github/container.training
1
0
Fork 0
mirror of https://github.com/jpetazzo/container.training.git synced 2026-07-28 09:11:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'bridgetkromhout-master-updates'

Browse Source
This commit is contained in:
Jerome Petazzoni
2018-02-26 17:52:14 -06:00
parent aab519177d 614f10432e
commit 630e275d99
8 changed files with 56 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
slides/kube/daemonset.md
View File

@@ -4,16 +4,30 @@
- We want one (and exactly one) instance of `rng` per node
- If we just scale `deploy/rng` to 4, nothing guarantees that they spread
- What if we just scale up `deploy/rng` to the number of nodes?
- nothing guarantees that the `rng` containers will be distributed evenly
- if we add nodes later, they will not automatically run a copy of `rng`
- if we remove (or reboot) a node, one `rng` container will restart elsewhere
- Instead of a `deployment`, we will use a `daemonset`
---
## Daemon sets in practice
- Daemon sets are great for cluster-wide, per-node processes:
- `kube-proxy`
- `weave` (our overlay network)
- monitoring agents
- hardware management tools (e.g. SCSI/FC HBA agents)
- etc.
- They can also be restricted to run [only on some nodes](https://kubernetes.io/docs/concepts/workloads/controllers/daemonset/#running-pods-on-only-some-nodes)
@@ -22,7 +36,7 @@
## Creating a daemon set
- Unfortunately, as of Kubernetes 1.8, the CLI cannot create daemon sets
- Unfortunately, as of Kubernetes 1.9, the CLI cannot create daemon sets
--
@@ -406,4 +420,4 @@ The timestamps should give us a hint about how many pods are currently receiving
- Bonus exercise 1: clean up the pods of the "old" daemon set
- Bonus exercise 2: how could we have done to avoid creating new pods?
- Bonus exercise 2: how could we have done this to avoid creating new pods?

22
slides/kube/dashboard.md
View File

@@ -89,12 +89,30 @@ The goo.gl URL expands to:
- Connect to https://oneofournodes:3xxxx/
(You will have to work around the TLS certificate validation warning)
- You will have to work around the TLS certificate validation warning
<!-- ```open https://node1:3xxxx/``` -->
]
The dashboard will then ask you which authentication you want to use.
.warning[Make sure that you use `https`! Otherwise, you'll get this error:]
```
This page isn’t working
<oneofournodes> sent an invalid response.
ERR_INVALID_HTTP_RESPONSE
```
.warning[Chrome 63 (and later) as well as recent versions of Edge will refuse to connect.]
We do not know how to work around that issue for the moment.
---
## Dashboard authentication
- We have three authentication options at this point:
- token (associated with a role that has appropriate permissions)
@@ -109,7 +127,7 @@ The goo.gl URL expands to:
## Granting more rights to the dashboard
- The dashboard documentation [explains how to do](https://github.com/kubernetes/dashboard/wiki/Access-control#admin-privileges)
- The dashboard documentation [explains how to do this](https://github.com/kubernetes/dashboard/wiki/Access-control#admin-privileges)
- We just need to load another YAML file!

7
slides/kube/kubectlget.md
View File

@@ -48,7 +48,7 @@
.exercise[
- Give us more info about them nodes:
- Give us more info about the nodes:
```bash
kubectl get nodes -o wide
```
@@ -136,7 +136,8 @@ There is already one service on our cluster: the Kubernetes API itself.
```
- `-k` is used to skip certificate verification
- Make sure to replace 10.96.0.1 with the CLUSTER-IP shown earlier
- Make sure to replace 10.96.0.1 with the CLUSTER-IP shown by `kubectl get svc`
]
@@ -173,7 +174,7 @@ The error that we see is expected: the Kubernetes API requires authentication.
## Namespaces
- Namespaces allow to segregate resources
- Namespaces allow us to segregate resources
.exercise[

2
slides/kube/ourapponkube.md
View File

@@ -40,7 +40,7 @@ In this part, we will:
- We could use the Docker Hub
- Or a service offered by our cloud provider (GCR, ECR...)
- Or a service offered by our cloud provider (ACR, GCR, ECR...)
- Or we could just self-host that registry

6
slides/kube/rollout.md
View File

@@ -149,7 +149,7 @@ Our rollout is stuck. However, the app is not dead (just 10% slower).
- We want to:
- revert to `v0.1`
- revert to `v0.1` (which we now realize we didn't tag - yikes!)
- be conservative on availability (always have desired number of available workers)
- be aggressive on rollout speed (update more than one pod at a time)
- give some time to our workers to "warm up" before starting more
@@ -163,7 +163,7 @@ spec:
spec:
containers:
- name: worker
image: $REGISTRY/worker:v0.1
image: $REGISTRY/worker:latest
strategy:
rollingUpdate:
maxUnavailable: 0
@@ -192,7 +192,7 @@ spec:
spec:
containers:
- name: worker
image: $REGISTRY/worker:v0.1
image: $REGISTRY/worker:latest
strategy:
rollingUpdate:
maxUnavailable: 0

13
slides/kube/setup-k8s.md
View File

@@ -42,20 +42,21 @@
## Other deployment options
- If you are on Google Cloud:
[GKE](https://cloud.google.com/container-engine/)
- If you are on Azure:
[AKS](https://azure.microsoft.com/services/container-service/)
Empirically the best Kubernetes deployment out there
- If you are on Google Cloud:
[GKE](https://cloud.google.com/kubernetes-engine/)
- If you are on AWS:
[EKS](https://aws.amazon.com/eks/)
or
[kops](https://github.com/kubernetes/kops)
... But with AWS re:invent just around the corner, expect some changes
- On a local machine:
[minikube](https://kubernetes.io/docs/getting-started-guides/minikube/),
[kubespawn](https://github.com/kinvolk/kube-spawn),
[Docker4Mac (coming soon)](https://beta.docker.com/)
[Docker4Mac](https://docs.docker.com/docker-for-mac/kubernetes/)
- If you want something customizable:
[kubicorn](https://github.com/kris-nova/kubicorn)

6
slides/kube/versions-k8s.md
View File

@@ -1,8 +1,8 @@
## Brand new versions!
- Kubernetes 1.8
- Docker Engine 17.11
- Docker Compose 1.17
- Kubernetes 1.9.3
- Docker Engine 18.02.0-ce
- Docker Compose 1.18.0
.exercise[

2
slides/kube/whatsnext.md
View File

@@ -160,7 +160,7 @@ Sorry Star Trek fans, this is not the federation you're looking for!
- Raft recommends low latency between nodes
- What if our cluster spreads multiple regions?
- What if our cluster spreads to multiple regions?
--