vzrmond Configuration File

The configuration for the vzrmond daemon is stored in the /etc/vzrmond.conf file. The daemon runs on the Monitor Server, provides remote monitoring capabilities for servers registered with it, and sends alerts to the specified e-mail addresses. Alerts can also be sent from external applications (e.g., via ICQ or by SMS). The configuration file lists a number of parameters, some of which should have user-provided values (from HOSTS to CUSTOM_LIST in the table below). These values must be enclosed in double quotes and separated by white spaces. The remaining parameters are set to default values which may also be changed. These values are not enclosed in quotes.

Note: To install the daemon, install the vzrmon package from the official repository.

Parameters

Name	Description	Default value
HOSTS	The list of hosts to monitor delimited by white spaces. Both hostnames and IP addresses are allowed.
EMAIL_ADDRESSES	E-mail address(es) to send alerts to. Must be separated by white spaces.
EMAIL_NOTIFICATIONS	Notification types which can be sent to the specified e-mail address(es).	`SYSTEM_UP` `SYSTEM_DOWN` `DISK_OK` `DISK_BAD` `INODES_NORM` `INODES_HIGH` `HDDBUSY_NORM` `HDDBUSY_HIGH` `SSH_UP` `SSH_DOWN` `VZSTAT_OK` `VZSTAT_BAD` `LOADAVG_NORM` `LOADAVG_HIGH` `UNINT_NORM` `UNINT_HIGH` `MEMLATM_NORM` `MEMLATM_HIGH` `MEMLATA_NORM` `MEMLATA_HIGH` `CPULATM_NORM` `CPULATM_HIGH` `CPULATA_NORM` `CPULATA_HIGH` `SWAPIN_NORM` `SWAPIN_HIGH` `SWAPOUT_NORM` `SWAPOUT_HIGH`
CUSTOM_ACTION	A custom program to send alerts from (e.g., via ICQ or by SMS).
CUSTOM_LIST	Command-line options for the program specified in `CUSTOM_ACTION` . Must be separated by white spaces.
POLL_PERIOD	Registered servers polling internal, in seconds.	`15`
CHK_MAX_FAILS	The maximum number of failed attempts to reach a server, after which the “Server is dead” alert will be sent.	`4`
LOAD_AVG	The average number of processes on the server. When this value is exceeded, an alert is sent.	`30`
PROC_UNINT	The number of uninterruptable sleeping processes (in the “D” state). When this value is exceeded, an alert is sent.	`20`
CPU_LAT_MAX_ERR	The maximum process scheduling latency, in milliseconds. When this value is exceeded, an alert is sent.	`1000`
CPU_LAT_AVG_ERR	The average process scheduling latency, in milliseconds. When this value is exceeded, an alert is sent.	`750`
MEM_LAT_MAX_ERR	The maximum memory allocation latency, in milliseconds. When this value is exceeded, an alert is sent.	`500`
MEM_LAT_AVG_ERR	The average memory allocation latency, in milliseconds. When this value is exceeded, an alert is sent.	`400`
SWAP_IN_ERR	The swap in activity, in Mb/s. When this value is exceeded, an alert is sent.	`1.0`
SWAP_OUT_ERR	The swap out activity, in Mb/s. When this value is exceeded, an alert is sent.	`1.0`
DISK_FREE_INODES_ERR	The percentage of free disk inodes. When the actual value becomes less than this value, an alert is sent.	`5`
DISK_FREE_SPACE_ERR	The percentage of free disk space. When the actual value becomes less than this value, an alert is sent.	`5`

To start monitoring a server, you should provide valid values for the HOSTS and EMAIL parameters. An external program to send alerts from must be installed on the Monitor Server and its name and command-line options must be set in the CUSTOM_ACTION and CUSTOM_LIST parameters. Alerts will be sent as messages standard for the specified program.

You should increase the value of the POLL_PERIOD parameter together with the increase in the number of monitored servers not to create an overload on the Monitor Server. The parameters related to the scheduling latency, memory allocation latency, and swap in/out activity serve to have an alert generated if the system’s performance plummets due to the abnormal values of these parameters.

Restart the vzrmond daemon after editing its configuration file.

Top of page

Feedback