Page 1 of 1

Reporting Tab performance in large scale Monitoring

Posted: Wed Sep 26, 2018 4:07 am
by salami
Hi,
I have Nagios XI 5.4.13 with 12K Host. I offloading DB to remote Linux centos 7 with 10 core CPU and 8 GB RAM on MariaDB 5.5. I need to create Reports about availability in Reports Tab but when I select a host or hostgroup and run the report, avail.cgi process runs but nothing has been generated for a long time. Is it possible to have better performance in reporting Tab?

my.cnf file is as below:

Code: Select all

[mysqld]
datadir=/var/lib/mysql
socket=/var/lib/mysql/mysql.sock
symbolic-links=0
bind-address={mysql_IP_Address}
port=3306
skip_name_resolve
key-buffer-size = 32M
myisam-recover-options = FORCE,BACKUP
max-allowed-packet = 16M
max-connect-errors = 1000000
expire-logs-days = 14
innodb-flush-log-at-trx-commit = 1
innodb-file-per-table          = 1
innodb-buffer-pool-size        = 2G
tmp-table-size                 = 32M
max-heap-table-size            = 32M
query-cache-size               = 1024M
max-connections                = 2000
thread-cache-size              = 512
open-files-limit               = 65535
table-definition-cache         = 4096
table-open-cache               = 4096

[mysqld_safe]
log-error=/var/log/mariadb/mariadb.log
pid-file=/var/run/mariadb/mariadb.pid

Re: Reporting Tab performance in large scale Monitoring

Posted: Wed Sep 26, 2018 3:51 pm
by lmiltchev
The avail.cgi is a bottleneck, and we are looking into removing its usage in XI 6. See this post:

https://support.nagios.com/forum/viewto ... =6&t=49516

Re: Reporting Tab performance in large scale Monitoring

Posted: Tue Oct 02, 2018 3:54 am
by salami
Thanks for your reply.
Is there any time estimation for releasing new version of Nagios XI (6 version)?

Also, I have a suggestion for new reporting system for large scale environments. We need to separate reporting DB from Main DB in large scales for increasing performance, avoiding affects on monitoring system and increasing reliability. Is it possible to plan in development team to create a new feature for determining report server in Nagios XI web UI?

We can create a replica server for reporting purpose and in this situation we can determine for Nagios XI to queried reports from slave DB server instead of main server.

Thank you so much

Re: Reporting Tab performance in large scale Monitoring

Posted: Tue Oct 02, 2018 11:31 am
by lmiltchev
Is there any time estimation for releasing new version of Nagios XI (6 version)?
Nagios XI 6 is the next planned major release for Q3 of 2019.
https://www.nagios.com/roadmaps/
Also, I have a suggestion for new reporting system for large scale environments. We need to separate reporting DB from Main DB in large scales for increasing performance, avoiding affects on monitoring system and increasing reliability. Is it possible to plan in development team to create a new feature for determining report server in Nagios XI web UI?

We can create a replica server for reporting purpose and in this situation we can determine for Nagios XI to queried reports from slave DB server instead of main server.
I filed an internal feature request for adding this functionality, currently pending approval. Please keep in mind that the decision to implement the enhancement is at the sole discretion of our development team.