[![Lines of Code](https://img.shields.io/badge/lines%20of%20code-36k-lightgrey)](https://github.com/HariSekhon/DevOps-Bash-tools#hari-sekhon---devops-bash-tools)
[![Lines of Code](https://sonarcloud.io/api/project_badges/measure?project=HariSekhon_DevOps-Bash-tools&metric=ncloc)](https://sonarcloud.io/dashboard?id=HariSekhon_DevOps-Bash-tools)
shields.io gets throttled after a few reloads :-/ - which is bad because shields latency is high and oftens fails to load some badges which requires you to reload the page a few times until they load, so have gone back to the generic wercker badge
[![Codeship Status for HariSekhon/DevOps-Bash-tools](https://app.codeship.com/projects/1b13b490-38c2-0138-ed09-4a04e28a9ab8/status?branch=master)](https://app.codeship.com/projects/386508)
[![CI Mac 10.15](https://github.com/HariSekhon/DevOps-Bash-tools/workflows/CI%20Mac%2010.15/badge.svg)](https://github.com/HariSekhon/DevOps-Bash-tools/actions?query=workflow%3A%22CI+Mac+10.15%22)
Used heavily by all my [GitHub repos](https://github.com/harisekhon), dozens of [DockerHub builds](https://hub.docker.com/r/harisekhon)([Dockerfiles](https://github.com/HariSekhon/Dockerfiles)) and 300+ [CI builds](https://bitbucket.org/harisekhon/devops-bash-tools/src/master/STATUS.md).
- Advanced Bash environment - `.bashrc` + `.bash.d/*.sh` - tonnes of advanced customizations, aliases and functions. See [.bash.d/README.md](https://github.com/HariSekhon/DevOps-Bash-tools/blob/master/.bash.d/README.md)
- Advanced configuration files for common tools like [vim](https://www.vim.org/), [screen](https://www.gnu.org/software/screen/), [tmux](https://github.com/tmux/tmux/wiki), installs the best sysadmin packages like those above plus [AWS CLI](https://aws.amazon.com/cli/), [Azure CLI](https://docs.microsoft.com/en-us/cli/azure/?view=azure-cli-latest), [GCloud SDK](https://cloud.google.com/sdk), [jq](https://stedolan.github.io/jq/) and many others, adds dynamic Git and shell behaviour enhancements, colouring, functions, aliases and automatic pathing of many common installation locations for many major languages like Python, Perl, Ruby, NodeJS, Golang...
- Adds sourcing to `.bashrc`/`.bash_profile` to automatically inherit all `.bash.d/*.sh` environment enhancements for all technologies (see [Inventory Overview](https://github.com/harisekhon/devops-bash-tools#Inventory-Overview) below)
- Symlinks all `.*` config files to `$HOME` for [git](https://git-scm.com/), [vim](https://www.vim.org/), top, [htop](https://hisham.hm/htop/), [screen](https://www.gnu.org/software/screen/), [tmux](https://github.com/tmux/tmux/wiki), [editorconfig](https://editorconfig.org/), [Ansible](https://www.ansible.com/) etc.
`make install` effectively does `make system-packages bash python aws`, but if you want to pick and choose from different sections, see [Individual Setup Parts](https://github.com/harisekhon/devops-bash-tools#Individual-Setup-Parts) below.
- installation scripts for various OS packages (RPM, Deb, Apk) for various Linux distros ([Redhat RHEL](https://www.redhat.com/en/technologies/linux-platforms/enterprise-linux) / [CentOS](https://www.centos.org/) / [Fedora](https://getfedora.org/), [Debian](https://www.debian.org/) / [Ubuntu](https://ubuntu.com/), [Alpine](https://alpinelinux.org/))
- install if absent scripts for Python, Perl, Ruby, NodeJS and Golang packages - good for minimizing the number of source code installs by first running the OS install scripts and then only building modules which aren't already detected as installed (provided by system packages), speeding up builds and reducing the likelihood of compile failures
- install scripts for [Jython](https://www.jython.org/) and build tools like [Gradle](https://gradle.org/) and [SBT](https://www.scala-sbt.org/) for when Linux distros don't provide packaged versions or where the packaged versions are too old
-`.*` - dot conf files for lots of common software eg. advanced `.vimrc`, `.gitconfig`, massive `.gitignore`, `.editorconfig`, `.screenrc`, `.tmux.conf` etc.
-`.vimrc` - contains many awesome [vim](https://www.vim.org/) tweaks, plus hotkeys for linting lots of different file types in place, including Python, Perl, Bash / Shell, Dockerfiles, JSON, YAML, XML, CSV, INI / Properties files, LDAP LDIF etc without leaving the editor!
-`.screenrc` - fancy [screen](https://www.gnu.org/software/screen/) configuration including advanced colour bar, large history, hotkey reloading, auto-blanking etc.
-`.tmux.conf` - fancy [tmux](https://github.com/tmux/tmux/wiki) configuration include advanced colour bar and plugins, settings, hotkey reloading etc.
- automatic GPG and SSH agent handling for handling encrypted private keys without re-entering passwords, and lazy evaluation to only prompt key load the first time SSH is called
-`aws*.sh` - various [AWS](https://aws.amazon.com/) scripts for EC2 metadata, Spot Termination, SSM Parameter Store secret put from prompt, IAM Credential Reports on IAM users without MFA, old access keys and passwords, old user accounts that haven't logged in or used an access key recently, show password policy / set hardened password policy, show unattached IAM policies, show account summary to check various details including root account MFA enabled and no access keys, KMS keys rotation status, CloudTrail & Config status etc.
-`curl_auth.sh` - wraps curl to send your username and password from environment variables or interactive prompt through a ram file descriptor to avoid using the `-u`/`--user` which might otherwise expose your credentials in the process list or OS audit log files. Used by other API querying scripts
-`k8s_api.sh` - finds Kubernetes API and runs your curl arguments against it, auto-getting authorization token and populating `Authorization: Bearer` header
-`ldap_user_recurse.sh` / `ldap_group_recurse.sh` - recurse Active Directory LDAP users upwards to find all parent groups, or groups downwards to find all nested users (useful for debugging LDAP integration and group-based permissions)
-`kafka_*.sh` - scripts to make [Kafka](http://kafka.apache.org/) CLI usage easier including auto-setting Kerberos to source TGT from environment and auto-populating broker and zookeeper addresses. These are auto-added to the `$PATH` when `.bashrc` is sourced. For something similar for [Solr](https://lucene.apache.org/solr/), see `solr_cli.pl` in the [DevOps Perl Tools](https://github.com/harisekhon/devops-perl-tools) repo.
-`zookeeper_client.sh` - wraps zookeeper-client, auto-populating the zookeeper quorum from the environment variable `$ZOOKEEPERS` or else parsing the zookeeper quorum from `/etc/**/*-site.xml` to make it faster and easier to connect
-`zookeeper_shell.sh` - wraps Kafka's zookeeper-shell, auto-populating the zookeeper quorum from the environment variable `$KAFKA_ZOOKEEPERS` and optionally `$KAFKA_ZOOKEEPER_ROOT` to make it faster and easier to connect
-`beeline.sh` - connects to [HiveServer2](https://cwiki.apache.org/confluence/display/Hive/HiveServer2+Overview) via beeline, auto-populating Kerberos and SSL settings, zookeepers for HiveServer2 HA discovery if the environment variable `$HIVE_HA` is set or using the `$HIVESERVER_HOST` environment variable so you can connect with no arguments (prompts for HiveServer2 address if you haven't set `$HIVESERVER_HOST` or `$HIVE_HA`)
-`beeline_zk.sh` - connects to [HiveServer2](https://cwiki.apache.org/confluence/display/Hive/HiveServer2+Overview) HA via beeline, auto-populating SSL and ZooKeeper service discovery settings (specify `$HIVE_ZOOKEEPERS` environment variable to override). Automatically called by `beeline.sh` if either `$HIVE_ZOOKEEPERS` or `$HIVE_HA` is set (the latter parses `hive-site.xml` for the ZooKeeper addresses)
-`hive_foreach_table.sh` - executes a SQL query against every table, replacing `{db}` and `{table}` in each iteration eg. `select count(*) from {table}`
-`hive_*.sh` - various scripts using `beeline.sh` to list databases, tables, for all tables: row counts, DDL metadata field extraction, table locations etc.
-`impala_shell.sh` - connects to [Impala](https://impala.apache.org/) via impala-shell, parsing the Hadoop topology map and selecting a random datanode to connect to its Impalad. This is mostly for convenience to shorten commands and while it acts as a poor man's load balancer, you might want to instead use my real load balancer [HAProxy config for Impala](https://github.com/HariSekhon/HAProxy-configs) (and many other Big Data & NoSQL technologies). Optional environment variables `$IMPALA_HOST` (eg. point to HAProxy load balancer) and `IMPALA_SSL=1` (or use regular impala-shell `--ssl` argument pass through)
-`impala_foreach_table.sh` - executes a SQL query against every table, replacing `{db}` and `{table}` in each iteration eg. `select count(*) from {table}`
-`impala_*.sh` - various scripts using `impala_shell.sh` to list databases, tables, for all tables: row counts, DDL metadata field extraction, table locations etc.
-`mysql.sh` - connects to [MySQL](https://www.mysql.com/) via `mysql`, auto-populating settings from both standard environment variables like `$MYSQL_TCP_PORT`, `$DBI_USER`, `$MYSQL_PWD` (see [doc](https://dev.mysql.com/doc/refman/8.0/en/environment-variables.html)) and other common environment variables like `$MYSQL_HOST` / `$HOST`, `$MYSQL_USER` / `$USER`, `$MYSQL_PASSWORD` / `$PASSWORD`, `$MYSQL_DATABASE` / `$DATABASE`
-`mysql_foreach_table.sh` - executes a SQL query against every table, replacing `{db}` and `{table}` in each iteration eg. `select count(*) from {table}`
-`mysql_*.sh` - various scripts using `mysql.sh` for row counts, iterating each table, or outputting clean lists of databases and tables for quick scripting
-`psql.sh` - connects to [PostreSQL](https://www.postgresql.org/) via `psql`, auto-populating settings from environment variables, using both standard postgres supported environment variables like `$PG*` (see [doc](https://www.postgresql.org/docs/9.0/libpq-envars.html)) as well as other common environment variables like `$POSTGRESQL_HOST` / `$POSTGRES_HOST` / `$HOST`, `$POSTGRESQL_USER` / `$POSTGRES_USER` / `$USER`, `$POSTGRESQL_PASSWORD` / `$POSTGRES_PASSWORD` / `$PASSWORD`, `$POSTGRESQL_DATABASE` / `$POSTGRES_DATABASE` / `$DATABASE`
-`postgres_foreach_table.sh` - executes a SQL query against every table, replacing `{db}`, `{schema}` and `{table}` in each iteration eg. `select count(*) from {table}`
-`postgres_*.sh` - various scripts using `psql.sh` for row counts, iterating each table, or outputting clean lists of databases, schemas and tables for quick scripting
-`hdfs_checksum*.sh` - walks an HDFS directory tree and outputs HDFS native checksums, MD5-of-MD5 or the portable externally comparable CRC32, in serial or in parallel to save time
-`hdfs_find_replication_factor_1.sh` / `hdfs_set_replication_factor_3.sh` - finds HDFS files with replication factor 1 / sets HDFS files with replication factor <=2 to replication factor 3 to repair replication safety and avoid no replica alarms during maintenance operations (see also Python API version in the [DevOps Python Tools](https://github.com/harisekhon/devops-python-tools) repo)
-`hdfs_file_size.sh` / `hdfs_file_size_including_replicas.sh` - quickly differentiate HDFS files raw size vs total replicated size
-`cloudera_manager_api.sh` - script to simplify querying [Cloudera Manager](https://www.cloudera.com/products/product-components/cloudera-manager.html) API using environment variables, prompts, authentication and sensible defaults. Built on top of `curl_auth.sh`
-`cloudera_manager_impala_queries*.sh` - queries [Cloudera Manager](https://www.cloudera.com/products/product-components/cloudera-manager.html) for recent [Impala](https://impala.apache.org/) queries, failed queries, exceptions, DDL statements, metadata stale errors, metadata refresh calls etc. Built on top of `cloudera_manager_api.sh`
-`cloudera_manager_yarn_apps.sh` - queries [Cloudera Manager](https://www.cloudera.com/products/product-components/cloudera-manager.html) for recent [Yarn](https://hadoop.apache.org/docs/current/hadoop-yarn/hadoop-yarn-site/YARN.html) apps. Built on top of `cloudera_manager_api.sh`
-`cloudera_navigator_api.sh` - script to simplify querying [Cloudera Navigator](https://www.cloudera.com/products/product-components/cloudera-navigator.html) API using environment variables, prompts, authentication and sensible defaults. Built on top of `curl_auth.sh`
-`cloudera_navigator_audit_logs.sh` - fetches [Cloudera Navigator](https://www.cloudera.com/products/product-components/cloudera-navigator.html) audit logs for given service eg. hive/impala/hdfs via the API, simplifying date handling, authentication and common settings. Built on top of `cloudera_navigator_api.sh`
-`cloudera_navigator_audit_logs_download.sh` - downloads [Cloudera Navigator](https://www.cloudera.com/products/product-components/cloudera-navigator.html) audit logs for each service by year. Skips existing logs, deletes partially downloaded logs on failure, generally retry safe (while true, Control-C, not `kill -9` obviously). Built on top of `cloudera_navigator_audit_logs.sh`
-`git_foreach_branch.sh` - runs a command on all branches (useful in heavily version branched repos like in my [Dockerfiles](https://github.com/HariSekhon/Dockerfiles) repo)
-`git_foreach_repo.sh` - runs a command on all adjacent repos from a given repolist (useful for updating all your github projects across work and home computers)
-`git_merge_all.sh` / `git_merge_master.sh` / `git_merge_master_pull.sh` - merges updates from master branch to all other branches to avoid drift on longer lived feature branches / version branches (eg. [Dockerfiles](https://github.com/HariSekhon/Dockerfiles) repo)
-`git_remotes_add_public_repos.sh` - auto-creates a checkout's remotes to the 3 major public repositories ([GitHub](https://github.com/)/[GitLab](https://gitlab.com/)/[Bitbucket](https://bitbucket.org))
-`git_remotes_set_multi_origin.sh` - sets up multi-remote origin for unified push to all 3 major public repositories
-`git_repos_update.sh` - updates multiple repos based on a source file mapping list - useful for easily sync'ing lots of Git repos among computers
-`git_submodules_update_repos.sh` - submodule handling, including updating and committing latest submodule updates - used on all my repos for updating shared code submodules
-`github_api.sh` - querying the GitHub API while inferring github repo from local remotes and authenticating using `$GITHUB_TOKEN` or token from git checkout's remote github url when available. Used as a base for several other scripts that use the GitHub API. Built on top of `curl_auth.sh`
-`github_get_user_ssh_public_key.sh` / `github_get_user_ssh_public_key_api.sh` - fetches GitHub users public SSH keys for quick local installation to `~/.ssh/authorized_keys`
-`github_generate_status_page.sh` - generates a [STATUS.md](https://bitbucket.org/harisekhon/devops-bash-tools/src/master/STATUS.md) page by merging all the README.md headers for all a user's non-forked GitHub repos or a given list of any repos etc.
-`jenkins_cli.sh` - runs Jenkins CLI, auto-inferring basic configuations, auto-downloads `jenkins-cli.jar` from Jenkins server if not present, infers a bunch of Jenkins related variables like `$JENKINS_URL` and authentication from `$JENKINS_USER`/`$JENKINS_PASSWORD`, or finds admin password from inside local docker container. Used heavily by `jenkins.sh` one-shot setup
-`jenkins_password.sh` - gets Jenkins admin password from local docker container. Used by `jenkins_cli.sh`
-`jenkins.sh` - one-touch [Jenkins CI](https://jenkins.io/), launches in docker, installs plugins, validates `Jenkinsfile`, configures jobs from `$PWD/setup/jenkins-job.xml` and sets Pipeline to git remote origin's `Jenkinsfile`, triggers build, tails results in terminal. Call from any repo top level directory with a `Jenkinsfile` pipeline and `setup/jenkins-job.xml` (all mine have it)
-`concourse.sh` - one-touch [Concourse CI](https://concourse-ci.org/), launches in docker, configures pipeline from `$PWD/.concourse.yml`, triggers build, tails results in terminal, prints recent build statuses at end. Call from any repo top level directory with a `.concourse.yml` config (all mine have it), mimicking structure of fully managed CI systems
-`gocd.sh` - one-touch [GoCD](https://www.gocd.org/), launches in docker, (re)creates config repo (`$PWD/setup/gocd_config_repo.json`) from which to source pipeline(s) (`.gocd.yml`), detects and enables agent(s) to start building. Call from any repo top level directory with a `.gocd.yml` config (all mine have it), mimicking structure of fully managed CI systems
-`perl_cpanm_install.sh` - bulk installs CPAN modules from mix of arguments / file lists / stdin, accounting for User vs System installs, root vs user sudo, [Perlbrew](https://perlbrew.pl/) / Google Cloud Shell environments, Mac vs Linux library paths, ignore failure option, auto finds and reads build failure log for quicker debugging showing root cause error in CI builds logs etc
-`perl_cpanm_install_if_absent.sh` - installs CPAN modules not already in Perl libary path (OS or CPAN installed) for faster installations only where OS packages are already providing some of the modules, reducing time and failure rates in CI builds
-`perlpath.sh` - prints all Perl libary search paths, one per line
-`perl_find_library_executable.sh` - finds directory where a CPAN module's CLI program is installed (system vs user, useful when it gets installed to a place that isn't in your `$PATH`, where `which` won't help)
-`perl_find_duplicate_cpan_requirements.sh` - finds duplicate CPAN modules listed for install more than once under the directory tree (useful for deduping module installs in a project and across submodules)
-`python_compile.sh` - byte-compiles Python scripts and libraries into `.pyo` optimized files
-`python_pip_install.sh` - bulk installs PyPI modules from mix of arguments / file lists / stdin, accounting for User vs System installs, root vs user sudo, VirtualEnvs / Anaconda / GitHub Workflows/ Google Cloud Shell, Mac vs Linux library paths, and ignore failure option
-`python_pip_install_if_absent.sh` - installs PyPI modules not already in Python libary path (OS or pip installed) for faster installations only where OS packages are already providing some of the modules, reducing time and failure rates in CI builds
-`python_pip_reinstall_all_modules.sh` - reinstalls all PyPI modules which can fix some issues
-`pythonpath.sh` - prints all Python libary search paths, one per line
-`python_find_library_executable.sh` - finds directory where a PyPI module's CLI program is installed (system vs user, useful when it gets installed to a place that isn't in your `$PATH`, where `which` won't help)
-`python_find_unused_pip_modules.sh` - finds PyPI modules that aren't used by any programs in the current directory tree
-`python_find_duplicate_pip_requirements.sh` - finds duplicate PyPI modules listed for install under the directory tree (useful for deduping module installs in a project and across submodules)
-`python_pyinstaller.sh` - creates [PyInstaller](https://www.pyinstaller.org/) self-contained Python programs with Python interpreter and all PyPI modules included
-`spotify_*.sh` - [Spotify](https://www.spotify.com/) API scripts to list playlists, playlist contents as track URIs / `Artists - Track` / CSV, convert Spotify track URIs to `Artist - Track` / CSV, create backups of all Spotify playlists, iterate any command against all playlists, query any API endpoint with authentication etc.
- all builds across all my GitHub repos now `make system-packages` before `make pip` / `make cpan` to shorten how many packages need installing, reducing chances of build failures
* [The Advanced Nagios Plugins Collection](https://github.com/harisekhon/nagios-plugins) - 450+ programs for Nagios monitoring your Hadoop & NoSQL clusters. Covers every Hadoop vendor's management API and every major NoSQL technology (HBase, Cassandra, MongoDB, Elasticsearch, Solr, Riak, Redis etc.) as well as message queues (Kafka, RabbitMQ), continuous integration (Jenkins, Travis CI) and traditional infrastructure (SSL, Whois, DNS, Linux)
[Pre-built Docker images](https://hub.docker.com/u/harisekhon/) are available for those repos (which include this one as a submodule) and the ["docker available"](https://hub.docker.com/r/harisekhon/centos-github/) icon above links to an [uber image](https://hub.docker.com/r/harisekhon/centos-github/) which contains all my github repos pre-built. There are [Centos](https://hub.docker.com/r/harisekhon/centos-github/), [Alpine](https://hub.docker.com/r/harisekhon/alpine-github/), [Debian](https://hub.docker.com/r/harisekhon/debian-github/) and [Ubuntu](https://hub.docker.com/r/harisekhon/ubuntu-github/) versions of this uber Docker image containing all repos.
make build installs all dependencies - OS packages and any language libraries via native tools eg. pip, cpanm, gem, go etc that are not available via OS packages
make system-packages installs OS packages only (detects OS via whichever package manager is available)
make test run tests
make clean removes compiled / generated files, downloaded tarballs, temporary files etc.
make submodules initialize and update submodules to the right release (done automatically by build / system-packages)
make cpan install any modules listed in any cpan-requirements.txt files if not already installed
make pip install any modules listed in any requirements.txt files if not already installed
make python-compile compile any python files found in the current directory and 1 level of subdirectory
make pycompile
make github open browser at github project
make readme open browser at github's README
make github-url print github url and copy to clipboard
make ls-files print list of files in project
make ls-code print list of code files, excluding READMEs and other peripheral files
make wc show line counts of the files and grand total
make wc-code show line counts of only code files and total
Repo specific options:
make install builds all script dependencies, installs AWS CLI, symlinks all config files to $HOME and adds sourcing of bash profile
make link symlinks all config files to $HOME and adds sourcing of bash profile
make unlink removes all symlinks pointing to this repo's config files and removes the sourcing lines from .bashrc and .bash_profile
make python-desktop installs all Python Pip packages for desktop workstation listed in setup/pip-packages-desktop.txt
make perl-desktop installs all Perl CPAN packages for desktop workstation listed in setup/cpan-packages-desktop.txt
make ruby-desktop installs all Ruby Gem packages for desktop workstation listed in setup/gem-packages-desktop.txt
make golang-desktop installs all Golang packages for desktop workstation listed in setup/go-packages-desktop.txt
make mac-desktop all of the above + installs a bunch of major common workstation software packages like Ansible, Terraform, MiniKube, MiniShift, SDKman, Travis CI, CCMenu, Parquet tools etc.
(`make help` exits with error code 3 like most of my programs to differentiate from build success to make sure a stray `help` argument doesn't cause silent build failure with exit code 0)