:51.250 --> 00:53.170
Well, that's actually kind of inefficient.

00:53.380 --> 00:59.050
And there is a protocol known as our P C that actually addresses this problem.

00:59.050 --> 01:04.600
And I'm not going to go through it in detail, but I will post the link to go to the course resources

01:04.600 --> 01:05.290
for this lecture.

01:05.650 --> 01:11.350
And you can have a look at and there are examples for go for C++, for Java, for Python and for others.

01:11.780 --> 01:13.640
So that's one thing you might want to look at.

01:14.770 --> 01:20.890
The other thing to bear in mind is that our micro servers, for example, it actually fires off go routines

01:20.890 --> 01:22.270
that send email.

01:22.570 --> 01:28.210
And it's entirely possible that we could shut down that application for maintenance or whatever while

01:28.210 --> 01:32.650
it's in the process of constructing in about to send an email so that email never gets sent.

01:32.650 --> 01:33.380
And that's bad.

01:33.880 --> 01:38.560
So one thing you might want to look into are signals, signals and traps.

01:38.680 --> 01:43.990
And the ones you really want to pay attention to are ones like Sigge Term where software is terminated.

01:43.990 --> 01:48.280
It sent a chill on Linux and there's an equivalent thing on Windows, of course.

01:48.970 --> 01:50.440
Well, you can actually trap for those.

01:50.470 --> 01:56.500
And I'll post a link to this page as well as to this page, which is a blog post that goes through how

01:56.500 --> 01:59.440
you can shut down worker go routines gracefully.

01:59.450 --> 02:04.750
In other words, keep the program running until all of the existing routines complete, whatever task

02:04.750 --> 02:07.090
they have to complete and then terminate the program.

02:07.330 --> 02:10.780
And it's relatively straightforward, but it's something you might want to look into.

02:11.430 --> 02:16.360
Almost certainly have a follow up to this course at some point in the future where we go through these

02:16.360 --> 02:17.410
and other things.

02:17.770 --> 02:21.670
But in the interim, if you have some spare time, you might want to give these things a look.

02:22.000 --> 02:22.470
All right.

02:22.690 --> 02:23.410
Happy coding.