[Name]
hanetobserv - Setting, modifying, deleting, and displaying the information for the communication target monitoring function
[Synopsis]
/opt/FJSVhanet/usr/sbin/hanetobserv command [args]
[Feature description]
The hanetobserv command sets, modifies, deletes, and displays the monitoring destination information required for the operation in GS linkage mode.
Command | Process outline | Authority |
|---|---|---|
create | Sets a monitoring destination information | Super user |
delete | Deletes monitoring destination information | Super user |
Displays monitoring destination information | General user | |
param | Modifies the monitoring destination information. | Super user |
(1) create command
The operation in GS linkage mode requires the monitoring of the communication target. This enables the system to continue communication using other communication paths when a failure occurs. Use the create command to generate a communication target. The following is the command format for generating a monitoring destination:
GS communication /opt/FJSVhanet/usr/sbin/hanetobserv create -n node -i ipaddress -t nicaddress1[,nicaddress2,...] |
Specify a name by which to identify the node of a communication target, using up to 16 one-byte characters.
Specify a host name or IP address of a virtual interface held by the communication target. Up to 128 can be set. This host name must correspond to an IP address in a network database such as the /etc/hosts files. You can directly specify an IP address instead of a host name. In this case, you must specify the IP address in dotted decimal notation.
Specify the IP addresses of physical interfaces bundled by a virtual interface held by the communication target, using a comma (",") to separate those IP addresses. Up to 32 IP addresses can be set.
In addition, if you perform remote network communication via router to connect to the communication target, specify the IP addresses in the format of "IP address of a neighboring router + IP address of a physical interface."
Specify the IP addresses of physical interfaces bundled by a virtual interface of a destination cluster node and the IP addresses of neighboring switches, using a comma (",") to separate those IP addresses.
Note that, you can also specify the switches with the host names instead of IP addresses.
Specify the host name or IP address of a physical interface bundled by a virtual interface.
Specify the IP address or host name of the router for the local system. This option can be omitted if you do not perform remote network communication via router to connect to GS.
(2) delete command
The following is the format of the delete command used to delete the monitoring destination information created using the create command:
To delete all the monitoring destination information: /opt/FJSVhanet/usr/sbin/hanetobserv delete -n all To delete the monitoring destination information by specifying the name of the monitoring destination node: /opt/FJSVhanet/usr/sbin/hanetobserv delete -n node1[,node2,...] To delete the monitoring destination information by specifying the virtual IP address of the monitoring destination: /opt/FJSVhanet/usr/sbin/hanetobserv delete -n node -i ipaddress1[,ipaddress2,...] To delete the monitoring destination information by specifying the physical IP address and router IP address of the monitoring destination: /opt/FJSVhanet/usr/sbin/hanetobserv delete -n node -i ipaddress -t [routeraddress1+]nicaddress1[,[routeraddress2+]nicaddress2] |
Specify the name of the remote host. You can specify more than one name by delimiting them with a comma.
If all is specified, all monitoring destination information is deleted.
Specify a remote node name that is set in the monitoring destination information and should be deleted. You can specify more than one remote node name by listing them delimited with a comma.
Specify the name of the remote host you want to delete that is set in the monitoring destination information.
Specify the virtual IP address or host name of the virtual interface on the remote host. The definition information of the remote host is also deleted if only one virtual interface is defined on the remote host.
Specify the IP addresses or host names to be deleted. You can use the print command of hanetobserv to confirm the combination of the IP addresses or host names to be deleted.
Specify the IP addresses or host names of the physical interfaces assigned to the virtual interface.
Specify the IP address or host name of the router for the local system.
(3) print command
Use the print command to display the current monitoring destination information. The following is the format of the print command. If no option is specified, information on both the monitoring destination and the relay destination is output.
/opt/FJSVhanet/usr/sbin/hanetobserv print |
The following shows an example of displaying monitoring destination information:
# /opt/FJSVhanet/usr/sbin/hanetobserv print
interval(s) = 5 sec
times(c) = 5 times
idle(p) = 60 sec
repair_time(b) = 5 sec
repair_retry(r) = 0 times
fail over mode(f) = YES
Destination Host Virtual Address (Router Address+)NIC Address
+----------------+-------------------+--------------------------------+
hostA 192.168.91.1 192.168.70.254+192.168.80.2,
192.168.71.254+192.168.81.2
hostB ipaddress3 ipaddress4,ipaddress5 |
Item | Explanation | |
|---|---|---|
Interval | Displays the monitoring interval in the stationary status. | |
Idle | Displays in seconds the wait time that elapses after monitoring starts and before the HUB links up. | |
times | Displays the monitoring count. | |
repair_time | Displays the recovery monitoring interval in seconds. | |
repair_retry | Displays the retry count of the recovery monitoring. | |
fail over mode | YES | If the virtual interface is registered in the cluster resource, node switching is performed when all the transfer routes fail. (default) |
NO | If the virtual interface is registered in the cluster resource, node switching is not performed when all the transfer routes fail. | |
Destination Host | Outputs the host name of the communication target. | |
Virtual Address | Displays the host name or IP address set for the virtual interface of the communication target. | |
(Router addr+)NIC Address | Displays the host name or IP address of the physical interfaces assigned to the virtual interface of the communication target, and the host name or IP address of a local router. | |
(4) param command
Use this command to modify each parameter value for the remote host monitoring function. The command format is as follows.
/opt/FJSVhanet/usr/sbin/hanetobserv param [-s sec] [-c times] [-p sec] [-b sec] [-r times] [-f {yes | no}] |
Specify the monitoring time in seconds. A value from 1 to 300 can be specified (note that the product of sec and time must be 300 or less). If this option is not specified, the previous setting is enabled. Initially, 5 (seconds) is specified.
Specify the monitoring count. A value from 1 to 300 can be specified (note that the product of sec and time must be 300 or less). If this option is not specified, the previous setting is enabled. Initially, 5 (times) is specified.
Specify in seconds the wait time that should elapse after monitoring starts and before the HUB links up in GS linkage mode. A value from 1 to 300 can be specified. If this option is not specified, the previous setting is enabled. Initially, 60 (seconds) is specified. If the specified value is less than the monitoring interval multiplied by the monitoring count, the system ignores the specified link-up time and adopts the time obtained by multiplying the monitoring interval by the monitoring count.
When detected an error in communication target monitoring, specify an interval to monitor recovery. The range possible to set is zero to 300. If not specified this option, the values set the last time become valid. 5 (seconds) is set as the initial set value.
Specify the retry count to return to the regular monitoring if recovery monitoring has been consecutively successful after detecting an error in recovery monitoring by remote host monitoring. A value from 0 to 300 can be specified. The default value is 0 (times). (The monitoring target is considered as recovered if the ping monitoring succeeds once and no retry occurs.)
Specify the operation used when node switching occurs due to a line failure during cluster operation. If this option is not specified, the previous setting is enabled. Initially, "yes" is specified. (This parameter is enabled only when a takeover virtual interface is set for cluster operation.)
yes: Node switching is performed if a line monitoring failure occurs.
no: No node switching is performed if a line monitoring failure occurs.
Note
Setting "no" restricts switching caused by an error occurred in transfer routes. This does not restrict node switching caused by other errors such as an activation failure for virtual interfaces.
If the cluster application is switched when all the transfer paths for the virtual interface are failed, resources will fail even if "no" is set.
To use "no" for maintenance purpose of nodes in the cluster, see "I.6.5 Maintenance procedure performed when the communication target stopped" and perform the procedure.
[Notes]
To change the monitoring destination, delete it first, and then re-create it.
To add, delete, or change a monitoring destination, the virtual interface in GS linkage mode (operation mode "c") must be inactivated.
An IP address or host name to be specified when the communication target monitoring function is set or changed must be defined in /etc/hosts.
The node name information must not be specified as "all".
Up to 32 physical interfaces can be specified to be bundled by the virtual interface of the communication target to be specified in the monitoring destination information.
When specified a host name to where to set a host name or an IP address with this command, it is not possible to change the corresponding host name on the host database of such as /etc/hosts files. To change the information of the host name, it is necessary to temporarily delete a definition of a Redundant Line Control function to use the corresponding host name and to set the definition again.
Do not use characters other than alphanumeric characters, period, and hyphen for the host name. If characters other than the above are used, re-write the host names in /etc/hosts so that it does not contain any other characters. Also, the first and last character for the host name must be alphanumeric character.
[Examples]
The following shows a setting example in which monitoring is performed while the communication target host hahostA has virtual IP address "vip1", which bundles two physical IP addresses ipaddress1 and ipaddress2. The host name is assumed to be associated with the IP address in the /etc/hosts file.
# /opt/FJSVhanet/usr/sbin/hanetobserv create -n hahostA -i vip1 -t ipaddress1,ipaddress2 |
The following shows a setting example in which monitoring is performed while the monitoring information for the virtual IP address "vip1" of the communication target host hahostA is specified and then the two physical IP addresses ipaddress3, and ipaddress4 for the virtual IP address vip1 are added. The host name is assumed to be associated with the IP address in the /etc/hosts file.
# /opt/FJSVhanet/usr/sbin/hanetobserv create -n hahostA -i vip1 -t ipaddress3,ipaddress4 |
The following example shows the settings in which there exist routers rt1 and rt2 for the local system, there exists the virtual IP address "vip1" on the communication target host hahostA, and the "vip1" is assigned to the physical IP addresses ipaddress1 and ipaddress2. The host name is assumed to be associated with the IP address in the /etc/hosts file.
# /opt/FJSVhanet/usr/sbin/hanetobserv create -n hahostA -i vip1 -t rt1+ipaddress1,rt2+ipaddress2 |
The following is an example for setting destination cluster node monitoring information and switches monitoring information.
It shows when the destination node name is cl_node, a take-over IP address is cl_vip, the physical IP addresses of the destination node are cl_ipaddress1 and cl_ipaddress2, and the IP addresses of neighboring switches are sw1 and sw2. The host name is assumed to be associated with the IP address in the /etc/hosts file.
# /opt/FJSVhanet/usr/sbin/hanetobserv create -n cl_node -i cl_vip -t cl_ipaddress1,cl_ipaddress2,sw1,sw2 |
The following shows an example of deleting all the monitoring destination information.
# /opt/FJSVhanet/usr/sbin/hanetobserv delete -n all |
The following shows an example of deleting all the information held by the monitored host (hahostA).
# /opt/FJSVhanet/usr/sbin/hanetobserv delete -n hahostA |
The following shows an example of deleting the information under the virtual IP address "vip1" held by the monitored host (hahostA).
# /opt/FJSVhanet/usr/sbin/hanetobserv delete -n hahostA -i vip1 |
The following shows an example of deleting the physical IP addresses (ipaddress1, ipaddress2) under the virtual IP address "vip1".
# /opt/FJSVhanet/usr/sbin/hanetobserv delete -i vip1 -t ipaddress1,ipaddress2 |
The following shows an example of specifying and deleting the physical IP and router information in the virtual IP address "vip1" that the monitoring destination remote host hahostA has.
# /opt/FJSVhanet/usr/sbin/hanetobserv delete -n hahostA -i vip1 -t rt1+ipaddress1,rt2+ipaddress2 |
The following shows an example of displaying the configuration information list of a virtual interface.
# /opt/FJSVhanet/usr/sbin/hanetobserv print |
The following shows an example of setting the monitoring interval and monitoring count for the remote host monitoring function to 3 seconds and 2 times respectively.
# /opt/FJSVhanet/usr/sbin/hanetobserv param -s 3 -c 2 |
The following shows an example of setting the remote host monitoring function to perform node switching when all the transfer routs fail. (when you set the node switching task to default)
# /opt/FJSVhanet/usr/sbin/hanetobserv param -f yes |
