github/linuxkit
1
0
Fork 0
mirror of https://github.com/linuxkit/linuxkit.git synced 2025-07-21 18:11:35 +00:00
Code Issues Projects Releases Wiki Activity

projects: Remove the etcd projects

Browse Source 
It was last used close to a year ago and is not longer
maintained/used.

Signed-off-by: Rolf Neugebauer <rolf.neugebauer@docker.com>
This commit is contained in:
Rolf Neugebauer 2018-01-05 15:19:42 +00:00
parent 07c2f49edc
commit db52123afc
15 changed files with 0 additions and 423 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

105
projects/etcd/README.md
View File

@ -1,105 +0,0 @@
This directory contains files used in Moby/LinuxKit DockerCon 2017
keynote etcd cluster demo. They mostly serve as examples and probably
need adjustments to your specific environment. They may also break
over time :)
## Prerequisites
Most of the scripts/files assume you are on a Mac.
- Recent Docker for Mac installed (We used 17.05.0-ce-rc1-mac8 from the edge channel)
- For the GCP portion: `brew install google-cloud-sdk`
- Infrakit: Clone [infrakit](https://github.com/docker/infrakit) and
the [GCP plugin](https://github.com/docker/infrakit.gcp) for
infrakit. The GCP plugin, needs to be v0.1. For each, `make
build-in-container` and then copy the contents of `./build`
somewhere in your path.
## etcd cluster setup
An `etcd` cluster can be bootstrapped in different ways (see the [Documentation](https://coreos.com/etcd/docs/latest/op-guide/clustering.html) for more details. For the demo we use configuration via static IP addresses. With Infrakit these are managed by assigning `LogicalID`s to cluster members. The `LogicalID` is interpreted as a IP address.
The `etcd` package takes the official `etcd` container and adds a
[script](./pkg/etcd.sh) to start `etcd`. [etcd.sh](./pkg/etcd.sh)
first attempts to join a new cluster. If that fails it attempts to
join an existing cluster. Note, the number and members of the cluster
are somewhat hard coded in the script.
Each node is also configured with a disk, which is mounted inside the
`etcd` container. `etcd` uses it to keep some state to help with
restarts.
## GCP Setup
You probably want to change the project/zone
```
export CLOUDSDK_CORE_PROJECT=docker4x
export CLOUDSDK_COMPUTE_ZONE=europe-west1-d
gcloud auth login
gcloud auth application-default login
```
You may also want to create ssh-keys and upload them. See the [Generating a new SSH key-pair section](https://cloud.google.com/compute/docs/instances/connecting-to-instance)
Note, the demo uses static IP addresses and they are specific to our
setup. The IP addresses need to be changed in the `infrakit-gcp.json`
config file.
In order to use the static IP addresses we created a custom network:
```
gcloud compute networks create rneugeba-demo --mode auto
gcloud compute networks subnets list
# get IP subnet for rneugeba-demo
gcloud compute firewall-rules create rneugeba-demo-internal --network \
rneugeba-demo --allow tcp,udp,icmp --source-ranges 10.132.0.0/9
```
The firewall setup means that all our projects networks can talk to the demo
network.
## Preparation
We create a number of local packages, not pulled from Hub. To build them, invoke `./build-pkg.sh` in the `./pkg` directory.
Then build the various YAML files using the `linuxkit` tool and package/upload them to Google Cloud using the `linuxkit` tool.
## InfraKit cluster setup
This should create a HyperKit based, InfraKit managed `etcd` cluster with 5 `etcd` instances.
Start InfraKit:
```
./start-infrakit
```
Note: The HyperKit InfraKit plugin must be started from the directory
where the `etcd` mobylinux image is located.
Now, commit the new config:
```
infrakit group commit infrakit.json
```
To check if everything is fine, we created (above) a local `etcd.local` docker image which already has the environment set up to contact the cluster:
```
docker run --rm -ti etcd.local etcdctl member list
docker run --rm -ti etcd.local etcdctl cluster-health
```
You can perform rolling updates, by for example, switching the kernel version in `etcd.yml`, build a new LinuxKit, e.g., `linuxkit build -name etcd-4.10 etcd`, update `infrakit.json`, and then commit the new configuration to InfraKit: `infrakit group commit infrakit.json`.
## Infrakit GCP setup
Start infrakit as above:
```
./start-infrakit
```
Commit the configuration:
```
infrakit group commit infrakit-gcp.json
```
## Prometheus server
The etcd nodes use the Prometheus node exported. You can use the prometheus server image, also in this directory, to collect statistics from etc node. We currently build a specific Prometheus images with hard coded IP addresses. Ideally, the information should be passed in via the metadata/userdata.

44
projects/etcd/etcd.yml
View File

@ -1,44 +0,0 @@
kernel:
image: linuxkit/kernel:4.9.74
cmdline: "console=ttyS0 console=tty0 page_poison=1"
init:
- linuxkit/init:5a577d070817b4f17821657823082651baafd4ed
- linuxkit/runc:abc3f292653e64a2fd488e9675ace19a55ec7023
- linuxkit/containerd:e58a382c33bb509ba3e0e8170dfaa5a100504c5b
- linuxkit/ca-certificates:de21b84d9b055ad9dcecc57965b654a7a24ef8e0
onboot:
- name: sysctl
image: linuxkit/sysctl:4c1ef93bb5eb1a877318db4b2daa6768ed002e21
- name: format
image: linuxkit/format:e945016ec780a788a71dcddc81497d54d3b14bc7
- name: mount
image: linuxkit/mount:b346ec277b7074e5c9986128a879c10a1d18742b
command: ["/usr/bin/mountie", "/var/lib/etcd"]
- name: dhcpcd
image: linuxkit/dhcpcd:0d59a6cc03412289ef4313f2491ec666c1715cc9
command: ["/sbin/dhcpcd", "--nobackground", "-f", "/dhcpcd.conf", "-1"]
- name: metadata
image: linuxkit/metadata:2af15c9f4b0e73515c219b7cc14e6e65e1d4fd6d
services:
- name: rngd
image: linuxkit/rngd:94e01a4b16fadb053455cdc2269c4eb0b39199cd
- name: ntpd
image: linuxkit/openntpd:536e5947607c9e6a6771957c2ff817230cba0d3c
- name: node_exporter
image: linuxkit/node_exporter:db8ba40196c010d06f15ffd231132d360614d507
- name: etcd
image: moby/etcd
capabilities:
- CAP_CHOWN
- CAP_SETUID
- CAP_SETGID
- CAP_DAC_OVERRIDE
- CAP_SYS_ADMIN
- CAP_MKNOD
net: host
binds:
- /var/lib/etcd:/var/lib/etcd
- /var/config/etcd:/etc/etcd
trust:
org:
- linuxkit

53
projects/etcd/infrakit-gcp.json
View File

@ -1,53 +0,0 @@
{
"ID": "etcd.gcp",
"Properties": {
"Allocation": {
"LogicalIDs": [
"10.132.0.200",
"10.132.0.201",
"10.132.0.202",
"10.132.0.203",
"10.132.0.204"
]
},
"Instance": {
"Plugin": "instance-gcp",
"Properties": {
"NamePrefix": "etcd-demo",
"Description": "ETCD test cluster",
"Network": "rneugeba-demo",
"Connect" : true,
"MachineType": "n1-standard-1",
"DiskType": "pd-standard",
"DiskSizeMb": 60,
"DiskImage": "https://www.googleapis.com/compute/v1/projects/docker4x/global/images/etcd",
"AutoDeleteDisk": true,
"ReuseExistingDisk": false,
"Scopes": [ ]
}
},
"Flavor": {
"Plugin": "flavor-vanilla",
"Properties": {
"Init": [
"{",
" \"etcd\": {",
" \"name_prefix\": {",
" \"perm\": \"0644\",",
" \"content\": \"server\"",
"},",
" \"init_cluster\": {",
" \"perm\": \"0644\",",
" \"content\": \"server200=http://10.132.0.200:2380,server201=http://10.132.0.201:2380,server202=http://10.132.0.202:2380,server203=http://10.132.0.203:2380,server204=http://10.132.0.204:2380\"",
"}",
"}",
"}"
],
"Tags": {
"tier": "etcd-cluster",
"project": "infrakit"
}
}
}
}
}

46
projects/etcd/infrakit.json
View File

@ -1,46 +0,0 @@
{
"ID": "etcd",
"Properties": {
"Allocation": {
"LogicalIDs": [
"192.168.65.200",
"192.168.65.201",
"192.168.65.202",
"192.168.65.203",
"192.168.65.204"
]
},
"Instance": {
"Plugin": "instance-hyperkit",
"Properties": {
"kernel+initrd": "etcd",
"Disk" : 2048,
"CPUs" : 1,
"Memory" : 1024
}
},
"Flavor": {
"Plugin": "flavor-vanilla",
"Properties": {
"Init": [
"{",
" \"etcd\": {",
" \"name_prefix\": {",
" \"perm\": \"0644\",",
" \"content\": \"server\"",
"},",
" \"init_cluster\": {",
" \"perm\": \"0644\",",
" \"content\": \"server200=http://192.168.65.200:2380,server201=http://192.168.65.201:2380,server202=http://192.168.65.202:2380,server203=http://192.168.65.203:2380,server204=http://192.168.65.204:2380\"",
"}",
"}",
"}"
],
"Tags": {
"tier": "etcd-cluster",
"project": "infrakit"
}
}
}
}
}

5
projects/etcd/pkg/Dockerfile.etcd
View File

@ -1,5 +0,0 @@
FROM quay.io/coreos/etcd:v3.1.5
COPY ./etcd.sh /
ENTRYPOINT ["/etcd.sh"]

6
projects/etcd/pkg/Dockerfile.etcd.local
View File

@ -1,6 +0,0 @@
# A dockerfile to build an etcd container image from the upstream one
#
# It sets the environment to talk to the local cluster
FROM quay.io/coreos/etcd:v3.1.5
ENV ETCDCTL_ENDPOINTS="http://192.168.65.200:2379,http://192.168.65.201:2379,http://192.168.65.202:2379,http://192.168.65.203:2379,http://192.168.65.204:2379"

2
projects/etcd/pkg/Dockerfile.prom.local
View File

@ -1,2 +0,0 @@
FROM prom/prometheus
ADD etc/prometheus-local.yml /etc/prometheus/prometheus.yml

2
projects/etcd/pkg/Dockerfile.prom.us-central1-f
View File

@ -1,2 +0,0 @@
FROM prom/prometheus
ADD etc/prometheus-us-central.yml /etc/prometheus/prometheus.yml

7
projects/etcd/pkg/build-pkg.sh
View File

@ -1,7 +0,0 @@
#! /bin/sh
docker build -t moby/etcd -f Dockerfile.etcd .
docker build -t etcd.local -f Dockerfile.etcd.local .
docker build -t moby/prom-us-central1-f -f Dockerfile.prom.us-central1-f .
docker build -t moby/prom-local -f Dockerfile.prom.local .

19
projects/etcd/pkg/etc/prometheus-local.yml
View File

@ -1,19 +0,0 @@
global:
scrape_configs:
# - job_name: 'prometheus'
# scrape_interval: 20s
# static_configs:
# - targets: ['localhost:9090']
# Scrape the Node Exporter
- job_name: 'node'
scrape_interval: 20s
static_configs:
- targets: ['localhost:9100', 'localhost:9101', 'localhost:9102', 'localhost:9103', 'localhost:9104']
# # Scrape the containerd exporter
# - job_name: 'containerd'
# scrape_interval: 15s
# static_configs:
# - targets: ['192.168.65.200:13337', '192.168.65.201:13337', '192.168.65.202:13337', '192.168.65.203:13337', '192.168.65.204:13337']

13
projects/etcd/pkg/etc/prometheus-us-central.yml
View File

@ -1,13 +0,0 @@
global:
scrape_configs:
- job_name: 'prometheus'
scrape_interval: 20s
static_configs:
- targets: ['localhost:9090']
# Scrape the Node Exporter
- job_name: 'node'
scrape_interval: 20s
static_configs:
- targets: ['10.128.0.200:9100', '10.128.0.201:9100', '10.128.0.202:9100', '10.128.0.203:9100', '10.128.0.204:9100']

13
projects/etcd/pkg/etc/prometheus.yml
View File

@ -1,13 +0,0 @@
global:
scrape_configs:
- job_name: 'prometheus'
scrape_interval: 20s
static_configs:
- targets: ['localhost:9090']
# Scrape the Node Exporter
- job_name: 'node'
scrape_interval: 20s
static_configs:
- targets: ['10.128.0.200:9100', '10.128.0.201:9100', '10.128.0.202:9100', '10.128.0.203:9100', '10.128.0.204:9100']

58
projects/etcd/pkg/etcd.sh
View File

@ -1,58 +0,0 @@
#! /bin/sh
# debug
set -x
set -v
# Get initial cluster data from metadata
NAME_PREFIX="$(cat /etc/etcd/name_prefix)"
INIT_CLUSTER="$(cat /etc/etcd/init_cluster)"
# Wait till we have an IP address
IP=""
while [ -z "$IP" ]; do
IP=$(ifconfig eth0 2>/dev/null|awk '/inet addr:/ {print $2}'|sed 's/addr://')
sleep 1
done
# Where to store data
DATADIR=/var/lib/etcd
# Name is NAME_PREFIX+last octet of IP address
NUM=$(echo ${IP} | cut -d . -f 4)
NAME=${NAME_PREFIX}${NUM}
# We currently have no easy way to determine if we join a cluster for
# the first time or if we got restarted and need to join an existing
# cluster. So we first try joining a *new* cluster. This fails if we
# had already joined it previously. As a fallback we then try to join
# an existing cluster.
# Try to join an new cluster
/usr/local/bin/etcd \
--name ${NAME} \
--debug \
--log-package-levels etcdmain=DEBUG,etcdserver=DEBUG \
--data-dir $DATADIR \
--initial-advertise-peer-urls http://${IP}:2380 \
--listen-peer-urls http://${IP}:2380 \
--listen-client-urls http://${IP}:2379,http://127.0.0.1:2379 \
--advertise-client-urls http://${IP}:2379 \
--initial-cluster-token etcd-cluster-1 \
--initial-cluster ${INIT_CLUSTER} \
--initial-cluster-state new
[ $? -eq 0 ] && exit 0
# Try to join an existing cluster
/usr/local/bin/etcd \
--name ${NAME} \
--debug \
--log-package-levels etcdmain=DEBUG,etcdserver=DEBUG \
--data-dir $DATADIR \
--initial-advertise-peer-urls http://${IP}:2380 \
--listen-peer-urls http://${IP}:2380 \
--listen-client-urls http://${IP}:2379,http://127.0.0.1:2379 \
--advertise-client-urls http://${IP}:2379 \
--initial-cluster ${INIT_CLUSTER} \
--initial-cluster-state existing

30
projects/etcd/prom-us-central1-f.yml
View File

@ -1,30 +0,0 @@
kernel:
image: mobylinux/kernel:4.9.x
cmdline: "console=ttyS0 page_poison=1"
init:
- linuxkit/init:5a577d070817b4f17821657823082651baafd4ed
- mobylinux/runc:b0fb122e10dbb7e4e45115177a61a3f8d68c19a9
- mobylinux/containerd:18eaf72f3f4f9a9f29ca1951f66df701f873060b
- mobylinux/ca-certificates:eabc5a6e59f05aa91529d80e9a595b85b046f935
onboot:
- name: sysctl
image: linuxkit/sysctl:4c1ef93bb5eb1a877318db4b2daa6768ed002e21
- name: dhcpcd
image: linuxkit/dhcpcd:0d59a6cc03412289ef4313f2491ec666c1715cc9
command: ["/sbin/dhcpcd", "--nobackground", "-f", "/dhcpcd.conf", "-1"]
- name: metadata
image: linuxkit/metadata:2af15c9f4b0e73515c219b7cc14e6e65e1d4fd6d
services:
- name: rngd
image: mobylinux/rngd:3dad6dd43270fa632ac031e99d1947f20b22eec9
- name: prometheus
image: moby/prom-us-central1-f
binds:
- /dev:/dev
- /var/lib/misc:/data
capabilities:
- all
net: host
trust:
image:
- mobylinux/kernel

20
projects/etcd/start-infrakit.sh
View File

@ -1,20 +0,0 @@
# !/bin/sh
# Start the infrakit plugins, save their PID
INFRAKIT_HOME=~/.infrakit
IK_PLUGINS=$INFRAKIT_HOME/plugins
rm -rf $INFRAKIT_HOME
mkdir -p $INFRAKIT_HOME/cli
infrakit-flavor-vanilla &
infrakit-instance-hyperkit &
infrakit-instance-gcp --project $CLOUDSDK_CORE_PROJECT --zone $CLOUDSDK_COMPUTE_ZONE &
# start the group plugin in the foreground. If it exits, it will take
# the others down as well.
infrakit-group-default
rm -rf $INFRAKIT_HOME