11 KiB
Monitorix pour YunoHost
Lire le README dans d'autres langues.
Ce package vous permet d’installer Monitorix rapidement et simplement sur un serveur YunoHost.
Si vous n’avez pas YunoHost, consultez ce guide pour savoir comment l’installer et en profiter.
Vue d’ensemble
Monitorix is a free, open source, lightweight system monitoring tool designed to monitor as many services and system resources as possible. It has been created to be used under production Linux/UNIX servers, but due to its simplicity and small size can be used on embedded devices as well.
Version incluse : 3.15.0~ynh3
Démo : https://www.fibranet.cat/monitorix/
Captures d’écran
Avertissements / informations importantes
Upgrade
By default, a backup is performed before upgrading. To avoid this, you have the following options:
- Pass the
NO_BACKUP_UPGRADEenv variable with1at each upgrade. For exampleNO_BACKUP_UPGRADE=1 yunohost app upgrade monitorix. - Set
disable_backup_before_upgradeto1. You can set it with this command:
yunohost app setting monitorix disable_backup_before_upgrade -v 1
After that, the settings will be applied for all the next updates.
From command line:
yunohost app upgrade monitorix
Backup
This application now uses the core-only feature of the backup. To keep the integrity of the data and to have a better guarantee of the restoration it is recommended to proceed as follows:
- Stop Monitorix service with this command:
systemctl stop monitorix.service
- Launch Monitorix backup with this command:
yunohost backup create --app monitorix
- Backup your data with your specific strategy (could be with rsync, borg backup or just cp). The data is generally stored in
/var/lib/monitorix. - Restart Monitorix service with theses command:
systemctl start monitorix.service
Remove
Due of the backup core only feature the data directory in /var/lib/monitorix is not removed. It must be manually deleted to purge user data from the app.
More sensor
If you want to see the temperature of some sensor you can install the lm-sensor packet. For disk temperature you can instal the hddtemp packet.
Custom config
If you want do custom the monitorix config for more personnal information you can add a file in /etc/monitorix/conf.d/. This config file will be overwritte the original config in /etc/monitorix/monitorix.conf.
You will have a full complete documentation for monitorix config here : https://www.monitorix.org/documentation.html
By example you can extends the basic config by this :
priority = 5
<graph_enable>
disk = y
lmsens = y
gensens = y
mail = y
</graph_enable>
# LMSENS graph
# -----------------------------------------------------------------------------
<lmsens>
<list>
core0 = temp1
core1 =
mb0 =
cpu0 =
fan0 =
fan1 =
fan2 =
volt0 =
volt1 =
volt2 =
volt3 =
volt4 =
volt5 =
volt6 =
volt7 =
</list>
</lmsns>
# GENSENS graph
# -----------------------------------------------------------------------------
<gensens>
<list>
0 = cpu_temp
1 = cpu0_freq, cpu1_freq, cpu2_freq, cpu3_freq
</list>
<desc>
cpu_temp = /sys/class/thermal/thermal_zone0/temp
cpu0_freq = /sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_cur_freq
cpu1_freq = /sys/devices/system/cpu/cpu1/cpufreq/cpuinfo_cur_freq
cpu2_freq = /sys/devices/system/cpu/cpu2/cpufreq/cpuinfo_cur_freq
cpu3_freq = /sys/devices/system/cpu/cpu3/cpufreq/cpuinfo_cur_freq
</desc>
<unit>
cpu_temp = 1000
cpu0_freq = 0.001
cpu1_freq = 0.001
cpu2_freq = 0.001
cpu3_freq = 0.001
</unit>
<map>
cpu_temp = CPU Temperature
cpu0_freq = CPU 0 Frequency
cpu1_freq = CPU 1 Frequency
cpu2_freq = CPU 2 Frequency
cpu3_freq = CPU 3 Frequency
</map>
<alerts>
cpu_temp = 300, 65, /etc/monitorix/monitorix_alerts_scripts/cpu_temp.sh
</alerts>
</gensens>
# DISK graph
# -----------------------------------------------------------------------------
<disk>
<list>
0 = /dev/sda
</list>
<alerts>
realloc_enabled = y
realloc_timeintvl = 0
realloc_threshold = 1
realloc_script = /etc/monitorix/monitorix_alerts_scripts/disk_realloc.sh
pendsect_enabled = y
pendsect_timeintvl = 0
pendsect_threshold = 1
pendsect_script = /etc/monitorix/monitorix_alerts_scripts/disk_pendsect.sh
</alerts>
</disk>
# FS graph
# -----------------------------------------------------------------------------
<fs>
<list>
0 = /, /home, /var, /$tempdir, swap
</list>
<desc>
/ = Root FS
/home = home
/var = var
/$tempdir = tmp
</desc>
<devmap>
</devmap>
rigid = 2, 0, 2, 0
limit = 100, 1000, 100, 1000
<alerts>
/ = 3600, 98, /etc/monitorix/monitorix_alerts_scripts/fs_rootfs.sh
/home = 3600, 98, /etc/monitorix/monitorix_alerts_scripts/fs_home.sh
/var = 3600, 98, /etc/monitorix/monitorix_alerts_scripts/fs_var.sh
/$tempdir = 3600, 98, /etc/monitorix/monitorix_alerts_scripts/fs_tmp.sh
swap = 3600, 98, /etc/monitorix/monitorix_alerts_scripts/fs_swap.sh
</alerts>
</fs>
# MAIL graph
# -----------------------------------------------------------------------------
<mail>
mta = postfix
greylist = postgrey
stats_rate = real
rigid = 0, 0, 0, 0, 0
limit = 1, 1000, 1000, 1000, 1000
<alerts>
delvd_enabled = y
delvd_timeintvl = 60
delvd_threshold = 100
delvd_script = /etc/monitorix/monitorix_alerts_scripts/mail_delvd.sh
mqueued_enabled = y
mqueued_timeintvl = 3600
mqueued_threshold = 100
mqueued_script = /etc/monitorix/monitorix_alerts_scripts/mail_mqueued.sh
</alerts>
</mail>
# NET graph
# -----------------------------------------------------------------------------
<net>
list = eth0,lo
<desc>
eth0 = FastEthernet LAN, 0, 10000000
lo = loopback, 0, 10000000
</desc>
gateway = eth0
</net>
# PROCESS graph
# -----------------------------------------------------------------------------
<process>
<list>
0 = sshd, ntpd, monitorix, monitorix-httpd
1 = openvpn, ...
...
6 = mysqld, slapd, postgresql
</list>
<desc>
master = Postfix
imap = Dovecot
</desc>
rigid = 2, 0, 0, 0, 0, 0, 0, 0
limit = 100, 1000, 1000, 1000, 1000, 1000, 1000, 1000
</process>
<emailreports>
enabled = y
url_prefix = http://127.0.0.1:8081/monitorix
smtp_hostname = localhost
from_address = noreply@domain.tld
hour = 2
minute = 7
<daily>
enabled = y
graphs = system, fs, gensens, disk, netstat, port, nginx
to = user@domain.tld
</daily>
<weekly>
enabled = y
graphs = system, fs, gensens, disk, kern, proc, net, netstat, process, serv, port, user, nginx, mysql, fail2ban, int
to = user@domain.tld
</weekly>
<monthly>
enabled = y
graphs = system, fs, gensens, disk, kern, proc, net, netstat, process, serv, port, user, nginx, mysql, fail2ban, int
to = user@domain.tld
</monthly>
<yearly>
enabled = y
graphs = system, fs, gensens, disk, kern, proc, net, netstat, process, serv, port, user, nginx, mysql, fail2ban, int
to = user@domain.tld
</yearly>
</emailreports>
In this config we have :
- We set the process priority to 5 (which mean that it will be lower priority than the other process).
- We get the lmsensor sensor data.
- We get some sensors data not accessible with lmsensor (with gensens)
- We check the disk health and send an email if any error happens. For that you need to make some script. An example is available in
/usr/share/doc/monitorix/monitorix-alert.sh. - We check the filesystem.
- We check the traffic in the network card.
- We check some process.
- We send every day, week, month and year a rapport.
Documentations et ressources
- Site officiel de l’app : http://monitorix.org
- Documentation officielle de l’admin : https://www.monitorix.org/documentation.html
- Dépôt de code officiel de l’app : https://github.com/mikaku/Monitorix
- YunoHost Store : https://apps.yunohost.org/app/monitorix
- Signaler un bug : https://github.com/YunoHost-Apps/monitorix_ynh/issues
Informations pour les développeurs
Merci de faire vos pull request sur la branche testing.
Pour essayer la branche testing, procédez comme suit :
sudo yunohost app install https://github.com/YunoHost-Apps/monitorix_ynh/tree/testing --debug
ou
sudo yunohost app upgrade monitorix -u https://github.com/YunoHost-Apps/monitorix_ynh/tree/testing --debug
Plus d’infos sur le packaging d’applications : https://yunohost.org/packaging_apps