1
00:00:06,900 --> 00:00:10,380
- In this video, we'll talk
about system D self-healing.

2
00:00:10,380 --> 00:00:11,970
I just have one demo,

3
00:00:11,970 --> 00:00:14,850
but before doing the demo, a short story.

4
00:00:14,850 --> 00:00:18,930
In the early days, I made
good money doing consultancy.

5
00:00:18,930 --> 00:00:22,080
Consultancy for
high-availability clustering.

6
00:00:22,080 --> 00:00:23,250
You know what that is?

7
00:00:23,250 --> 00:00:25,110
In high-availability clustering

8
00:00:25,110 --> 00:00:27,540
you are going to protect a vital service.

9
00:00:27,540 --> 00:00:29,460
And if the vital service goes down

10
00:00:29,460 --> 00:00:30,870
the highly available cluster

11
00:00:30,870 --> 00:00:34,110
make sure that it's started
somewhere else again.

12
00:00:34,110 --> 00:00:36,420
And that's pretty important to ensure

13
00:00:36,420 --> 00:00:39,510
that your services are up
and running all the time.

14
00:00:39,510 --> 00:00:41,250
And then came system D.

15
00:00:41,250 --> 00:00:42,960
And what is system D doing?

16
00:00:42,960 --> 00:00:45,510
Well system D is offering a simplified way

17
00:00:45,510 --> 00:00:47,490
of high availability.

18
00:00:47,490 --> 00:00:49,209
We call it self-healing.

19
00:00:49,209 --> 00:00:50,670
Now the difference is that

20
00:00:50,670 --> 00:00:54,480
a high availability cluster
protects an entire server.

21
00:00:54,480 --> 00:00:57,450
Server goes down; no worries.

22
00:00:57,450 --> 00:00:59,760
Critical services are being started again

23
00:00:59,760 --> 00:01:01,110
on another service.

24
00:01:01,110 --> 00:01:02,760
That's something that you can't obtain

25
00:01:02,760 --> 00:01:05,250
through system D self-healing.

26
00:01:05,250 --> 00:01:06,720
But what you can obtain,

27
00:01:06,720 --> 00:01:09,210
is that if a service is going down,

28
00:01:09,210 --> 00:01:12,330
system D is starting
it again automatically.

29
00:01:12,330 --> 00:01:14,030
And that's all I want to show you.

30
00:01:15,948 --> 00:01:16,950
(typing computer keys)

31
00:01:16,950 --> 00:01:21,950
Alright, system CTL
edits on HTTPD.service.

32
00:01:23,700 --> 00:01:28,700
And we are going to
create a service section.

33
00:01:28,980 --> 00:01:31,683
And in the service
section, oops, it's a typo.

34
00:01:33,810 --> 00:01:38,810
In the surface section, I am
including a restart, as always.

35
00:01:42,180 --> 00:01:47,180
And restart sec is, well,
the slide says three

36
00:01:47,730 --> 00:01:49,530
then I need to be real fast.

37
00:01:49,530 --> 00:01:50,880
Let's make it 10.

38
00:01:50,880 --> 00:01:53,223
So restart sec is 10.

39
00:01:56,250 --> 00:02:00,970
Good. Next, system CTL status on HTTPD

40
00:02:02,310 --> 00:02:04,860
where we don't really see anything.

41
00:02:04,860 --> 00:02:06,750
And what is going to happen

42
00:02:06,750 --> 00:02:11,750
if I use kill minus nine, Of what?

43
00:02:11,970 --> 00:02:13,140
Of 14.

44
00:02:13,140 --> 00:02:14,263
971

45
00:02:14,263 --> 00:02:17,520
That's the HTTPD process.

46
00:02:17,520 --> 00:02:19,623
And I'm doing my status again.

47
00:02:20,700 --> 00:02:21,600
Now what do we see?

48
00:02:21,600 --> 00:02:24,960
We see active, activating, auto restart,

49
00:02:24,960 --> 00:02:26,100
and this auto restart,

50
00:02:26,100 --> 00:02:28,260
well, that was one second ago.

51
00:02:28,260 --> 00:02:30,840
We just need to give it
a little bit of time.

52
00:02:30,840 --> 00:02:33,240
On the slide, you see
two steps that are not

53
00:02:33,240 --> 00:02:35,520
really required anymore.

54
00:02:35,520 --> 00:02:37,950
Just for backward compatibility

55
00:02:37,950 --> 00:02:39,060
I included them.

56
00:02:39,060 --> 00:02:41,610
And that's a system CTL daemon reload.

57
00:02:41,610 --> 00:02:43,470
You don't need to daemon reload,

58
00:02:43,470 --> 00:02:45,420
because system D has made progress

59
00:02:45,420 --> 00:02:47,640
and it picks it up automatically.

60
00:02:47,640 --> 00:02:51,480
So if I do my system CTL
status again, what do we see?

61
00:02:51,480 --> 00:02:54,360
We see it's active and it is running.

62
00:02:54,360 --> 00:02:58,560
Because the system D
self-healing has helped to

63
00:02:58,560 --> 00:03:00,423
repair the failing service.