mgmt/monitoring/Icinga-memory_not_enough.md

= PHP =

vi /etc/opt/rh/rh-php73/php.ini ; 2024-05-16 * for IcingaWeb to deploy vvvveeeeeeeery big config /A ; * still not enough /A ; memory_limit = 1024M memory_limit = 2048M

vi /etc/opt/rh/rh-php73/php-fpm.d/www.conf ; Default 128M is not enough for reports ;php_admin_value[memory_limit] = 1024M ; 2024-05-16 * 1024 was not enough for IPAM query execition /A php_admin_value[memory_limit] = 2048M

= MariaDB / MySQL=

vi /etc/my.cnf.d/server.cnf
[mysqld]
max_allowed_packet=100M

and restart DB

systemctl | grep db
systemctl restart rh-mariadb103-mariadb.service

or

mysql -u root -p
MariaDB [(none)]> SET GLOBAL max_allowed_packet=100000000;
MariaDB [(none)]> SHOW VARIABLES LIKE 'max_allowed_packet';

and remember to restart DB client, to renew session, in our case, IcingaWeb (which is PHP)

systemctl | grep php
systemctl restart rh-php73-php-fpm.service

= InfluxDB =

vi /etc/influxdb/influxdb.conf

# 2024-05-22  * because of two /16 subnets make for than 100k host objects, and tags in timeseries, limit need to be increased /A
max-values-per-tag=200000

systemctl restart influxdb