Commit e18a7da7 authored by Alexandre MERLIN's avatar Alexandre MERLIN
Browse files

update doc

parent b65ddf16
Pipeline #268536 passed with stages
in 15 minutes and 55 seconds
No preview for this file type
......@@ -601,6 +601,7 @@ environments:
extraction_dir: /mnt/dest
tarball_dir: /tmp
rambin_dir: /rambin
allowed_name_regex: '[A-Za-z0-9\-\_]+'
max_postinstall_size: 10
max_preinstall_size: 10
pxe:
......@@ -768,6 +769,7 @@ external:
\item \ypath{/environments}
\begin{itemize}
\item \yfieldd{allowed\_name\_regex}{String} A regex that control how an environement should be name.
\item \yfieldd{max\_preinstall\_size}{Integer}{20} maximum size (MB) of the preinstall files
\item \yfieldd{max\_postinstall\_size}{Integer}{20} maximum size (MB) of the postinstall files
\end{itemize}
......@@ -981,13 +983,12 @@ To define the specific configuration of a cluster, you must create a specific fi
\begin{verbatim}
---
partitioning:
block_device: /dev/sda
partitions:
swap: 1
prod: 2
deploy: 3
tmp: 5
script: parted_sample
formating_script: formating.sh
deploy_label: DEPLOY
block_device: disk0
disk_path:
disk0: /dev/disk/by-path/pci-0000:00:0d.0-ata-1
trusted_deployment:
user: deploy
partition: "/dev/sda3"
......@@ -1133,17 +1134,12 @@ automata:
\begin{itemize}
\item \ypath{/partitioning}
\begin{itemize}
\item \yfield{block\_device}{String}block device of the disk used on the nodes
\item \yfield{block\_device}{String}default disk used on the nodes
\item \yfield{deploy\_label}{String}default label used on the nodes
\item \yfieldd{disable\_swap}{Boolean}{false} disable the swap partition on the disk
\item \yfield{script}{String} Path to a script that creates the partition table on the nodes. You can use Kadeploy3 environment variables (see section \ref{sec:env_vars}) in this script. Please refer to section \ref{sec:partscript}) for further informations.
\end{itemize}
\item \ypath{/partitioning/partitions}
\begin{itemize}
\item \yfieldd{swap}{Integer}{1} number of the swap partition on the disk
\item \yfieldd{prod}{Integer}{-1} number of the production partition on the disk
\item \yfieldd{deploy}{Integer}{2} number of the deployment partition on the disk
\item \yfieldd{tmp}{Integer}{-1} number of the tmp partition on the disk
\item \yfield{formating\_script}{String} Path to a script that is used to format the partitions. You can use Kadeploy3 environment variables (see section \ref{sec:env_vars}) in this script. Please refer to section \ref{sec:formatscript}) for further informations.
\item \yfield{disk\_path}{Hash} Map the disks name to the real path on the nodes.
\end{itemize}
\item \ypath{/trusted\_deployment}
......@@ -1367,6 +1363,8 @@ Here is the list of the macro-step and their implementation:
\item format\_tmp\_part \texttt{[only if non-multipart]}
\item format\_swap\_part \texttt{[only if non-multipart]}
\end{itemize}
\item \texttt{SetDeploymentTrusted} \\
Check if the environement is trusted based on the cluster configuration if so, use \texttt{SetDeploymentEnvKexec} else use \texttt{SetDeploymentEnvUntrusted}.
\item \texttt{SetDeploymentEnvUntrustedCustomPreInstall}
\begin{itemize}
\item switch\_pxe
......@@ -1509,12 +1507,19 @@ A partitioning script must be provided for each cluster.
You can use the Kadeploy3 environment variables in your script (see section \ref{sec:env_vars}). The tools you can use in your script are the ones that are installed in your deployment environment (see section \ref{sec:deployenv}).
Be careful to write the partition type on MSDOS partitions, some bootloaders will fail to boot unless this information is written in the partition table.
The script must write GPT label on each partition use by Kadeploy3. The label must be prefixed by \texttt{KDPL\_} and end by the disk name (which is given by the \texttt{KADEPLOY\_DISK\_NAME} environment variable).
For exemple, \texttt{KDPL\_SWAP\_disk0} is a valid label.
The \texttt{SWAP} and \texttt{TMP} label are used to identify the swap and tmp partition. The default deploy partition is specify in the cluster configuration of Kadeploy3.
You should also use the \emph{/sbin/partprobe} at the end of your script to inform the OS (deployment environment) of partition table changes.
Examples of scripts are provided in the distribution (in the directory \texttt{scripts/partitioning/}).
\section{Formating script}\label{sec:formatscript}
A formating script can be provided for each cluster. You can use the Kadeploy3 environment variables in your script (see section \ref{sec:env_vars}).
The same script is used for all the formating operations (swap, tmp and deploy partitions). The \texttt{KADEPLOY\_FORMAT\_PART} environment variable allow you to distinguish each operation.
\section{Specific commands configuration files}
In the part~\ref{sec:specific_config} we saw that generic commands can be given to all the nodes that belong to a cluster. It is also possible to override these generic values for some specific nodes. To do this, you must fill the file named \texttt{command.conf} in the configuration directory.
......@@ -2313,12 +2318,11 @@ When writing a script for an admin pre-install, an admin/user post-install or so
\item \texttt{KADEPLOY\_ENV\_KERNEL\_PARAMS} : the deployed environment kernel parameters
\item \texttt{KADEPLOY\_ENV\_HYPERVISOR} : the path to the hypervisor's kernel file inside the deployed environment directory (usefull when deploying a Xen environment for example)
\item \texttt{KADEPLOY\_ENV\_HYPERVISOR\_PARAMS} : the deployed environment hypervisor parameters
\item \texttt{KADEPLOY\_DEPLOY\_PART} : deployment partition
\item \texttt{KADEPLOY\_BLOCK\_DEVICE} : the block device used in deployment
\item \texttt{KADEPLOY\_DEPLOY\_PART\_NUM} : the deployment parition number
\item \texttt{KADEPLOY\_SWAP\_PART\_NUM} : the swap partition number
\item \texttt{KADEPLOY\_PROD\_PART\_NUM} : the production partition number
\item \texttt{KADEPLOY\_TMP\_PART\_NUM} : the tmp partition number
\item \texttt{KADEPLOY\_DEPLOY\_PART} : path to the deployment partition
\item \texttt{KADEPLOY\_TMP\_PART} : path to the tmp partition
\item \texttt{KADEPLOY\_SWAP\_PART} : path to the swap partition
\item \texttt{KADEPLOY\_DISK\_NAME} : the Kadeploy3 name of the disk used in deployment
\item \texttt{KADEPLOY\_BLOCK\_DEVICE} : the path to the block device used in deployment
\item \texttt{KADEPLOY\_ENV\_EXTRACTION\_DIR} : path where the environment tarball is extracted
\item \texttt{KADEPLOY\_PREPOST\_EXTRACTION\_DIR} : path where the pre/post tarball are extracted
\item \texttt{KADEPLOY\_TMP\_DIR} : a temporary directory (to be used for your scripts)
......@@ -2327,6 +2331,7 @@ When writing a script for an admin pre-install, an admin/user post-install or so
\item \texttt{KADEPLOY\_FS\_TYPE} : the filesystem format of the deployment partition (the value of the \emph{filesystem} field of the environment description)
\item \texttt{KADEPLOY\_FS\_TYPE\_TMP} : the filesystem format of the tmp partition (if the tmp partition is reformated with the option --reformat-tmp)
\item \texttt{KADEPLOY\_KEXEC\_KIND} : used for kexec script. Specify what kind of kexec is happening. value are \emph{to\_deployed\_env} or \emph{to\_deploy\_kernel}
\item \texttt{KADEPLOY\_FORMAT\_PART} : used for formating script. Specify what disk is being formated. value are \emph{DEPLOY},\emph{TMP} or \emph{SWAP}
\item \texttt{KADEPLOY\_DEPLOY\_ENV\_KERNEL\_PARAMS} : the deployment kernel kernel parameters
\item \texttt{KADEPLOY\_ENV\_KEXEC\_REPOSITORY} : the location where deploy kernel are copied for kexec purpose
\end{itemize}
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment