kudos/check_health
kudos/check_health
2
1
Fork 0
Code Issues 7 Pull Requests Projects Releases 1 Wiki Activity

* Added check_exadata_* plugins

Browse Source 
* Smaller fixes to other plugins
This commit is contained in:
Patrick Van der Veken 2019-02-18 14:48:02 +01:00
parent ef266c72e6
commit c690671896
18 changed files with 1738 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

90
build/exadata/SPECS/hc-exadata-platform.spec Normal file
View File

@ -0,0 +1,90 @@
%define build_timestamp %(date +"%Y%m%d")
Name: hc-exadata-platform
Version: %{build_timestamp}
Release: 1
Summary: The KUDOS Health Checker (HC) for UNIX (platform plugins)
Group: Tools/Monitoring
License: GNU General Public License either version 2 of the License, or (at your option) any later version
URL: http://www.kudos.be
Requires: ksh,hc-linux
BuildArch: noarch
BuildRoot: %{_topdir}/%{name}-%{version}-root
%description
The Health Checker is collection of scripts (plugins) designed to perform regular - but not intensive - health checks on UNIX/Linux systems. It provides plugins for AIX, HP-UX and Linux as well customer specific checks. Checks may include topics such file system mounts, process checks, file consistency etc.
This package contains platform/OS specific plugins.
%prep
%build
%install
rm -rf $RPM_BUILD_ROOT
install -d -m 755 $RPM_BUILD_ROOT/opt/hc/lib
install -d -m 755 $RPM_BUILD_ROOT/opt/hc/lib/platform
install -d -m 755 $RPM_BUILD_ROOT/opt/hc/lib/platform/exadata
cp ../SOURCES/lib/platform/exadata/check_exadata_zfs_logs.sh $RPM_BUILD_ROOT/opt/hc/lib/platform/exadata/check_exadata_zfs_logs.sh
cp ../SOURCES/lib/platform/exadata/check_exadata_zfs_services.sh $RPM_BUILD_ROOT/opt/hc/lib/platform/exadata/check_exadata_zfs_services.sh
cp ../SOURCES/lib/platform/exadata/check_exadata_zfs_share_replication.sh $RPM_BUILD_ROOT/opt/hc/lib/platform/exadata/check_exadata_zfs_share_replication.sh
cp ../SOURCES/lib/platform/exadata/check_exadata_zfs_share_usage.sh $RPM_BUILD_ROOT/opt/hc/lib/platform/exadata/check_exadata_zfs_share_usage.sh
install -d -m 755 $RPM_BUILD_ROOT/etc/opt/hc
cp ../SOURCES/etc/check_exadata_zfs_logs.conf.dist $RPM_BUILD_ROOT/etc/opt/hc/check_exadata_zfs_logs.conf.dist
cp ../SOURCES/etc/check_exadata_zfs_services.conf.dist $RPM_BUILD_ROOT/etc/opt/hc/check_exadata_zfs_services.conf.dist
cp ../SOURCES/etc/check_exadata_zfs_share_replication.conf.dist $RPM_BUILD_ROOT/etc/opt/hc/check_exadata_zfs_share_replication.conf.dist
cp ../SOURCES/etc/check_exadata_zfs_share_usage.conf.dist $RPM_BUILD_ROOT/etc/opt/hc/check_exadata_zfs_share_usage.conf.dist
%post
# ------------------------- CONFIGURATION starts here -------------------------
# location of check_health.sh
HC_BIN="/opt/hc/bin/check_health.sh"
PATH="$PATH:/usr/bin:/etc:/usr/sbin:/usr/ucb:/usr/bin/X11:/sbin"
# ------------------------- CONFIGURATION ends here ---------------------------
echo "INFO: starting post-install script ..."
# refresh symbolic FPATH links
if [[ -x ${HC_BIN} ]]
then
${HC_BIN} --fix-symlinks
(( $? == 0 )) || echo "WARN: updating symlinks failed"
fi
echo "INFO: finished post-install script"
%postun
# ------------------------- CONFIGURATION starts here -------------------------
# location of check_health.sh
HC_BIN="/opt/hc/bin/check_health.sh"
PATH="$PATH:/usr/bin:/etc:/usr/sbin:/usr/ucb:/usr/bin/X11:/sbin"
# ------------------------- CONFIGURATION ends here ---------------------------
echo "INFO: starting post-uninstall script ..."
# refresh symbolic FPATH links
if [[ -x ${HC_BIN} ]]
then
${HC_BIN} --fix-symlinks
(( $? == 0 )) || echo "WARN: updating symlinks failed"
fi
echo "INFO: finished post-uninstall script"
%files
%defattr(-,root,root,755)
%dir /opt/hc/lib
%dir /opt/hc/lib/platform
%dir /opt/hc/lib/platform/exadata
%attr(755, root, root) /opt/hc/lib/platform/exadata/check_exadata_zfs_logs.sh
%attr(755, root, root) /opt/hc/lib/platform/exadata/check_exadata_zfs_services.sh
%attr(755, root, root) /opt/hc/lib/platform/exadata/check_exadata_zfs_share_replication.sh
%attr(755, root, root) /opt/hc/lib/platform/exadata/check_exadata_zfs_share_usage.sh
%dir /etc/opt/hc
%attr(644, root, root) /etc/opt/hc/check_exadata_zfs_logs.dist
%attr(644, root, root) /etc/opt/hc/check_exadata_zfs_services.dist
%attr(644, root, root) /etc/opt/hc/check_exadata_zfs_share_replication.conf.dist
%attr(644, root, root) /etc/opt/hc/check_exadata_zfs_share_usage.conf.dist
%changelog
* Mon Feb 18 2019 <patrick@kudos.be> - 0.0.1
- Initial build

52
build/exadata/hc_build_linux_rpms.sh Normal file
View File

@ -0,0 +1,52 @@
#!/usr/bin/env ksh
#******************************************************************************
# @(#) build script for HC RPM packages (uses 'rpmbuild')
#******************************************************************************
# @(#) Copyright (C) 2014 by KUDOS BVBA (info@kudos.be). All rights reserved.
#
# This program is a free software; you can redistribute it and/or modify
# it under the same terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful, but WITHOUT
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
# FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details
#******************************************************************************
#******************************************************************************
# Requires following build (dir) structures:
#
# hc_build_linux_rpms.sh
# BUILD/
# BUILDROOT/
# RPMS/
# SOURCES/
# SOURCES/bin/<hc_scripts>
# SOURCES/lib/*/<hc_plugins>
# SPECS/<spec_files>
# SRPMS/
#
# Build order:
# 1) Copy sources/scripts to the correct locations
# 2) Copy template, build and installer script files into correct locations
# 3) Execute hc_build_linux_rpms.sh
# 4) RPM packages may be found in the RPMS directory
#******************************************************************************
BUILD_DIR="$(dirname $0)"
# clean up previous packages
rm -f ${BUILD_DIR}/RPMS/*/* >/dev/null
# build main packages
rpmbuild -bb ${BUILD_DIR}/SPECS/hc-exadata-platform.spec
print "List of built packages:"
ls -l ${BUILD_DIR}/RPMS/*/*
exit 0
#******************************************************************************
# END of script
#******************************************************************************

44
configs/etc/check_exadata_zfs_logs.conf.dist Normal file
View File

@ -0,0 +1,44 @@
#******************************************************************************
# @(#) check_exadata_zfs_share_usage.conf
#******************************************************************************
# This is a configuration file for the check_exadata_zfs_share_usage HC plugin.
# All lines starting with a '#' are comment lines.
# [default: indicates hardcoded script values if no value is defined here]
#******************************************************************************
# specify whether to also log passed health checks
# (warning: this may rapidly grow the HC log)
# [default: no]
log_healthy="yes"
# specify the user account for the SSH session to the ZFS appliance(s)
# [default: root]
ssh_user=""
# specify the private key file for the SSH session to the ZFS appliance(s)
# [default: ~root/.ssh/id_rsa]
ssh_key_file=""
# specify additional options for the SSH session to the ZFS appliance(s)
# [default: null]
ssh_opts=""
# specify the ZFS hostname(s), log name(s) & alert levels. Filters
# should be comma-separated. Following logs are supported (filters in brackets)
# alert (critical,major,minor
# fltlog (critical,major,minor)
# system (error)
# scrk (failed)
# Format:
# zfs:<host_name>:<alert|fltlog|audit|scrk|system>:<minor,major,critical>]
# Examples:
# check all major events in the fltlog on myzfs1
# zfs:myzfs1:fltlog:major
# check for all major & critical events in the audit and fltlog on myzfs2
# zfs:myzfs2:audit:major,critical
# zfs:myzfs2:fltlog:major,critical
#******************************************************************************
# End of FILE
#******************************************************************************

38
configs/etc/check_exadata_zfs_services.conf.dist Normal file
View File

@ -0,0 +1,38 @@
#******************************************************************************
# @(#) check_exadata_zfs_services.conf
#******************************************************************************
# This is a configuration file for the check_exadata_zfs_services HC plugin.
# All lines starting with a '#' are comment lines.
# [default: indicates hardcoded script values if no value is defined here]
#******************************************************************************
# specify whether to also log passed health checks
# (warning: this may rapidly grow the HC log)
# [default: no]
log_healthy="yes"
# specify the user account for the SSH session to the ZFS appliance(s)
# [default: root]
ssh_user=""
# specify the private key file for the SSH session to the ZFS appliance(s)
# [default: ~root/.ssh/id_rsa]
ssh_key_file=""
# specify additional options for the SSH session to the ZFS appliance(s)
# [default: null]
ssh_opts=""
# specify the ZFS hostname(s), services name(s) and their desired state.
# In order to check share(s) for a given ZFS appliance at least one configuration
# entry must be present: either a wildcard or custom entry.
# Format:
# zfs:<host_name>:<service_name>:<online|disabled>
# Examples:
# check on myzfs1 that svc1 is online
# zfs:myzfs1:svc1:online
#******************************************************************************
# End of FILE
#******************************************************************************

55
configs/etc/check_exadata_zfs_share_replication.conf.dist Normal file
View File

@ -0,0 +1,55 @@
#******************************************************************************
# @(#) check_exadata_zfs_share_replication.conf
#******************************************************************************
# This is a configuration file for the check_exadata_zfs_share_replication HC plugin.
# All lines starting with a '#' are comment lines.
# [default: indicates hardcoded script values if no value is defined here]
#******************************************************************************
# specify whether to also log passed health checks
# (warning: this may rapidly grow the HC log)
# [default: no]
log_healthy="yes"
# specify the user account for the SSH session to the ZFS appliance(s)
# [default: root]
ssh_user=""
# specify the private key file for the SSH session to the ZFS appliance(s)
# [default: ~root/.ssh/id_rsa]
ssh_key_file=""
# specify additional options for the SSH session to the ZFS appliance(s)
# [default: null]
ssh_opts=""
# specify the maximum replication in seconds (general threshold)
# [default: 300]
max_replication_lag=300
# specify the ZFS hostname(s), replication name(s) and their maximum lag (in seconds)
# When not defining a threshold for a given share, the general threshold will
# be used (see above). When defining a threshold of 0 (zero), then the check
# will for this give share will be skipped (this allows for exclusion of shares)
# In order to check share(s) for a given ZFS appliance at least one configuration
# entry must be present: either a wildcard or custom entry.
# Format:
# zfs:<host_name>:<replication_name|*>:<true|false|*>:<success|failed|*>:[<max_replication_lag>]
# Examples:
# check rep_share1 on myzfs1 with a custom threshold of 300 seconds
# zfs:myzfs1:rep_share1:*:*:600
# check all shares of myzfs2 with a custom threshold of 1200 seconds
# zfs:myzfs2:*:*:*:1200
# check all shares of myzfs3 with the general threshold
# zfs:myzfs3:*:*:*:
# disable all shares of myzfs4 from checking
# zfs:myzfs4:*:*:*:0
# disable check of rep_share7 on myzfs5
# zfs:myzfs5:rep_share7:*:*:0
# check that rep_share4 on myzfs6 is inactive
# zfs:myzfs6:rep_share4:false:*:
#******************************************************************************
# End of FILE
#******************************************************************************

53
configs/etc/check_exadata_zfs_share_usage.conf.dist Normal file
View File

@ -0,0 +1,53 @@
#******************************************************************************
# @(#) check_exadata_zfs_share_usage.conf
#******************************************************************************
# This is a configuration file for the check_exadata_zfs_share_usage HC plugin.
# All lines starting with a '#' are comment lines.
# [default: indicates hardcoded script values if no value is defined here]
#******************************************************************************
# specify whether to also log passed health checks
# (warning: this may rapidly grow the HC log)
# [default: no]
log_healthy="yes"
# specify the user account for the SSH session to the ZFS appliance(s)
# [default: root]
ssh_user=""
# specify the private key file for the SSH session to the ZFS appliance(s)
# [default: ~root/.ssh/id_rsa]
ssh_key_file=""
# specify additional options for the SSH session to the ZFS appliance(s)
# [default: null]
ssh_opts=""
# specify the maximum % of space usage a share may reach (general threshold)
# [default: 90]
max_space_usage=90
# specify the ZFS hostname(s), project+share name(s) and their maximum usage (in %)
# When not defining a threshold for a given share, the general threshold will
# be used (see above). When defining a threshold of 0 (zero), then the check
# will for this give share will be skipped (this allows for exclusion of shares)
# In order to check share(s) for a given ZFS appliance at least one configuration
# entry must be present: either a wildcard or custom entry.
# Format:
# zfs:<host_name>:<project_name|*>:<share_name|*>:[<usage_threshold_%>]
# Examples:
# check share1 on myzfs1 with a custom threshold of 75%
# zfs:myzfs1:projectX:share1:75
# check all shares of myzfs2 with a custom threshold of 85%
# zfs:myzfs2:*:*:85
# check all shares of myzfs3 with the general threshold
# zfs:myzfs3:*:*:
# disable all shares of myzfs4 from checking
# zfs:myzfs4:*:*:0
# disable check of share7 on myzfs5
# zfs:myzfs5:projectY:share7:0
#******************************************************************************
# End of FILE
#******************************************************************************

2
configs/etc/check_hpux_fs_usage.conf.dist
View File

@ -32,7 +32,7 @@ max_space_usage=90
# defining a threshold of 0 (zero), then the check will for this give filesystem # defining a threshold of 0 (zero), then the check will for this give filesystem
# will be skipped (this allows for exclusion of filesystems) # will be skipped (this allows for exclusion of filesystems)
# Format: # Format:
# [fs:<mount_point>:<max_inode_usage_%>:<max_space_usage_%] # fs:<mount_point>:[<max_inode_usage_%>]:[<max_space_usage_%]
# Examples: # Examples:
# check myfs1 with a custom inodes + space usage threshold # check myfs1 with a custom inodes + space usage threshold
#fs:myfs1:70:95 #fs:myfs1:70:95

2
configs/etc/check_linux_fs_usage.conf.dist
View File

@ -32,7 +32,7 @@ max_space_usage=90
# defining a threshold of 0 (zero), then the check will for this give filesystem # defining a threshold of 0 (zero), then the check will for this give filesystem
# will be skipped (this allows for exclusion of filesystems) # will be skipped (this allows for exclusion of filesystems)
# Format: # Format:
# [fs:<mount_point>:<max_inode_usage_%>:<max_space_usage_%] # fs:<mount_point>:[<max_inode_usage_%>]:[<max_space_usage_%]
# Examples: # Examples:
# check myfs1 with a custom inodes + space usage threshold # check myfs1 with a custom inodes + space usage threshold
#fs:myfs1:70:95 #fs:myfs1:70:95

2
sources/bin/check_health.sh
View File

@ -38,7 +38,7 @@
# ------------------------- CONFIGURATION starts here ------------------------- # ------------------------- CONFIGURATION starts here -------------------------
# define the version (YYYY-MM-DD) # define the version (YYYY-MM-DD)
typeset -r SCRIPT_VERSION="2019-02-13" typeset -r SCRIPT_VERSION="2019-02-18"
# location of parent directory containing KSH functions/HC plugins # location of parent directory containing KSH functions/HC plugins
typeset -r FPATH_PARENT="/opt/hc/lib" typeset -r FPATH_PARENT="/opt/hc/lib"
# location of custom HC configuration files # location of custom HC configuration files

3
sources/lib/core/include_data.sh
View File

@ -41,7 +41,8 @@ _LVALUE=$(grep -i "^${_PARAMETER} *=" ${_CONFIG_FILE} | cut -f2- -d'=')
if [[ -n "${_LVALUE}" ]] if [[ -n "${_LVALUE}" ]]
then then
print "$(data_dequote \"${_LVALUE}\")" # do not escape inside quotes
print -R "$(data_dequote "${_LVALUE}")"
else else
_RC=1 _RC=1
fi fi

33
sources/lib/core/include_os.sh
View File

@ -114,7 +114,7 @@ return 0
# EXPECTS: n/a # EXPECTS: n/a
# OUTPUTS: 0=not active/installed; 1=active/installed # OUTPUTS: 0=not active/installed; 1=active/installed
# RETURNS: 0=success; 1=error # RETURNS: 0=success; 1=error
# REQUIRES: n/a # REQUIRES: Corosync
function linux_has_crm function linux_has_crm
{ {
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set "${DEBUG_OPTS}" (( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set "${DEBUG_OPTS}"
@ -149,7 +149,7 @@ return 0
# EXPECTS: n/a # EXPECTS: n/a
# OUTPUTS: 0=not active/installed; 1=active/installed # OUTPUTS: 0=not active/installed; 1=active/installed
# RETURNS: 0=success; 1=error # RETURNS: 0=success; 1=error
# REQUIRES: n/a # REQUIRES: Docker
function linux_has_docker function linux_has_docker
{ {
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set "${DEBUG_OPTS}" (( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set "${DEBUG_OPTS}"
@ -184,7 +184,7 @@ return 0
# EXPECTS: n/a # EXPECTS: n/a
# OUTPUTS: 0=not active/installed; 1=active/installed # OUTPUTS: 0=not active/installed; 1=active/installed
# RETURNS: 0=success; 1=error # RETURNS: 0=success; 1=error
# REQUIRES: n/a # REQUIRES: NetworkManager
function linux_has_nm function linux_has_nm
{ {
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set "${DEBUG_OPTS}" (( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set "${DEBUG_OPTS}"
@ -218,7 +218,7 @@ return 0
# EXPECTS: name of service [string] # EXPECTS: name of service [string]
# OUTPUTS: 0=not installed; 1=installed # OUTPUTS: 0=not installed; 1=installed
# RETURNS: 0=success; 1=error # RETURNS: 0=success; 1=error
# REQUIRES: n/a # REQUIRES: systemd
function linux_has_systemd_service function linux_has_systemd_service
{ {
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set "${DEBUG_OPTS}" (( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set "${DEBUG_OPTS}"
@ -230,6 +230,31 @@ _RC=$?
return ${_RC} return ${_RC}
} }
# -----------------------------------------------------------------------------
# @(#) FUNCTION: linux_exec_ssh()
# DOES: execute a shell command remotely via SSH
# EXPECTS: 1=options [string], 2=user [string], 3=host [string], 4=command [string]
# RETURNS: exit code of remote command
# OUTPUTS: STDOUT from SSH call
# REQUIRES: ssh command-line utility
function linux_exec_ssh
{
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set ${DEBUG_OPTS}
typeset _SSH_OPTS="${1}"
typeset _SSH_USER="${2}"
typeset _SSH_HOST="${3}"
typeset _SSH_COMMAND="${4}"
if [[ -z "${_SSH_USER}" || -z "${_SSH_HOST}" || -z "${_SSH_COMMAND}" ]]
then
return 255
fi
# shellcheck disable=SC2086
ssh ${_SSH_OPTS} -l ${_SSH_USER} ${_SSH_HOST} ${_SSH_COMMAND} 2>>${HC_STDERR_LOG} </dev/null
return $?
}
#****************************************************************************** #******************************************************************************
# END of script # END of script
#****************************************************************************** #******************************************************************************

476
sources/lib/platform/exadata/check_exadata_zfs_logs.sh Normal file
View File

@ -0,0 +1,476 @@
#!/usr/bin/env ksh
#******************************************************************************
# @(#) check_exadata_zfs_logs.sh
#******************************************************************************
# @(#) Copyright (C) 2019 by KUDOS BVBA (info@kudos.be). All rights reserved.
#
# This program is a free software; you can redistribute it and/or modify
# it under the same terms of the GNU General Public License as published by
# the Free Software Foundation; either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful, but WITHOUT
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
# FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details
#******************************************************************************
#
# DOCUMENTATION (MAIN)
# -----------------------------------------------------------------------------
# @(#) MAIN: check_exadata_zfs_logs
# DOES: see _show_usage()
# EXPECTS: see _show_usage()
# REQUIRES: data_comma2space(), dump_logs(), init_hc(), linux_exec_ssh(),
# log_hc(), warn()
#
# @(#) HISTORY:
# @(#) 2019-02-18: initial version [Patrick Van der Veken]
# -----------------------------------------------------------------------------
# DO NOT CHANGE THIS FILE UNLESS YOU KNOW WHAT YOU ARE DOING!
#******************************************************************************
# -----------------------------------------------------------------------------
function check_exadata_zfs_logs
{
# ------------------------- CONFIGURATION starts here -------------------------
typeset _CONFIG_FILE="${CONFIG_DIR}/$0.conf"
typeset _VERSION="2019-02-18" # YYYY-MM-DD
typeset _SUPPORTED_PLATFORMS="Linux" # uname -s match
# ------------------------- CONFIGURATION ends here ---------------------------
# set defaults
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set ${DEBUG_OPTS}
init_hc "$0" "${_SUPPORTED_PLATFORMS}" "${_VERSION}"
typeset _ARGS=$(data_comma2space "$*")
typeset _ARG=""
typeset _MSG=""
typeset _STC=0
typeset _AWK_RC=""
typeset _FILTER=""
typeset _FILTERS=""
typeset _CFG_HEALTHY=""
typeset _LOG_HEALTHY=0
typeset _CFG_SSH_KEY_FILE=""
typeset _CFG_SSH_OPTS=""
typeset _CFG_SSH_USER=""
typeset _CFG_ZFS_HOSTS=""
typeset _CFG_ZFS_HOST=""
typeset _LAST_LOG_ENTRY=""
typeset _MSG_DESC=""
typeset _MSG_ID=""
typeset _MSG_MODULE=""
typeset _MSG_PRIO=""
typeset _MSG_RESULT=""
typeset _MSG_TEXT=""
typeset _MSG_TYPE=""
typeset _NEW_LAST_LOG_ENTRY=""
typeset _SSH_BIN=""
typeset _SSH_OUTPUT=""
typeset _STATE_FILE=""
typeset _ZFS_SCRIPT=""
typeset _ZFS_LOG=""
typeset _ZFS_DATA=""
# handle arguments (originally comma-separated)
for _ARG in ${_ARGS}
do
case "${_ARG}" in
help)
_show_usage $0 ${_VERSION} ${_CONFIG_FILE} && return 0
;;
esac
done
# handle configuration file
[[ -n "${ARG_CONFIG_FILE}" ]] && _CONFIG_FILE="${ARG_CONFIG_FILE}"
if [[ ! -r ${_CONFIG_FILE} ]]
then
warn "unable to read configuration file at ${_CONFIG_FILE}"
return 1
fi
# read configuration values
_CFG_HEALTHY=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'log_healthy')
case "${_CFG_HEALTHY}" in
yes|YES|Yes)
_LOG_HEALTHY=1
;;
*)
# do not override hc_arg
(( _LOG_HEALTHY > 0 )) || _LOG_HEALTHY=0
;;
esac
_CFG_SSH_USER=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_user')
if [[ -z "${_CFG_SSH_USER}" ]]
then
_CFG_SSH_USER="root"
fi
_CFG_SSH_KEY_FILE=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_key_file')
_CFG_SSH_OPTS=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_opts')
# add quiet mode
_CFG_SSH_OPTS="${_CFG_SSH_OPTS} -q"
if [[ -n "${_CFG_SSH_KEY_FILE}" ]]
then
if [[ -r "${_CFG_SSH_KEY_FILE}" ]]
then
log "will use SSH key ${_CFG_SSH_KEY_FILE}"
_CFG_SSH_OPTS="${_CFG_SSH_OPTS} -i ${_CFG_SSH_KEY_FILE}"
else
warn "will use SSH key ${_CFG_SSH_KEY_FILE}, but file does not exist"
return 1
fi
fi
# log_healthy
(( ARG_LOG_HEALTHY > 0 )) && _LOG_HEALTHY=1
if (( _LOG_HEALTHY > 0 ))
then
if (( ARG_LOG > 0 ))
then
log "logging/showing passed health checks"
else
log "showing passed health checks (but not logging)"
fi
else
log "not logging/showing passed health checks"
fi
# check ssh
_SSH_BIN="$(which ssh 2>>${HC_STDERR_LOG})"
if [[ ! -x ${_SSH_BIN} || -z "${_SSH_BIN}" ]]
then
warn "SSH is not installed here"
return 1
fi
# gather ZFS hostnames
_CFG_ZFS_HOSTS=$(grep -i -E -e '^zfs:' ${_CONFIG_FILE} 2>/dev/null | cut -f2 -d':' 2>/dev/null | sort -u 2>/dev/null)
if [[ -z "${_CFG_ZFS_HOSTS}" ]]
then
warn "no monitoring rules defined in ${_CONFIG_FILE}"
return 1
fi
# gather ZFS log data
print "${_CFG_ZFS_HOSTS}" | while read -r _CFG_ZFS_HOST
do
# for which log(s)?
grep -i -E -e "^zfs:${_CFG_ZFS_HOST}:" ${_CONFIG_FILE} 2>/dev/null |\
while IFS=':' read -r _ _ _ZFS_LOG _FILTERS
do
# validate _ZFS_LOG & define script settings
case "${_ZFS_LOG}" in
alert|ALERT|Alert)
# define log query script -- DO NOT CHANGE --
_ZFS_SCRIPT="
script
run('maintenance logs select alert');
entries = list();
for (i = 0; i < entries.length; i++) {
try { run('select ' + entries[i]);
printf('%s|%s|%s|%s|%s\n', entries[i], get('timestamp'),
get('uuid'), get('description'), get('type'));
run('cd ..');
} catch (err) { }
}"
# validate _FILTERS
for _FILTER in $(data_comma2space "${_FILTERS}")
do
case "${_FILTER}" in
minor|MINOR|major|MAJOR|CRITICAL|critical)
:
;;
*)
warn "filter value is incorrect for ${_CFG_ZFS_HOST}/${_ZFS_LOG} in configuration file ${_CONFIG_FILE} "
return 1
;;
esac
done
;;
FLTLOG|fltlog|Fltlog)
# define log query script -- DO NOT CHANGE --
_ZFS_SCRIPT="
script
run('maintenance logs select fltlog');
entries = list();
for (i = 0; i < entries.length; i++) {
try { run('select ' + entries[i]);
printf('%s|%s|%s|%s|%s\n', entries[i], get('timestamp'),
get('uuid'), get('desc'), get('type'));
run('cd ..');
} catch (err) { }
}"
# validate _FILTERS
for _FILTER in $(data_comma2space "${_FILTERS}")
do
case "${_FILTER}" in
minor|MINOR|major|MAJOR|CRITICAL|critical)
:
;;
*)
warn "filter value is incorrect for ${_CFG_ZFS_HOST}/${_ZFS_LOG} in configuration file ${_CONFIG_FILE} "
return 1
;;
esac
done
;;
SCRK|scrk|Scrk)
# define log query script -- DO NOT CHANGE --
_ZFS_SCRIPT="
script
run('maintenance logs select scrk');
entries = list();
for (i = 0; i < entries.length; i++) {
try { run('select ' + entries[i]);
printf('%s|%s|%s|%s\n', entries[i], get('timestamp'),
get('description'), get('result'));
run('cd ..');
} catch (err) { }
}"
# validate _FILTERS
for _FILTER in $(data_comma2space "${_FILTERS}")
do
case "${_FILTER}" in
failed|FAILED|OK|ok)
:
;;
*)
warn "filter value is incorrect for ${_CFG_ZFS_HOST}/${_ZFS_LOG} in configuration file ${_CONFIG_FILE} "
return 1
;;
esac
done
;;
SYSTEM|system|System)
# define log query script -- DO NOT CHANGE --
_ZFS_SCRIPT="
script
run('maintenance logs select system');
entries = list();
for (i = 0; i < entries.length; i++) {
try { run('select ' + entries[i]);
printf('%s|%s|%s|%s|%s\n', entries[i], get('timestamp'),
get('module'), get('priority'), get('text'));
run('cd ..');
} catch (err) { }
}"
_FILTERS="error"
;;
*)
warn "log name value is incorrect for ${_CFG_ZFS_HOST}/${_ZFS_LOG} in configuration file ${_CONFIG_FILE} "
return 1
;;
esac
# check state file
_STATE_FILE="${STATE_PERM_DIR}/${_CFG_ZFS_HOST}.${_ZFS_LOG}.logs"
(( ARG_DEBUG > 0 )) && debug "checking/reading state file at ${_STATE_FILE}"
if [[ -r ${_STATE_FILE} ]]
then
_LAST_LOG_ENTRY=$(<"${_STATE_FILE}")
if [[ -z "${_LAST_LOG_ENTRY}" ]]
then
(( ARG_DEBUG > 0 )) && debug "no recorded last log entry for ${_CFG_ZFS_HOST}/${_ZFS_LOG}"
else
(( ARG_DEBUG > 0 )) && debug "recorded last log entry for ${_CFG_ZFS_HOST}/${_ZFS_LOG}: ${_LAST_LOG_ENTRY}"
fi
else
: >${_STATE_FILE}
(( $? > 0 )) && {
warn "failed to create new state file at ${_STATE_FILE}"
return 1
}
log "created new state file at ${_STATE_FILE}"
fi
(( ARG_DEBUG > 0 )) && debug "executing remote ZFS script on ${_CFG_ZFS_HOST} for log ${_ZFS_LOG}"
_SSH_OUTPUT=$(linux_exec_ssh "${_CFG_SSH_OPTS}" "${_CFG_SSH_USER}" "${_CFG_ZFS_HOST}" "${_ZFS_SCRIPT}" 2>>${HC_STDERR_LOG})
if (( $? > 0 )) || [[ -z "${_SSH_OUTPUT}" ]]
then
warn "unable to discover ${_ZFS_LOG} log data on ${_CFG_ZFS_HOST}"
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && dump_logs
continue
else
# parse log lines based on the log source they originated from
_ZFS_DATA=$(print -R "${_SSH_OUTPUT}" |\
awk -F'|' -v zfs_log="${_ZFS_LOG}" -v filters="${_FILTERS}" -v last_entry="${_LAST_LOG_ENTRY}" '
BEGIN {
found_last_entry = 0;
# build search needles/regexes
split (filters, filter_array, ",");
for (word in filter_array) {
filter_needle = filter_needle filter_array[word]"|";
}
# chop last "|"
gsub (/\|$/, "", filter_needle);
zfs_log = tolower(zfs_log);
}
{
# match log data against needle & $_LAST_LOG_ENTRY pointer
if (last_entry == "") {
# match against needle
if (zfs_log == "alert" || zfs_log == "fltlog") {
if (tolower ($5) ~ filter_needle) { print $0 };
}
if (zfs_log == "scrk" || zfs_log == "system") {
if (tolower ($4) ~ filter_needle) { print $0 };
}
} else {
# match against the $_LAST_LOG_ENTRY pointer
if ($1 ~ last_entry) {
found_last_entry = 1;
next;
}
if (found_last_entry > 0) {
# match against needle
if (zfs_log == "alert" || zfs_log == "fltlog") {
if (tolower ($5) ~ filter_needle) { print $0 };
}
if (zfs_log == "scrk" || zfs_log == "system") {
if (tolower ($4) ~ filter_needle) { print $0 };
}
}
}
}
END {
# when we had a log pointer at the start but did not
# encounter it, then we have a problematic situation (could
# be that the log pointer got rotated past the 100 lines query)
# flag this by RC=255 and reset the log pointer to last
# discovered entry (which is only a stopgap solution but the
# best we can come up with)
if (last_entry != "" && found_last_entry == 0) {
exit 255;
}
}' 2>>${HC_STDERR_LOG})
_AWK_RC=$?
# check and reports results
if (( _AWK_RC == 255 ))
then
warn "lost the current log pointer for ${_CFG_ZFS_HOST}/${_ZFS_LOG}"
# rewrite log pointer from the last log entry we discovered
_NEW_LAST_LOG_ENTRY=$(print "${_SSH_OUTPUT}" | tail -1 2>/dev/null | awk -F'|' '{ print $1 }' 2>/dev/null)
if [[ -n "${_NEW_LAST_LOG_ENTRY}" ]]
then
if (( ARG_LOG > 0 ))
then
warn "resetting the current log pointer for ${_CFG_ZFS_HOST}/${_ZFS_LOG} to ${_NEW_LAST_LOG_ENTRY}"
print "${_NEW_LAST_LOG_ENTRY}" >${_STATE_FILE} 2>>${HC_STDERR_LOG}
fi
else
# zero the state file
if (( ARG_LOG > 0 ))
then
warn "null resetting the current log pointer for ${_CFG_ZFS_HOST}/${_ZFS_LOG}"
: >${_STATE_FILE} 2>>${HC_STDERR_LOG}
fi
fi
continue
elif (( _AWK_RC > 0 ))
then
warn "unable to parse log data from ${_CFG_ZFS_HOST}/${_ZFS_LOG}"
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && dump_logs
return 1
else
if [[ -n "${_ZFS_DATA}" ]]
then
# save data to STDOUT
print "${_ZFS_DATA}" >>${HC_STDOUT_LOG}
# filter data based on logs
case "${_ZFS_LOG}" in
alert|ALERT|Alert)
print -R "${_ZFS_DATA}" | while IFS='|' read -r _MSG_ID _ _ _MSG_DESC _MSG_TYPE
do
_MSG="${_MSG_ID} (${_MSG_TYPE}) ${_MSG_DESC}"
log_hc "$0" 1 "${_CFG_ZFS_HOST}/${_ZFS_LOG}: ${_MSG}"
done
;;
FLTLOG|fltlog|Fltlog)
print -R "${_ZFS_DATA}" | while IFS='|' read -r _MSG_ID _ _ _MSG_DESC _MSG_TYPE
do
_MSG="${_MSG_ID} (${_MSG_TYPE}) ${_MSG_DESC}"
log_hc "$0" 1 "${_CFG_ZFS_HOST}/${_ZFS_LOG}: ${_MSG}"
done
;;
SCRK|scrk|Scrk)
print -R "${_ZFS_DATA}" | while IFS='|' read -r _MSG_ID _ _MSG_DESC _MSG_RESULT
do
if [[ "${_MSG_RESULT}" = "OK" ]]
then
_STC=0
else
_STC=1
fi
_MSG="${_MSG_ID} (${_MSG_RESULT}) ${_MSG_DESC}"
if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then
log_hc "$0" ${_STC} "${_CFG_ZFS_HOST}/${_ZFS_LOG}: ${_MSG}"
fi
done
;;
SYSTEM|system|System)
print -R "${_ZFS_DATA}" | while IFS='|' read -r _MSG_ID _ _MSG_MODULE _MSG_PRIO _MSG_TEXT
do
_MSG="${_MSG_ID} (${_MSG_PRIO}) ${_MSG_MODULE}: ${_MSG_TEXT}"
log_hc "$0" 1 "${_CFG_ZFS_HOST}/${_ZFS_LOG}: ${_MSG}"
done
;;
esac
else
if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then
_MSG="no (new) messages discovered from ${_CFG_ZFS_HOST}:/${_ZFS_LOG}"
log_hc "$0" 0 "${_MSG}"
fi
fi
# rewrite log pointer from the last log entry we discovered
_NEW_LAST_LOG_ENTRY=$(print "${_SSH_OUTPUT}" | tail -1 2>/dev/null | awk -F'|' '{ print $1 }' 2>/dev/null )
if (( ARG_LOG > 0 )) && [[ -n "${_NEW_LAST_LOG_ENTRY}" ]]
then
(( ARG_DEBUG > 0 )) && debug "updating last log entry for ${_CFG_ZFS_HOST}/${_ZFS_LOG} to ${_NEW_LAST_LOG_ENTRY}"
print "${_NEW_LAST_LOG_ENTRY}" >${_STATE_FILE} 2>>${HC_STDERR_LOG}
fi
fi
fi
done
done
return 0
}
# -----------------------------------------------------------------------------
function _show_usage
{
cat <<- EOT
NAME : $1
VERSION : $2
CONFIG : $3 with parameters:
log_healthy=<yes|no>
ssh_user=<ssh_user_account>
ssh_key_file=<ssh_private_key_file>
and formatted stanzas of:
zfs:<host_name>:<alert|fltlog|scrk|system>:<filters>
PURPOSE : checks the ZFS logs for (new) entries with particular alert level(s)
Following logs are supported (filters in brackets):
* alert (critical,major,minor)
* fltlog (critical,major,minor)
* system (error)
* scrk (failed)
CLI: zfs > maintenance > logs > select (log) > show
CAVEAT: plugin will use state files to track 'seen' messages. However each
check will only retrieve the default 100 last log entries. So it
is possible that log entries are lost between health checks (this
can be avoided by scheduling the check quicker than the likely
rotation time for 100 log entries).
LOG HEALTHY : Supported
EOT
return 0
}
#******************************************************************************
# END of script
#******************************************************************************

257
sources/lib/platform/exadata/check_exadata_zfs_services.sh Normal file
View File

@ -0,0 +1,257 @@
#!/usr/bin/env ksh
#******************************************************************************
# @(#) check_exadata_zfs_services.sh
#******************************************************************************
# @(#) Copyright (C) 2019 by KUDOS BVBA (info@kudos.be). All rights reserved.
#
# This program is a free software; you can redistribute it and/or modify
# it under the same terms of the GNU General Public License as published by
# the Free Software Foundation; either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful, but WITHOUT
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
# FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details
#******************************************************************************
#
# DOCUMENTATION (MAIN)
# -----------------------------------------------------------------------------
# @(#) MAIN: check_exadata_zfs_services
# DOES: see _show_usage()
# EXPECTS: see _show_usage()
# REQUIRES: data_comma2space(), dump_logs(), init_hc(), linux_exec_ssh(),
# log_hc(), warn()
#
# @(#) HISTORY:
# @(#) 2019-02-18: initial version [Patrick Van der Veken]
# -----------------------------------------------------------------------------
# DO NOT CHANGE THIS FILE UNLESS YOU KNOW WHAT YOU ARE DOING!
#******************************************************************************
# -----------------------------------------------------------------------------
function check_exadata_zfs_services
{
# ------------------------- CONFIGURATION starts here -------------------------
typeset _CONFIG_FILE="${CONFIG_DIR}/$0.conf"
typeset _VERSION="2019-02-18" # YYYY-MM-DD
typeset _SUPPORTED_PLATFORMS="Linux" # uname -s match
# usage query script -- DO NOT CHANGE --
# svc1:online
# svc2:disabled
typeset _ZFS_SCRIPT="
script
run('configuration services');
var svcs = children();
for (var i = 0; i < svcs.length; ++i) {
run(svcs[i]);
try {
printf('%0s:%s\n', svcs[i], get('<status>'));
} catch (err) { };
run('done');
}"
# ------------------------- CONFIGURATION ends here ---------------------------
# set defaults
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set ${DEBUG_OPTS}
init_hc "$0" "${_SUPPORTED_PLATFORMS}" "${_VERSION}"
typeset _ARGS=$(data_comma2space "$*")
typeset _ARG=""
typeset _MSG=""
typeset _STC=0
typeset _CFG_HEALTHY=""
typeset _LOG_HEALTHY=0
typeset _CFG_SSH_KEY_FILE=""
typeset _CFG_SSH_OPTS=""
typeset _CFG_SSH_USER=""
typeset _CFG_ZFS_HOSTS=""
typeset _CFG_ZFS_HOST=""
typeset _CFG_ZFS_LINE=""
typeset _SERVICE_NAME=""
typeset _SERVICE_STATE=""
typeset _SSH_BIN=""
typeset _SSH_OUTPUT=""
typeset _ZFS_DATA=""
# handle arguments (originally comma-separated)
for _ARG in ${_ARGS}
do
case "${_ARG}" in
help)
_show_usage $0 ${_VERSION} ${_CONFIG_FILE} && return 0
;;
esac
done
# handle configuration file
[[ -n "${ARG_CONFIG_FILE}" ]] && _CONFIG_FILE="${ARG_CONFIG_FILE}"
if [[ ! -r ${_CONFIG_FILE} ]]
then
warn "unable to read configuration file at ${_CONFIG_FILE}"
return 1
fi
# read configuration values
_CFG_HEALTHY=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'log_healthy')
case "${_CFG_HEALTHY}" in
yes|YES|Yes)
_LOG_HEALTHY=1
;;
*)
# do not override hc_arg
(( _LOG_HEALTHY > 0 )) || _LOG_HEALTHY=0
;;
esac
_CFG_SSH_USER=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_user')
if [[ -z "${_CFG_SSH_USER}" ]]
then
_CFG_SSH_USER="root"
fi
_CFG_SSH_KEY_FILE=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_key_file')
_CFG_SSH_OPTS=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_opts')
# add quiet mode
_CFG_SSH_OPTS="${_CFG_SSH_OPTS} -q"
if [[ -n "${_CFG_SSH_KEY_FILE}" ]]
then
if [[ -r "${_CFG_SSH_KEY_FILE}" ]]
then
log "will use SSH key ${_CFG_SSH_KEY_FILE}"
_CFG_SSH_OPTS="${_CFG_SSH_OPTS} -i ${_CFG_SSH_KEY_FILE}"
else
warn "will use SSH key ${_CFG_SSH_KEY_FILE}, but file does not exist"
return 1
fi
fi
# log_healthy
(( ARG_LOG_HEALTHY > 0 )) && _LOG_HEALTHY=1
if (( _LOG_HEALTHY > 0 ))
then
if (( ARG_LOG > 0 ))
then
log "logging/showing passed health checks"
else
log "showing passed health checks (but not logging)"
fi
else
log "not logging/showing passed health checks"
fi
# check ssh
_SSH_BIN="$(which ssh 2>>${HC_STDERR_LOG})"
if [[ ! -x ${_SSH_BIN} || -z "${_SSH_BIN}" ]]
then
warn "SSH is not installed here"
return 1
fi
# gather ZFS hostnames (for this we need at least one data line, possibly with wildcards)
_CFG_ZFS_HOSTS=$(grep -i -E -e '^zfs:' ${_CONFIG_FILE} 2>/dev/null | cut -f2 -d':' 2>/dev/null | sort -u 2>/dev/null)
if [[ -z "${_CFG_ZFS_HOSTS}" ]]
then
warn "no monitoring rules defined in ${_CONFIG_FILE}"
return 1
fi
# gather ZFS usage data
print "${_CFG_ZFS_HOSTS}" | while read -r _CFG_ZFS_HOST
do
(( ARG_DEBUG > 0 )) && debug "executing remote ZFS script on ${_CFG_ZFS_HOST}"
_SSH_OUTPUT=$(linux_exec_ssh "${_CFG_SSH_OPTS}" "${_CFG_SSH_USER}" "${_CFG_ZFS_HOST}" "${_ZFS_SCRIPT}" 2>>${HC_STDERR_LOG})
if (( $? > 0 )) || [[ -z "${_SSH_OUTPUT}" ]]
then
warn "unable to discover services data on ${_CFG_ZFS_HOST}"
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && dump_logs
continue
else
# mangle SSH output by prefixing with hostname
print "${_SSH_OUTPUT}" | while read -r _SSH_LINE
do
if [[ -z "${_ZFS_DATA}" ]]
then
_ZFS_DATA="${_CFG_ZFS_HOST}:${_SSH_LINE}"
else
# shellcheck disable=SC1117
_ZFS_DATA="${_ZFS_DATA}\n${_CFG_ZFS_HOST}:${_SSH_LINE}"
fi
done
fi
done
# process usage status data
if [[ -z "${_ZFS_DATA}" ]]
then
_MSG="did not discover any ZFS services data"
_STC=2
if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then
log_hc "$0" ${_STC} "${_MSG}"
fi
return 1
fi
grep -E -e '^zfs:' ${_CONFIG_FILE} 2>/dev/null |\
while IFS=':' read -r _ _CFG_ZFS_HOST _CFG_SERVICE_NAME _CFG_SERVICE_STATE
do
if [[ -z "${_CFG_SERVICE_NAME}" ]]
then
warn "value of <service_name> for ${_CFG_ZFS_HOST} is not defined in configuration file ${_CONFIG_FILE}"
continue
fi
case "${_CFG_SERVICE_STATE}" in
online|ONLINE|Online|disabled|DISABLED|Disabled)
:
;;
*)
warn "value of <service_state> for ${_CFG_ZFS_HOST}/${_CFG_SERVICE_NAME} is not correct in configuration file ${_CONFIG_FILE}"
continue
esac
(( ARG_DEBUG > 0 )) && debug "parsing services data for service: ${_CFG_ZFS_HOST}/${_CFG_SERVICE_NAME}"
# perform check
_SERVICE_STATE=$(print "${_ZFS_DATA}" | grep -E -e "^${_CFG_ZFS_HOST}:${_CFG_SERVICE_NAME}:" 2>/dev/null | cut -f3 -d':' 2>/dev/null)
if [[ -n "${_SERVICE_STATE}" ]]
then
if [[ $(data_lc "${_SERVICE_STATE}") != $(data_lc "${_CFG_SERVICE_STATE}") ]]
then
_MSG="state of ${_CFG_ZFS_HOST}/${_CFG_SERVICE_NAME} is incorrect (${_SERVICE_STATE}!=${_CFG_SERVICE_STATE})"
_STC=1
else
_MSG="state of ${_CFG_ZFS_HOST}/${_CFG_SERVICE_NAME} is correct (${_SERVICE_STATE}=${_CFG_SERVICE_STATE})"
_STC=0
fi
if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then
log_hc "$0" ${_STC} "${_MSG}"
fi
else
warn "did not find services data for ${_CFG_ZFS_HOST}/${_CFG_SERVICE_NAME}"
continue
fi
done
return 0
}
# -----------------------------------------------------------------------------
function _show_usage
{
cat <<- EOT
NAME : $1
VERSION : $2
CONFIG : $3 with parameters:
log_healthy=<yes|no>
ssh_user=<ssh_user_account>
ssh_key_file=<ssh_private_key_file>
and formatted stanzas of:
zfs:<host_name>:<service_name>:<service_state>
PURPOSE : Checks the state of services for the configured ZFS hosts/shares
CLI: zfs > status > services > show
LOG HEALTHY : Supported
EOT
return 0
}
#******************************************************************************
# END of script
#******************************************************************************

328
sources/lib/platform/exadata/check_exadata_zfs_share_replication.sh Normal file
View File

@ -0,0 +1,328 @@
#!/usr/bin/env ksh
#******************************************************************************
# @(#) check_exadata_zfs_share_replication.sh
#******************************************************************************
# @(#) Copyright (C) 2019 by KUDOS BVBA (info@kudos.be). All rights reserved.
#
# This program is a free software; you can redistribute it and/or modify
# it under the same terms of the GNU General Public License as published by
# the Free Software Foundation; either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful, but WITHOUT
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
# FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details
#******************************************************************************
#
# DOCUMENTATION (MAIN)
# -----------------------------------------------------------------------------
# @(#) MAIN: check_exadata_zfs_share_replication
# DOES: see _show_usage()
# EXPECTS: see _show_usage()
# REQUIRES: data_comma2space(), dump_logs(), init_hc(), linux_exec_ssh(),
# log_hc(), warn()
#
# @(#) HISTORY:
# @(#) 2019-02-18: initial version [Patrick Van der Veken]
# -----------------------------------------------------------------------------
# DO NOT CHANGE THIS FILE UNLESS YOU KNOW WHAT YOU ARE DOING!
#******************************************************************************
# -----------------------------------------------------------------------------
function check_exadata_zfs_share_replication
{
# ------------------------- CONFIGURATION starts here -------------------------
typeset _CONFIG_FILE="${CONFIG_DIR}/$0.conf"
typeset _VERSION="2019-02-18" # YYYY-MM-DD
typeset _SUPPORTED_PLATFORMS="Linux" # uname -s match
# replication query script -- DO NOT CHANGE --
# prj1/share1:true:idle:success:111
# prj2/share2:true:idle:success:51
typeset _ZFS_SCRIPT="
script
run('shares replication actions');
actions = list();
for (i = 0; i < actions.length; i++) {
try { run('select ' + actions[i]);
printf('%s:%s:%s:%s\n', get('replication_of'),
get('enabled'),
get('last_result'),
get('replica_lag'));
run('cd ..');
} catch (err) { }
}"
# ------------------------- CONFIGURATION ends here ---------------------------
# set defaults
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set ${DEBUG_OPTS}
init_hc "$0" "${_SUPPORTED_PLATFORMS}" "${_VERSION}"
typeset _ARGS=$(data_comma2space "$*")
typeset _ARG=""
typeset _MSG=""
typeset _STC=0
typeset _CFG_HEALTHY=""
typeset _LOG_HEALTHY=0
typeset _CFG_MAX_REPLICA_LAG=""
typeset _CFG_REPLICATION_LAG=""
typeset _CFG_SSH_KEY_FILE=""
typeset _CFG_SSH_OPTS=""
typeset _CFG_SSH_USER=""
typeset _CFG_ZFS_HOSTS=""
typeset _CFG_ZFS_HOST=""
typeset _CFG_ZFS_LINE=""
typeset _REPLICATION_ENABLED=""
typeset _REPLICATION_LAG=""
typeset _REPLICATION_RESULT=""
typeset _SSH_BIN=""
typeset _SSH_OUTPUT=""
typeset _ZFS_DATA=""
# handle arguments (originally comma-separated)
for _ARG in ${_ARGS}
do
case "${_ARG}" in
help)
_show_usage $0 ${_VERSION} ${_CONFIG_FILE} && return 0
;;
esac
done
# handle configuration file
[[ -n "${ARG_CONFIG_FILE}" ]] && _CONFIG_FILE="${ARG_CONFIG_FILE}"
if [[ ! -r ${_CONFIG_FILE} ]]
then
warn "unable to read configuration file at ${_CONFIG_FILE}"
return 1
fi
# read configuration values
_CFG_HEALTHY=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'log_healthy')
case "${_CFG_HEALTHY}" in
yes|YES|Yes)
_LOG_HEALTHY=1
;;
*)
# do not override hc_arg
(( _LOG_HEALTHY > 0 )) || _LOG_HEALTHY=0
;;
esac
_CFG_SSH_USER=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_user')
if [[ -z "${_CFG_SSH_USER}" ]]
then
_CFG_SSH_USER="root"
fi
_CFG_SSH_KEY_FILE=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_key_file')
_CFG_SSH_OPTS=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_opts')
# add quiet mode
_CFG_SSH_OPTS="${_CFG_SSH_OPTS} -q"
if [[ -n "${_CFG_SSH_KEY_FILE}" ]]
then
if [[ -r "${_CFG_SSH_KEY_FILE}" ]]
then
log "will use SSH key ${_CFG_SSH_KEY_FILE}"
_CFG_SSH_OPTS="${_CFG_SSH_OPTS} -i ${_CFG_SSH_KEY_FILE}"
else
warn "will use SSH key ${_CFG_SSH_KEY_FILE}, but file does not exist"
return 1
fi
fi
_CFG_MAX_REPLICA_LAG=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'max_replica_lag')
if [[ -z "${_CFG_MAX_REPLICA_LAG}" ]]
then
# default
_CFG_MAX_REPLICA_LAG=90
fi
# log_healthy
(( ARG_LOG_HEALTHY > 0 )) && _LOG_HEALTHY=1
if (( _LOG_HEALTHY > 0 ))
then
if (( ARG_LOG > 0 ))
then
log "logging/showing passed health checks"
else
log "showing passed health checks (but not logging)"
fi
else
log "not logging/showing passed health checks"
fi
# check ssh
_SSH_BIN="$(which ssh 2>>${HC_STDERR_LOG})"
if [[ ! -x ${_SSH_BIN} || -z "${_SSH_BIN}" ]]
then
warn "SSH is not installed here"
return 1
fi
# gather ZFS hostnames
_CFG_ZFS_HOSTS=$(grep -i -E -e '^zfs:' ${_CONFIG_FILE} 2>/dev/null | cut -f2 -d':' 2>/dev/null | sort -u 2>/dev/null)
if [[ -z "${_CFG_ZFS_HOSTS}" ]]
then
warn "no monitoring rules defined in ${_CONFIG_FILE}"
return 1
fi
# gather ZFS replication data
print "${_CFG_ZFS_HOSTS}" | while read -r _CFG_ZFS_HOST
do
(( ARG_DEBUG > 0 )) && debug "executing remote ZFS script on ${_CFG_ZFS_HOST}"
_SSH_OUTPUT=$(linux_exec_ssh "${_CFG_SSH_OPTS}" "${_CFG_SSH_USER}" "${_CFG_ZFS_HOST}" "${_ZFS_SCRIPT}" 2>>${HC_STDERR_LOG})
if (( $? > 0 )) || [[ -z "${_SSH_OUTPUT}" ]]
then
warn "unable to discover replication data on ${_CFG_ZFS_HOST}"
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && dump_logs
continue
else
# mangle SSH output by prefixing with hostname
print "${_SSH_OUTPUT}" | while read -r _SSH_LINE
do
if [[ -z "${_ZFS_DATA}" ]]
then
_ZFS_DATA="${_CFG_ZFS_HOST}:${_SSH_LINE}"
else
# shellcheck disable=SC1117
_ZFS_DATA="${_ZFS_DATA}\n${_CFG_ZFS_HOST}:${_SSH_LINE}"
fi
done
fi
done
# process replication status data
if [[ -z "${_ZFS_DATA}" ]]
then
_MSG="did not discover any ZFS replication data"
_STC=2
if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then
log_hc "$0" ${_STC} "${_MSG}"
fi
return 1
fi
print "${_ZFS_DATA}" | while IFS=':' read -r _ZFS_HOST _REPLICATION_NAME _REPLICATION_ENABLED _REPLICATION_RESULT _REPLICATION_LAG
do
(( ARG_DEBUG > 0 )) && debug "parsing replication data for share: ${_ZFS_HOST}:${_REPLICATION_NAME}"
_CFG_REPLICATION_ENABLED=""
_CFG_REPLICATION_RESULT=""
_CFG_REPLICATION_LAG=""
# which values to use (general or custom?), keep in mind wildcards (custom will overrule wildcard entry)
_CFG_ZFS_LINE=$(grep -E -e "^zfs:${_ZFS_HOST}:[*]:" ${_CONFIG_FILE} 2>/dev/null)
if [[ -n "${_CFG_ZFS_LINE}" ]]
then
(( ARG_DEBUG > 0 )) && debug "found wilcard definition for ${_ZFS_HOST} in configuration file ${_CONFIG_FILE}"
_CFG_REPLICATION_ENABLED=$(print "${_CFG_ZFS_LINE}" | cut -f4 -d':' 2>/dev/null)
_CFG_REPLICATION_RESULT=$(print "${_CFG_ZFS_LINE}" | cut -f5 -d':' 2>/dev/null)
_CFG_REPLICATION_LAG=$(print "${_CFG_ZFS_LINE}" | cut -f6 -d':' 2>/dev/null)
# null value means general threshold
if [[ -z "${_CFG_REPLICATION_LAG}" ]]
then
(( ARG_DEBUG > 0 )) && debug "found empty lag threshold for ${_ZFS_HOST}, using general threshold"
_CFG_REPLICATION_LAG=${_CFG_MAX_REPLICATION_LAG}
fi
fi
_CFG_ZFS_LINE=$(grep -E -e "^zfs:${_ZFS_HOST}:${_REPLICATION_NAME}:" ${_CONFIG_FILE} 2>/dev/null)
if [[ -n "${_CFG_ZFS_LINE}" ]]
then
(( ARG_DEBUG > 0 )) && debug "found custom definition for ${_ZFS_HOST}:${_REPLICATION_NAME} in configuration file ${_CONFIG_FILE}"
_CFG_REPLICATION_ENABLED=$(print "${_CFG_ZFS_LINE}" | cut -f4 -d':' 2>/dev/null)
_CFG_REPLICATION_RESULT=$(print "${_CFG_ZFS_LINE}" | cut -f5 -d':' 2>/dev/null)
_CFG_REPLICATION_LAG=$(print "${_CFG_ZFS_LINE}" | cut -f6 -d':' 2>/dev/null)
# null value means general threshold
if [[ -z "${_CFG_REPLICATION_LAG}" ]]
then
(( ARG_DEBUG > 0 )) && debug "found empty lag threshold for ${_ZFS_HOST}, using general threshold"
_CFG_REPLICATION_LAG=${_CFG_MAX_REPLICATION_LAG}
fi
fi
if [[ -n "${_CFG_REPLICATION_LAG}" ]]
then
data_is_numeric "${_CFG_REPLICATION_LAG}"
if (( $? > 0 ))
then
warn "value for <max_replication_lag> is not numeric in configuration file ${_CONFIG_FILE}"
continue
fi
# zero value means disabled check
if (( _CFG_REPLICATION_LAG == 0 ))
then
(( ARG_DEBUG > 0 )) && debug "found zero lag threshold, disabling check"
continue
fi
else
(( ARG_DEBUG > 0 )) && debug "no custom space threshold for ${_ZFS_HOST}:${_REPLICATION_NAME}, using general threshold"
_CFG_REPLICATION_LAG=${_CFG_MAX_REPLICATION_LAG}
fi
# fixed defaults if missing
[[ -z "${_CFG_REPLICATION_ENABLED}" || "${_CFG_REPLICATION_ENABLED}" = '*' ]] && _CFG_REPLICATION_ENABLED="true"
[[ -z "${_CFG_REPLICATION_RESULT}" || "${_CFG_REPLICATION_RESULT}" = '*' ]] && _CFG_REPLICATION_RESULT="success"
# perform checks
# check replication enabled state (active or not?)
if [[ $(data_lc "${_REPLICATION_ENABLED}") != $(data_lc "${_CFG_REPLICATION_ENABLED}") ]]
then
_MSG="state for ${_ZFS_HOST}:${_REPLICATION_NAME} is incorrect [${_REPLICATION_ENABLED}!=${_CFG_REPLICATION_ENABLED}]"
_STC=1
else
_MSG="state for ${_ZFS_HOST}:${_REPLICATION_NAME} is correct [${_REPLICATION_ENABLED}=${_CFG_REPLICATION_ENABLED}]"
_STC=0
fi
if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then
log_hc "$0" ${_STC} "${_MSG}" "${_REPLICATION_ENABLED}" "${_CFG_REPLICATION_ENABLED}"
fi
# check replication last result (success or not?)
if [[ $(data_lc "${_REPLICATION_RESULT}") != $(data_lc "${_CFG_REPLICATION_RESULT}") ]]
then
_MSG="result for ${_ZFS_HOST}:${_REPLICATION_NAME} is incorrect [${_REPLICATION_RESULT}!=${_CFG_REPLICATION_RESULT}]"
_STC=1
else
_MSG="result for ${_ZFS_HOST}:${_REPLICATION_NAME} is correct [${_REPLICATION_RESULT}=${_CFG_REPLICATION_RESULT}]"
_STC=0
fi
if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then
log_hc "$0" ${_STC} "${_MSG}" "${_REPLICATION_RESULT}" "${_CFG_REPLICATION_RESULT}"
fi
# check replication lag
if (( _REPLICATION_LAG > _CFG_REPLICATION_LAG ))
then
_MSG="lag for ${_ZFS_HOST}:${_REPLICATION_NAME} is too big [${_REPLICATION_LAG}>${_CFG_REPLICATION_LAG}]"
_STC=1
else
_MSG="lag for ${_ZFS_HOST}:${_REPLICATION_NAME} is OK [${_REPLICATION_LAG}<=${_CFG_REPLICATION_LAG}]"
_STC=0
fi
if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then
log_hc "$0" ${_STC} "${_MSG}" "${_REPLICATION_LAG}" "${_CFG_REPLICATION_LAG}"
fi
done
return 0
}
# -----------------------------------------------------------------------------
function _show_usage
{
cat <<- EOT
NAME : $1
VERSION : $2
CONFIG : $3 with parameters:
log_healthy=<yes|no>
ssh_user=<ssh_user_account>
ssh_key_file=<ssh_private_key_file>
max_replication_lag=<general_max_replication>
and formatted stanzas of:
zfs:<host_name>:<replication_name>:<replication_enabled>:<replication_result>:<max_replication_lag>
PURPOSE : Checks the replication state, sync status and maximum lag of the configured ZFS hosts/shares
CLI: zfs > shares > replications > packages > select (action) > show
LOG HEALTHY : Supported
EOT
return 0
}
#******************************************************************************
# END of script
#******************************************************************************

300
sources/lib/platform/exadata/check_exadata_zfs_share_usage.sh Normal file
View File

@ -0,0 +1,300 @@
#!/usr/bin/env ksh
#******************************************************************************
# @(#) check_exadata_zfs_share_usage.sh
#******************************************************************************
# @(#) Copyright (C) 2019 by KUDOS BVBA (info@kudos.be). All rights reserved.
#
# This program is a free software; you can redistribute it and/or modify
# it under the same terms of the GNU General Public License as published by
# the Free Software Foundation; either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful, but WITHOUT
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
# FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details
#******************************************************************************
#
# DOCUMENTATION (MAIN)
# -----------------------------------------------------------------------------
# @(#) MAIN: check_exadata_zfs_share_usage
# DOES: see _show_usage()
# EXPECTS: see _show_usage()
# REQUIRES: data_comma2space(), dump_logs(), init_hc(), linux_exec_ssh(),
# log_hc(), warn()
#
# @(#) HISTORY:
# @(#) 2019-02-18: initial version [Patrick Van der Veken]
# -----------------------------------------------------------------------------
# DO NOT CHANGE THIS FILE UNLESS YOU KNOW WHAT YOU ARE DOING!
#******************************************************************************
# -----------------------------------------------------------------------------
function check_exadata_zfs_share_usage
{
# ------------------------- CONFIGURATION starts here -------------------------
typeset _CONFIG_FILE="${CONFIG_DIR}/$0.conf"
typeset _VERSION="2019-02-18" # YYYY-MM-DD
typeset _SUPPORTED_PLATFORMS="Linux" # uname -s match
# usage query script -- DO NOT CHANGE --
# prj1:share1:16
# prj2:share1:85
typeset _ZFS_SCRIPT="
script
run('shares');
projects = list();
for (i = 0; i < projects.length; i++) {
try { run('select ' + projects[i]);
shares = list();
for (j = 0; j < shares.length; j++) {
try { run('select ' + shares[j]);
printf('%s:%s:%d\n', projects[i], shares[j],
get('space_data')/get('space_available')*100);
run('cd ..');
} catch (err) { }
}
run('cd ..');
} catch (err) {
throw ('unexpected error occurred');
}
}"
# ------------------------- CONFIGURATION ends here ---------------------------
# set defaults
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && set ${DEBUG_OPTS}
init_hc "$0" "${_SUPPORTED_PLATFORMS}" "${_VERSION}"
typeset _ARGS=$(data_comma2space "$*")
typeset _ARG=""
typeset _MSG=""
typeset _STC=0
typeset _CFG_HEALTHY=""
typeset _LOG_HEALTHY=0
typeset _CFG_MAX_SPACE_USAGE=""
typeset _CFG_SSH_KEY_FILE=""
typeset _CFG_SSH_OPTS=""
typeset _CFG_SSH_USER=""
typeset _CFG_SPACE_THRESHOLD=""
typeset _CFG_ZFS_HOSTS=""
typeset _CFG_ZFS_HOST=""
typeset _CFG_ZFS_LINE=""
typeset _PROJECT_NAME=""
typeset _SHARE_NAME=""
typeset _SPACE_USAGE=""
typeset _SSH_BIN=""
typeset _SSH_OUTPUT=""
typeset _ZFS_DATA=""
# handle arguments (originally comma-separated)
for _ARG in ${_ARGS}
do
case "${_ARG}" in
help)
_show_usage $0 ${_VERSION} ${_CONFIG_FILE} && return 0
;;
esac
done
# handle configuration file
[[ -n "${ARG_CONFIG_FILE}" ]] && _CONFIG_FILE="${ARG_CONFIG_FILE}"
if [[ ! -r ${_CONFIG_FILE} ]]
then
warn "unable to read configuration file at ${_CONFIG_FILE}"
return 1
fi
# read configuration values
_CFG_HEALTHY=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'log_healthy')
case "${_CFG_HEALTHY}" in
yes|YES|Yes)
_LOG_HEALTHY=1
;;
*)
# do not override hc_arg
(( _LOG_HEALTHY > 0 )) || _LOG_HEALTHY=0
;;
esac
_CFG_SSH_USER=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_user')
if [[ -z "${_CFG_SSH_USER}" ]]
then
_CFG_SSH_USER="root"
fi
_CFG_SSH_KEY_FILE=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_key_file')
_CFG_SSH_OPTS=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'ssh_opts')
# add quiet mode
_CFG_SSH_OPTS="${_CFG_SSH_OPTS} -q"
if [[ -n "${_CFG_SSH_KEY_FILE}" ]]
then
if [[ -r "${_CFG_SSH_KEY_FILE}" ]]
then
log "will use SSH key ${_CFG_SSH_KEY_FILE}"
_CFG_SSH_OPTS="${_CFG_SSH_OPTS} -i ${_CFG_SSH_KEY_FILE}"
else
warn "will use SSH key ${_CFG_SSH_KEY_FILE}, but file does not exist"
return 1
fi
fi
_CFG_MAX_SPACE_USAGE=$(_CONFIG_FILE="${_CONFIG_FILE}" data_get_lvalue_from_config 'max_space_usage')
if [[ -z "${_CFG_MAX_SPACE_USAGE}" ]]
then
# default
_CFG_MAX_SPACE_USAGE=90
fi
# log_healthy
(( ARG_LOG_HEALTHY > 0 )) && _LOG_HEALTHY=1
if (( _LOG_HEALTHY > 0 ))
then
if (( ARG_LOG > 0 ))
then
log "logging/showing passed health checks"
else
log "showing passed health checks (but not logging)"
fi
else
log "not logging/showing passed health checks"
fi
# check ssh
_SSH_BIN="$(which ssh 2>>${HC_STDERR_LOG})"
if [[ ! -x ${_SSH_BIN} || -z "${_SSH_BIN}" ]]
then
warn "SSH is not installed here"
return 1
fi
# gather ZFS hostnames (for this we need at least one data line, possibly with wildcards)
_CFG_ZFS_HOSTS=$(grep -i -E -e '^zfs:' ${_CONFIG_FILE} 2>/dev/null | cut -f2 -d':' 2>/dev/null | sort -u 2>/dev/null)
if [[ -z "${_CFG_ZFS_HOSTS}" ]]
then
warn "no monitoring rules defined in ${_CONFIG_FILE}"
return 1
fi
# gather ZFS usage data
print "${_CFG_ZFS_HOSTS}" | while read -r _CFG_ZFS_HOST
do
(( ARG_DEBUG > 0 )) && debug "executing remote ZFS script on ${_CFG_ZFS_HOST}"
_SSH_OUTPUT=$(linux_exec_ssh "${_CFG_SSH_OPTS}" "${_CFG_SSH_USER}" "${_CFG_ZFS_HOST}" "${_ZFS_SCRIPT}" 2>>${HC_STDERR_LOG})
if (( $? > 0 )) || [[ -z "${_SSH_OUTPUT}" ]]
then
warn "unable to discover usage data on ${_CFG_ZFS_HOST}"
(( ARG_DEBUG > 0 && ARG_DEBUG_LEVEL > 0 )) && dump_logs
continue
else
# mangle SSH output by prefixing with hostname
print "${_SSH_OUTPUT}" | while read -r _SSH_LINE
do
if [[ -z "${_ZFS_DATA}" ]]
then
_ZFS_DATA="${_CFG_ZFS_HOST}:${_SSH_LINE}"
else
# shellcheck disable=SC1117
_ZFS_DATA="${_ZFS_DATA}\n${_CFG_ZFS_HOST}:${_SSH_LINE}"
fi
done
fi
done
# process usage status data
if [[ -z "${_ZFS_DATA}" ]]
then
_MSG="did not discover any ZFS share data"
_STC=2
if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then
log_hc "$0" ${_STC} "${_MSG}"
fi
return 1
fi
print "${_ZFS_DATA}" | while IFS=':' read -r _ZFS_HOST _PROJECT_NAME _SHARE_NAME _SPACE_USAGE
do
(( ARG_DEBUG > 0 )) && debug "parsing space data for share: ${_ZFS_HOST}:${_PROJECT_NAME}/${_SHARE_NAME}"
_CFG_SPACE_THRESHOLD=""
# which threshold to use (general or custom?), keep in mind wildcards (custom will overrule wildcard entry)
_CFG_ZFS_LINE=$(grep -E -e "^zfs:${_ZFS_HOST}:[*]:[*]:" ${_CONFIG_FILE} 2>/dev/null)
if [[ -n "${_CFG_ZFS_LINE}" ]]
then
(( ARG_DEBUG > 0 )) && debug "found wilcard definition for ${_ZFS_HOST} in configuration file ${_CONFIG_FILE}"
_CFG_SPACE_THRESHOLD=$(print "${_CFG_ZFS_LINE}" | cut -f5 -d':' 2>/dev/null)
# null value means general threshold
if [[ -z "${_CFG_SPACE_THRESHOLD}" ]]
then
(( ARG_DEBUG > 0 )) && debug "found empty space threshold for ${_ZFS_HOST}, using general threshold"
_CFG_SPACE_THRESHOLD=${_CFG_MAX_SPACE_USAGE}
fi
fi
_CFG_ZFS_LINE=$(grep -E -e "^zfs:${_ZFS_HOST}:${_PROJECT_NAME}:${_SHARE_NAME}:" ${_CONFIG_FILE} 2>/dev/null)
if [[ -n "${_CFG_ZFS_LINE}" ]]
then
(( ARG_DEBUG > 0 )) && debug "found custom definition for ${_ZFS_HOST}:${_PROJECT_NAME}/${_SHARE_NAME} in configuration file ${_CONFIG_FILE}"
_CFG_SPACE_THRESHOLD=$(print "${_CFG_ZFS_LINE}" | cut -f5 -d':' 2>/dev/null)
# null value means general threshold
if [[ -z "${_CFG_SPACE_THRESHOLD}" ]]
then
(( ARG_DEBUG > 0 )) && debug "found empty space threshold for ${_ZFS_HOST}:${_PROJECT_NAME}:${_SHARE_NAME}, using general threshold"
_CFG_SPACE_THRESHOLD=${_CFG_MAX_SPACE_USAGE}
fi
fi
if [[ -n "${_CFG_SPACE_THRESHOLD}" ]]
then
data_is_numeric "${_CFG_SPACE_THRESHOLD}"
if (( $? > 0 ))
then
warn "value for <max_space_threshold> is not numeric in configuration file ${_CONFIG_FILE}"
continue
fi
# zero value means disabled check
if (( _CFG_SPACE_THRESHOLD == 0 ))
then
(( ARG_DEBUG > 0 )) && debug "found zero space threshold, disabling check"
continue
fi
else
(( ARG_DEBUG > 0 )) && debug "no custom space threshold for ${_ZFS_HOST}:${_PROJECT_NAME}:${_SHARE_NAME}, using general threshold"
_CFG_SPACE_THRESHOLD=${_CFG_MAX_SPACE_USAGE}
fi
# perform check
if (( _SPACE_USAGE > _CFG_SPACE_THRESHOLD ))
then
_MSG="${_ZFS_HOST}:${_PROJECT_NAME}/${_SHARE_NAME} exceeds its space threshold (${_SPACE_USAGE}%>${_CFG_SPACE_THRESHOLD}%)"
_STC=1
else
_MSG="${_ZFS_HOST}:${_PROJECT_NAME}/${_SHARE_NAME} does not exceed its space threshold (${_SPACE_USAGE}%>${_CFG_SPACE_THRESHOLD}%)"
_STC=0
fi
if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then
log_hc "$0" ${_STC} "${_MSG}" "${_SPACE_USAGE}" "${_CFG_SPACE_THRESHOLD}"
fi
done
return 0
}
# -----------------------------------------------------------------------------
function _show_usage
{
cat <<- EOT
NAME : $1
VERSION : $2
CONFIG : $3 with parameters:
log_healthy=<yes|no>
ssh_user=<ssh_user_account>
ssh_key_file=<ssh_private_key_file>
max_space_usage=<general_max_space_treshold>
and formatted stanzas of:
zfs:<host_name>:<replication_name>:<replication_enabled>:<replication_result>:<max_space_threshold>
PURPOSE : Checks the space usage for the configured ZFS hosts/shares
CLI: zfs > shares > select (project) > (select share) > show
LOG HEALTHY : Supported
EOT
return 0
}
#******************************************************************************
# END of script
#******************************************************************************

8
sources/lib/platform/hp-ux/check_hpux_fs_usage.sh
View File

@ -26,6 +26,7 @@
# @(#) 2019-01-27: regex fix [Patrick Van der Veken] # @(#) 2019-01-27: regex fix [Patrick Van der Veken]
# @(#) 2019-01-30: refactored to support custom defintions with all # @(#) 2019-01-30: refactored to support custom defintions with all
# filesystems check [Patrick Van der Veken] # filesystems check [Patrick Van der Veken]
# @(#) 2019-02-18: fixes + help update
# ----------------------------------------------------------------------------- # -----------------------------------------------------------------------------
# DO NOT CHANGE THIS FILE UNLESS YOU KNOW WHAT YOU ARE DOING! # DO NOT CHANGE THIS FILE UNLESS YOU KNOW WHAT YOU ARE DOING!
#****************************************************************************** #******************************************************************************
@ -35,7 +36,7 @@ function check_hpux_fs_usage
{ {
# ------------------------- CONFIGURATION starts here ------------------------- # ------------------------- CONFIGURATION starts here -------------------------
typeset _CONFIG_FILE="${CONFIG_DIR}/$0.conf" typeset _CONFIG_FILE="${CONFIG_DIR}/$0.conf"
typeset _VERSION="2019-01-30" # YYYY-MM-DD typeset _VERSION="2019-02-18" # YYYY-MM-DD
typeset _SUPPORTED_PLATFORMS="HP-UX" # uname -s match typeset _SUPPORTED_PLATFORMS="HP-UX" # uname -s match
# ------------------------- CONFIGURATION ends here --------------------------- # ------------------------- CONFIGURATION ends here ---------------------------
@ -215,7 +216,7 @@ then
fi fi
if (( _LOG_HEALTHY > 0 || _STC > 0 )) if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then then
log_hc "$0" ${_STC} "${_MSG}" ${_INODES_USAGE} ${_CFG_INODES_THRESHOLD} log_hc "$0" ${_STC} "${_MSG}" "${_INODES_USAGE}" "${_CFG_INODES_THRESHOLD}"
fi fi
done done
fi fi
@ -276,7 +277,7 @@ then
fi fi
if (( _LOG_HEALTHY > 0 || _STC > 0 )) if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then then
log_hc "$0" ${_STC} "${_MSG}" ${_SPACE_USAGE} ${_CFG_SPACE_THRESHOLD} log_hc "$0" ${_STC} "${_MSG}" "${_SPACE_USAGE}" "${_CFG_SPACE_THRESHOLD}"
fi fi
done done
fi fi
@ -293,6 +294,7 @@ VERSION : $2
CONFIG : $3 with formatted stanzas (optional): CONFIG : $3 with formatted stanzas (optional):
fs:<fs_name>:<max_inodes_usage_in_%>:<max_space_usage_in%> fs:<fs_name>:<max_inodes_usage_in_%>:<max_space_usage_in%>
Other options: Other options:
log_healthy=<yes|no>
check_inodes_usage=<yes|no> check_inodes_usage=<yes|no>
check_space_usage=<yes|no> check_space_usage=<yes|no>
max_inodes_usage=<general_inodes_usage_treshold> max_inodes_usage=<general_inodes_usage_treshold>

8
sources/lib/platform/linux/check_linux_fs_usage.sh
View File

@ -27,6 +27,7 @@
# @(#) 2019-01-30: refactored to support custom definitions with all # @(#) 2019-01-30: refactored to support custom definitions with all
# filesystems check [Patrick Van der Veken] # filesystems check [Patrick Van der Veken]
# @(#) 2019-02-04: fix in cleanup # @(#) 2019-02-04: fix in cleanup
# @(#) 2019-02-18: fixes + help update
# ----------------------------------------------------------------------------- # -----------------------------------------------------------------------------
# DO NOT CHANGE THIS FILE UNLESS YOU KNOW WHAT YOU ARE DOING! # DO NOT CHANGE THIS FILE UNLESS YOU KNOW WHAT YOU ARE DOING!
#****************************************************************************** #******************************************************************************
@ -36,7 +37,7 @@ function check_linux_fs_usage
{ {
# ------------------------- CONFIGURATION starts here ------------------------- # ------------------------- CONFIGURATION starts here -------------------------
typeset _CONFIG_FILE="${CONFIG_DIR}/$0.conf" typeset _CONFIG_FILE="${CONFIG_DIR}/$0.conf"
typeset _VERSION="2019-02-04" # YYYY-MM-DD typeset _VERSION="2019-02-18" # YYYY-MM-DD
typeset _SUPPORTED_PLATFORMS="Linux" # uname -s match typeset _SUPPORTED_PLATFORMS="Linux" # uname -s match
# ------------------------- CONFIGURATION ends here --------------------------- # ------------------------- CONFIGURATION ends here ---------------------------
@ -231,7 +232,7 @@ then
fi fi
if (( _LOG_HEALTHY > 0 || _STC > 0 )) if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then then
log_hc "$0" ${_STC} "${_MSG}" ${_INODES_USAGE} ${_CFG_MAX_INODES_USAGE} log_hc "$0" ${_STC} "${_MSG}" "${_INODES_USAGE}" "${_CFG_MAX_INODES_USAGE}"
fi fi
done done
# add df output to stdout log_hc # add df output to stdout log_hc
@ -295,7 +296,7 @@ then
fi fi
if (( _LOG_HEALTHY > 0 || _STC > 0 )) if (( _LOG_HEALTHY > 0 || _STC > 0 ))
then then
log_hc "$0" ${_STC} "${_MSG}" ${_SPACE_USAGE} ${_CFG_SPACE_THRESHOLD} log_hc "$0" ${_STC} "${_MSG}" "${_SPACE_USAGE}" "${_CFG_SPACE_THRESHOLD}"
fi fi
done done
# add df output to stdout log_hc # add df output to stdout log_hc
@ -318,6 +319,7 @@ VERSION : $2
CONFIG : $3 with formatted stanzas (optional): CONFIG : $3 with formatted stanzas (optional):
fs:<fs_name>:<max_INODES_USAGE_in_%>:<max_space_usage_in%> fs:<fs_name>:<max_INODES_USAGE_in_%>:<max_space_usage_in%>
Other options: Other options:
log_healthy=<yes|no>
check_inodes_usage=<yes|no> check_inodes_usage=<yes|no>
check_space_usage=<yes|no> check_space_usage=<yes|no>
max_inodes_usage=<general_inodes_usage_treshold> max_inodes_usage=<general_inodes_usage_treshold>

1
sources/lib/platform/linux/check_linux_mysqld_status.sh
View File

@ -369,6 +369,7 @@ cat <<- EOT
NAME : $1 NAME : $1
VERSION : $2 VERSION : $2
CONFIG : $3 with parameters: CONFIG : $3 with parameters:
log_healthy=<yes|no>
mysql_user=<mysql_user_account> mysql_user=<mysql_user_account>
mysql_password=<mysql_user_password> mysql_password=<mysql_user_password>
mysql_host=<database_host> mysql_host=<database_host>