mrajashree

Host scaling using selector labels.
1. Create a host say `dev1` and add a unique label to it `foo=bar`.
All hosts with label `foo=bar` will fall under the same HostScalingGroup. While adding new hosts, one of the hosts with this label will be used for cloning
2. Create a webhook to scale hosts from scaling group with label `foo=bar`.  Creation fails if no hosts are found with that label.
3. In Execute, get all hosts sorted in descending order of created time. Find all hosts with label `foo=bar`. Number of such hosts is the length of the HostScalingGroup. If scaling up, consider the least recently created host with that label as the base host. Then from the HostScalingGroup, find the first host that matches base host prefix. Since it is the most recently created one, use its suffix to calculate suffix for the next clone host. Add the selector label `foo=bar` to this clone. Create as many clones as specified by the amount.
4. If scaling down, among all hosts sorted by created time, get the `amount` number of mo

mrajashree

# Webhooks

## New drivers
In 1.5, we have added two additional drivers for webhook-service, one for scaling hosts and the second for redeploying a service based on Dockerhub webhooks. Rest of the framwework remains the same.

## Scaling host
Driver for scaling hosts is implemented by cloning a base host. Hosts will be differentiated into different scaling groups by labels. This label will be provided by the users. The labels should be unique so as to distinguish properly between different host scaling groups. All hosts with the same label, whether created manually or added through webhook-service will be considered in a single host scaling group. This label is a necessary field while adding the receiver hook. It's not mandatory to have any hosts with this label while adding the hook, but while executing there must exist at least one such host. This host needs to be added by the user manually. The rest of the hosts added through webhook-service for any scaling group will be clones of the first added (least re

mrajashree

InvalidXMLException -> NameIDPolicy struct tag
ProtocolBinding -> Add to AuthnRequest
asn1 error: tags don't match -> x509.ParsePKCS1PrivateKey expects RSA private key. Meaning in place of key with:
-----BEGIN PRIVATE KEY-----
it should have
-----BEGIN RSA PRIVATE KEY-----
To convert your private key to RSA private key use this 
`command -> openssl rsa -in server.key -out server_new.key`
If key was generated with this command:
`openssl req -x509 -sha256 -nodes -days 365 -newkey rsa:2048 -keyout privateKey.key -out certificate.crt`

mrajashree

1. PingFederate server -> Server Configuration -> System settings -> Data Stores -> Add
2. PingFederate server -> IdP Configuration -> Create new -> Metadata should be Rancher's generated....-> ACS should be
updated to <rancherIP>:8080/v1-auth/saml/acs

mrajashree

    1  exit
    2  xhost local:root
    3  sudo DISPLAY=$DISPLAY gedit /etc/profile
    4  sudo DISPLAY=$DISPLY
    5  ls -l /tmp/.X11-unix/X1
    6  ps
    7  kill 8448
    8  ps
    9  xvfb-run --server-args='-screen 0, 1024x768x24' google-chrome -start-maximized http://www.example.com &> /dev/null &

mrajashree

Rancher's access control requires 7 fields Display Name Field, User Name Field, UID Field, Groups Field, Private Key, Certificate and Metadata XML

The first 4 fields, Display Name Field, User Name Field, UID Field and Groups Field, map to the attributes released from the data store used by the Identity Provider. These released attributes are specified in the IdP's Attribute Contract.

Generate private key and certificate for the Rancher server, and paste/upload them to the Private Key and Certificate fields respectively. Upload IdP metadata and click Save. This will save all the configuration in cattle db and generate the SP metadata.

mrajashree

1. In case of errors, log the error only if you handle it, if it's going to bubble up, then only return the error
2. If the error might expose too many details to the user, log the error, and return a human readable message
3. Follow import patterns specific to the programming language you use
4. It is a bad practice to use data structure names in your variable names. 
5. The code should be self-documenting or self-explanatory. If that means adding comments, add comments. 
But you should also strive to name and structure things so that the code is self-evident without comments.
6. Better commit messages
7. Close response, http, connections

mrajashree

create multiple AD groups
source: http://www.signalwarrant.com/create-active-directory-groups-bulk-csv-w-powershell/
```
$csv = Import-Csv -Path ".\bulk_input2.csv"

ForEach ($item In $csv) 
    {
	Write-Host "item: $($item)"
        $create_group = New-ADGroup -Name $item.GroupName -GroupCategory $item.GroupCategory -groupScope $item.GroupScope -Path $item.OU       
	Write-Host -ForegroundColor Green "Group $($item.GroupName) created!" 

mrajashree

dnf download <package>: downloads the binaries

mrajashree

https://kubernetes.io/docs/getting-started-guides/fedora/fedora_manual_config/ - ref

1. dnf config-manager --set-enabled updates-testing (to get k8s 1.7.3)
2. dnf install kubernetes
3. edit /etc/kubernetes/config

KUBE_LOGTOSTDERR="--logtostderr=true"

# journal message level, 0 is debug
KUBE_LOG_LEVEL="--v=4"