Aktionen
Setup zabbix-server22 » Historie » Revision 1
Revision 1/4
| Weiter »
Jeremias Keihsler, 13.01.2017 09:17
Install Zabbix Server 2.2.x¶
preliminary note¶
this procedure is based on [[vbip:software:operatingsystem:centos6:setup:zabbix-server2.2|Install Zabbix Server 2.2.x]]
you may also check
https://www.zabbix.com/documentation/2.2/manual/installation/install_from_packages
this procedure has not yet been used or tested, use with caution
remove this flag as soon as the first installation has succeeded.
Server-Install¶
yum update
yum install httpd httpd-devel
yum install zabbix22-server-mysql zabbix22-agent zabbix22-web-mysql
yum install mariadb-server
systemctl start httpd.service
systemctl enable httpd.service
systemctl start mariadb.service
systemctl enable mariadb.service
/usr/bin/mysql_secure_installation
mysql -u root -p
mysql> create database zabbix character set utf8 collate utf8_bin;
mysql> grant all privileges on zabbix.* to zabbix@localhost identified by 'password';
mysql> exit
cd /usr/share/zabbix-mysql/
mysql -u root -p zabbix < schema.sql
mysql -u root -p zabbix < images.sql
mysql -u root -p zabbix < data.sql
vim /etc/zabbix/zabbix_server.conf
# line 116: uncomment and add DB password for Zabbix DBPassword=password
vim /etc/zabbix/zabbix_agentd.conf
# line 133: change to your hostname Hostname=zabbix.server.com
vim /etc/php.ini
# line 384: change to Zabbix recommended max_execution_time = 300 # line 394: change to Zabbix recommended max_input_time = 300 # line 405: change to Zabbix recommended memory_limit = 128M # line 672: change to Zabbix recommended post_max_size = 16M # line 800: change to Zabbix recommended upload_max_filesize = 2M # line 879: uncomment and add your timezone date.timezone = Europe/Vienna
systemctl start zabbix-server
systemctl enable zabbix-server-mysql
systemctl start zabbix-agent
systemctl enable zabbix-agent
systemctl restart httpd
Firewall¶
if you are going to monitor other systems than the localhost then you have to open ports 10050 and 10051
firewall-cmd --permanent --add-port=10050/tcp
firewall-cmd --permanent --add-port=10051/tcp
systemctl restart firewalld
SELinux¶
as SELinux prevents the web-client to establish a connection to localhost:10050 you will get a wrong indication on Server-Status. Allow access to the server via
setsebool -P httpd_can_connect_zabbix 1
Backup Zabbix config¶
is taken from https://github.com/maxhq/zabbix-backup/
This script is a simple way to backup all configuration tables (eg. templates, hostgroups, hosts, triggers…) without the history data.
As the result is very small, is possible run this backup many times per day.
The backup-script may be invoked by cron
/etc/cron.d/zabbix-backup
# JKE 2016-04-10
#
# minute hour day month dayofweek
# m h d m d
# - - - - -
# | | | | |
# | | | | +-day of week (0-7) sunday=0 or 7
# | | | +---month (1-12)
# | | +-----day of month (1-31)
# | +-------hour (0-23)
# +---------min (0-59)
#
5 1 * * * root /usr/local/bin/zabbix-mysql-backupconf.sh -p password -o /var/cache/myzabbixdump -r 5 -q
zabbix-mysql-backupconf.sh
#!/usr/bin/env bash
#
# NAME
# zabbix-mysql-dump - Configuration Backup for Zabbix with MySQL
#
# SYNOPSIS
# This is a MySQL configuration backup script for Zabbix 1.x and 2.x.
# It does a full backup of all configuration tables, but only a schema
# backup of large data tables.
#
# The script is based on a script by Ricardo Santos
# (http://zabbixzone.com/zabbix/backuping-only-the-zabbix-configuration/)
#
# CONTRIBUTORS
# - Ricardo Santos
# - Jens Berthold (maxhq)
# - Oleksiy Zagorskyi (zalex)
# - Petr Jendrejovsky
# - Jonathan Bayer
# - Andreas Niedermann (dre-)
# - Mi?u Moldovan (dumol)
# - Daniel Schneller (dschneller)
#
# HISTORY
# 0.8.0 (2016-01-22)
# FIX: Only invoke `dig` if available
# ENH: Option -c to use a MySQL config ("options") file (suggested by Daniel Schneller)
# ENH: Option -r to rotate backup files (Daniel Schneller)
# ENH: Add database version to filename if available
# ENH: Add quiet mode. IP reverse lookup optional (Daniel Schneller)
# ENH: Bash related fixes (Misu Moldovan)
# CHG: Default output directory is now $PWD instead of script dir
#
# 0.7.1 (2015-01-27)
# NEW: Parsing of commandline arguments implemented
# ENH: Try reverse lookup of IPs and include hostname/IP in filename
# REV: Stop if database password is wrong
#
# 0.7.0 (2014-10-02)
# ENH: Complete overhaul to make script work with lots of Zabbix versions
#
# 0.6.0 (2014-09-15)
# REV: Updated the table list for use with zabbix v2.2.3
#
# 0.5.0 (2013-05-13)
# NEW: Added table list comparison between database and script
#
# 0.4.0 (2012-03-02)
# REV: Incorporated mysqldump options (suggested by Jonathan Bayer)
#
# 0.3.0 (2012-02-06)
# ENH: Backup of Zabbix 1.9.x / 2.0.0, removed unnecessary use of
# variables (DATEBIN etc) for commands that use to be in $PATH
#
# 0.2.0 (2011-11-05)
#
# AUTHOR
# Jens Berthold (maxhq), 2016
#
# DEFAULT VALUES
#
# DO NOT EDIT THESE VALUES!
# Instead, use command line parameters or a config file to specify options.
#
DUMPDIR="$PWD"
DBHOST="127.0.0.1"
DBNAME="zabbix"
DBUSER="zabbix"
DBPASS=""
QUIET="no"
REVERSELOOKUP="yes"
GENERATIONSTOKEEP=0
#
# SHOW HELP
#
if [ -z "$1" ]; then
cat <<EOF
USAGE
$(basename $BASH_SOURCE) [options]
OPTIONS
-h HOST
Hostname/IP of MySQL server.
Default: $DBHOST
-d DATABASE
Zabbix database name.
Default: $DBNAME
-u USER
MySQL user to access Zabbix database.
Default: $DBUSER
-p PASSWORD
MySQL user password (specify "-" for a prompt).
Default: no password
-o DIR
Save Zabbix MySQL dumps to DIR.
Default: $DUMPDIR
-c FILE
Use FILE for MySQL options (passed via --defaults-extra-file).
PLEASE NOTE:
mysqldump needs the database to be specified via command line.
So the first "database" options found in the config file is
used for mysqldump.
-r NUM
Rotate backups while keeping up to NUM generations.
Uses filename to match.
Default: keep all backups
-n
Skip reverse lookup of IP address for host.
-q
Quiet mode: no output except for errors (for batch/crontab use).
EXAMPLES
$(basename $BASH_SOURCE) -h 1.2.3.4 -d zabbixdb -u zabbix -p test
$(basename $BASH_SOURCE) -u zabbix -p - -o /tmp
$(basename $BASH_SOURCE) -c /etc/mysql/mysql.cnf
$(basename $BASH_SOURCE) -c /etc/mysql/mysql.cnf -d zabbixdb
EOF
exit 1
fi
#
# PARSE COMMAND LINE ARGUMENTS
#
DB_GIVEN=0
while getopts ":h:d:u:p:o:r:c:qn" opt; do
case $opt in
h) DBHOST="$OPTARG" ;;
d) DBNAME="$OPTARG"; DB_GIVEN=1 ;;
u) DBUSER="$OPTARG" ;;
p) DBPASS="$OPTARG" ;;
c) CNFFILE="$OPTARG" ;;
o) DUMPDIR="$OPTARG" ;;
r) GENERATIONSTOKEEP=$(printf '%.0f' "$OPTARG") ;;
n) REVERSELOOKUP="no" ;;
q) QUIET="yes" ;;
\?) echo "Invalid option: -$OPTARG" >&2; exit 1 ;;
:) echo "Option -$OPTARG requires an argument" >&2; exit 1 ;;
esac
done
# Password prompt
if [ "$DBPASS" = "-" ]; then
read -s -p "Enter MySQL password for user '$DBUSER' (input will be hidden): " DBPASS
echo ""
fi
# Config file validations
if [ ! -z "$CNFFILE" ]; then
if [ ! -r "$CNFFILE" ]; then
echo "ERROR: Cannot read configuration file $CNFFILE" >&2
exit 1
fi
# Database name needs special treatment:
# For mysqldump it has to be specified on the command line!
# Therefore we need to get it from the config file
if [ $DB_GIVEN -eq 0 ]; then
DBNAME=$(grep -m 1 ^database= "$CNFFILE" | cut -d= -f2)
fi
fi
#
# CONSTANTS
#
MYSQL_OPTS=()
[ ! -z "$CNFFILE" ] && MYSQL_OPTS=("${MYSQL_OPTS[@]}" --defaults-extra-file="$CNFFILE")
[ ! -z "$DBHOST" ] && MYSQL_OPTS=("${MYSQL_OPTS[@]}" -h $DBHOST)
[ ! -z "$DBUSER" ] && MYSQL_OPTS=("${MYSQL_OPTS[@]}" -u $DBUSER)
[ ! -z "$DBPASS" ] && MYSQL_OPTS=("${MYSQL_OPTS[@]}" -p"$DBPASS")
MYSQL_OPTS_BATCH=("${MYSQL_OPTS[@]}" --batch --silent)
[ ! -z "$DBNAME" ] && MYSQL_OPTS_BATCH=("${MYSQL_OPTS_BATCH[@]}" -D $DBNAME)
# Host name: try reverse lookup if IP is given
DBHOSTNAME="$DBHOST"
command -v dig >/dev/null 2>&1
FIND_DIG=$?
if [ "$REVERSELOOKUP" == "yes" -a $FIND_DIG -eq 0 ]; then
# Try resolving a given host ip
newHostname=$(dig +noall +answer -x $DBHOST | sed -r 's/((\S+)\s+)+([^\.]+)\..*/\3/')
test \! -z "$newHostname" && DBHOSTNAME="$newHostname"
fi
#
# CONFIG DUMP
#
if [ "$QUIET" == "no" ]; then
cat <<-EOF
Configuration:
- host: $DBHOST ($DBHOSTNAME)
- database: $DBNAME
- user: $DBUSER
- output: $DUMPDIR
EOF
fi
#
# FUNCTIONS
#
# Returns TRUE if argument 1 is part of the given array (remaining arguments)
elementIn () {
local e
for e in "${@:2}"; do [[ "$e" == "$1" ]] && return 0; done
return 1
}
#
# CHECKS
#
if [ ! -x /usr/bin/mysqldump ]; then
echo "mysqldump not found." >&2
echo "(with Debian, \"apt-get install mysql-client\" will help)" >&2
exit 1
fi
#
# READ TABLE LIST from __DATA__ section at the end of this script
# (http://stackoverflow.com/a/3477269/2983301)
#
DATA_TABLES=()
while read line; do
table=$(echo "$line" | cut -d" " -f1)
echo "$line" | cut -d" " -f5 | grep -qi "DATA"
test $? -eq 0 && DATA_TABLES+=($table)
done < <(sed '0,/^__DATA__$/d' "$BASH_SOURCE" | tr -s " ")
# paranoid check
if [ ${#DATA_TABLES[@]} -lt 5 ]; then
echo "ERROR: The number of large data tables configured in this script is less than 5." >&2
exit 1
fi
#
# BACKUP
#
# Read table list from database
[ "$QUIET" == "no" ] && echo "Fetching list of existing tables..."
DB_TABLES=$(mysql "${MYSQL_OPTS_BATCH[@]}" -e "SELECT table_name FROM information_schema.tables WHERE table_schema = '$DBNAME'" 2>&1)
if [ $? -ne 0 ]; then echo -e "ERROR while trying to access database:\n$DB_TABLES" 2>&1; exit 1; fi
DB_TABLES=$(echo "$DB_TABLES" | sort)
DB_TABLE_NUM=$(echo "$DB_TABLES" | wc -l)
# Query Zabbix database version
VERSION=""
DB_VER=$(mysql "${MYSQL_OPTS_BATCH[@]}" -N -e "select optional from dbversion;" 2>&1)
if [ $? -eq 0 ]; then
# version string is like: 02030015
re='(.*)([0-9]{2})([0-9]{4})'
if [[ $DB_VER =~ $re ]]; then
VERSION="_db-${BASH_REMATCH[1]}.$(( ${BASH_REMATCH[2]} + 0 )).$(( ${BASH_REMATCH[3]} + 0 ))"
fi
fi
# Assemble file name
DUMPFILENAME_PREFIX="zabbix_cfg_${DBHOSTNAME}"
DUMPFILEBASE="${DUMPFILENAME_PREFIX}_$(date +%Y%m%d-%H%M)${VERSION}.sql"
DUMPFILE="$DUMPDIR/$DUMPFILEBASE"
PROCESSED_DATA_TABLES=()
i=0
mkdir -p "${DUMPDIR}"
[ "$QUIET" == "no" ] && echo "Starting table backups..."
while read table; do
# large data tables: only store schema
if elementIn "$table" "${DATA_TABLES[@]}"; then
dump_opt="--no-data"
PROCESSED_DATA_TABLES+=($table)
# configuration tables: full dump
else
dump_opt="--extended-insert=FALSE"
fi
mysqldump "${MYSQL_OPTS[@]}" \
--routines --opt --single-transaction --skip-lock-tables \
$dump_opt \
$DBNAME --tables ${table} >> "$DUMPFILE"
if [ "$QUIET" == "no" ]; then
# show percentage
i=$((i+1)); i_percent=$(($i * 100 / $DB_TABLE_NUM))
if [ $(($i_percent % 12)) -eq 0 ]; then
echo -n "${i_percent}%"
else
if [ $(($i_percent % 2)) -eq 0 ]; then echo -n "."; fi
fi
fi
done <<<"$DB_TABLES"
#
# COMPRESS BACKUP
#
if [ "$QUIET" == "no" ]; then
echo -e "\n"
echo "For the following large tables only the schema (without data) was stored:"
for table in "${PROCESSED_DATA_TABLES[@]}"; do echo " - $table"; done
echo
echo "Compressing backup file..."
fi
gzip -f "$DUMPFILE"
if [ $? -ne 0 ]; then
echo -e "\nERROR: Could not compress backup file, see previous messages" >&2
exit 1
fi
[ "$QUIET" == "no" ] && echo -e "\nBackup Completed:\n${DUMPFILE}.gz"
#
# ROTATE OLD BACKUPS
#
if [ $GENERATIONSTOKEEP -gt 0 ]; then
[ "$QUIET" == "no" ] && echo "Removing old backups, keeping up to $GENERATIONSTOKEEP"
REMOVE_OLD_CMD="cd \"$DUMPDIR\" && ls -t \"${DUMPFILENAME_PREFIX}\"* | /usr/bin/awk \"NR>${GENERATIONSTOKEEP}\" | xargs rm -f "
eval ${REMOVE_OLD_CMD}
if [ $? -ne 0 ]; then
echo "ERROR: Could not rotate old backups" >&2
exit 1
fi
fi
exit 0
################################################################################
# List of all known table names and a flag indicating data (=large) tables
#
__DATA__
acknowledges 1.3.1 - 2.4.0 DATA
actions 1.3.1 - 2.4.0
alerts 1.3.1 - 2.4.0 DATA
application_template 2.1.0 - 2.4.0
applications 1.3.1 - 2.4.0
auditlog 1.3.1 - 2.4.0 DATA
auditlog_details 1.7 - 2.4.0 DATA
autoreg 1.3.1 - 1.3.4
autoreg_host 1.7 - 2.4.0
conditions 1.3.1 - 2.4.0
config 1.3.1 - 2.4.0
dbversion 2.1.0 - 2.4.0
dchecks 1.3.4 - 2.4.0
dhosts 1.3.4 - 2.4.0
drules 1.3.4 - 2.4.0
dservices 1.3.4 - 2.4.0
escalations 1.5.3 - 2.4.0
events 1.3.1 - 2.4.0 DATA
expressions 1.7 - 2.4.0
functions 1.3.1 - 2.4.0
globalmacro 1.7 - 2.4.0
globalvars 1.9.6 - 2.4.0
graph_discovery 1.9.0 - 2.4.0
graph_theme 1.7 - 2.4.0
graphs 1.3.1 - 2.4.0
graphs_items 1.3.1 - 2.4.0
group_discovery 2.1.4 - 2.4.0
group_prototype 2.1.4 - 2.4.0
groups 1.3.1 - 2.4.0
help_items 1.3.1 - 2.1.8
history 1.3.1 - 2.4.0 DATA
history_log 1.3.1 - 2.4.0 DATA
history_str 1.3.1 - 2.4.0 DATA
history_str_sync 1.3.1 - 2.2.6 DATA
history_sync 1.3.1 - 2.2.6 DATA
history_text 1.3.1 - 2.4.0 DATA
history_uint 1.3.1 - 2.4.0 DATA
history_uint_sync 1.3.1 - 2.2.6 DATA
host_discovery 2.1.4 - 2.4.0
host_inventory 1.9.6 - 2.4.0
host_profile 1.9.3 - 1.9.5
hostmacro 1.7 - 2.4.0
hosts 1.3.1 - 2.4.0
hosts_groups 1.3.1 - 2.4.0
hosts_profiles 1.3.1 - 1.9.2
hosts_profiles_ext 1.6 - 1.9.2
hosts_templates 1.3.1 - 2.4.0
housekeeper 1.3.1 - 2.4.0
httpstep 1.3.3 - 2.4.0
httpstepitem 1.3.3 - 2.4.0
httptest 1.3.3 - 2.4.0
httptestitem 1.3.3 - 2.4.0
icon_map 1.9.6 - 2.4.0
icon_mapping 1.9.6 - 2.4.0
ids 1.3.3 - 2.4.0
images 1.3.1 - 2.4.0
interface 1.9.1 - 2.4.0
interface_discovery 2.1.4 - 2.4.0
item_condition 2.3.0 - 2.4.0
item_discovery 1.9.0 - 2.4.0
items 1.3.1 - 2.4.0
items_applications 1.3.1 - 2.4.0
maintenances 1.7 - 2.4.0
maintenances_groups 1.7 - 2.4.0
maintenances_hosts 1.7 - 2.4.0
maintenances_windows 1.7 - 2.4.0
mappings 1.3.1 - 2.4.0
media 1.3.1 - 2.4.0
media_type 1.3.1 - 2.4.0
node_cksum 1.3.1 - 2.2.6
node_configlog 1.3.1 - 1.4.7
nodes 1.3.1 - 2.2.6
opcommand 1.9.4 - 2.4.0
opcommand_grp 1.9.2 - 2.4.0
opcommand_hst 1.9.2 - 2.4.0
opconditions 1.5.3 - 2.4.0
operations 1.3.4 - 2.4.0
opgroup 1.9.2 - 2.4.0
opmediatypes 1.7 - 1.8.21
opmessage 1.9.2 - 2.4.0
opmessage_grp 1.9.2 - 2.4.0
opmessage_usr 1.9.2 - 2.4.0
optemplate 1.9.2 - 2.4.0
profiles 1.3.1 - 2.4.0
proxy_autoreg_host 1.7 - 2.4.0
proxy_dhistory 1.5 - 2.4.0
proxy_history 1.5.1 - 2.4.0
regexps 1.7 - 2.4.0
rights 1.3.1 - 2.4.0
screens 1.3.1 - 2.4.0
screens_items 1.3.1 - 2.4.0
scripts 1.5 - 2.4.0
service_alarms 1.3.1 - 2.4.0
services 1.3.1 - 2.4.0
services_links 1.3.1 - 2.4.0
services_times 1.3.1 - 2.4.0
sessions 1.3.1 - 2.4.0
slides 1.3.4 - 2.4.0
slideshows 1.3.4 - 2.4.0
sysmap_element_url 1.9.0 - 2.4.0
sysmap_url 1.9.0 - 2.4.0
sysmaps 1.3.1 - 2.4.0
sysmaps_elements 1.3.1 - 2.4.0
sysmaps_link_triggers 1.5 - 2.4.0
sysmaps_links 1.3.1 - 2.4.0
timeperiods 1.7 - 2.4.0
trends 1.3.1 - 2.4.0 DATA
trends_uint 1.5 - 2.4.0 DATA
trigger_depends 1.3.1 - 2.4.0
trigger_discovery 1.9.0 - 2.4.0
triggers 1.3.1 - 2.4.0
user_history 1.7 - 2.4.0
users 1.3.1 - 2.4.0
users_groups 1.3.1 - 2.4.0
usrgrp 1.3.1 - 2.4.0
valuemaps 1.3.1 - 2.4.0
Von Jeremias Keihsler vor fast 8 Jahren aktualisiert · 1 Revisionen