TIB RV
Introduction Copied
Geneos TIB RV Plug-in monitors TIB Rendezvous Service parameters including network utilisation (transmission and re-transmission rates) and hosts.
The NetProbe is configured to monitor all nodes and users that publish and subscribe on configured service groups, including Multicast and Broadcast traffic.
It is not necessary to install NetProbe on every host to monitor RV. Although you may wish to do so to monitor other non-RV related host parameters. A typical installation has one NetProbe per subnet. The plug-in subscribes to advisory messages that are broadcast by all RVD’s that are subscribing or publishing to a particular configuration. A status advisory is sent every 90-seconds by each RVD on the network. Additional advisories are sent when there is an event such as an error condition on a node.
Views Copied
The TIB RV Plug-in produces four views: Fast-Senders, Hosts, Top-Problem-Receivers, and Top-Problem-Senders.
Fast-Senders View Copied
Shows the top ten RV senders on the selected service groups.
Fast-Senders Table Legend
Name | Description |
---|---|
group:Host | The name of the host and the group that it belongs to. |
address | The IP address of the host. |
msgTransmissionRate | The rate at which the host is sending messages (per second). |
kbitTransmissionRate | The rate at which the host is sending (kilobits per second). |
averageKBitTransmissionRate | Moving average of the last 5 kbitTransmissionRate values. |
byteTransmissionRate | The rate at which the host is sending (bytes per second). |
balance | The percentage of the total message transmission rate currently attributable to the host. |
Hosts View Copied
Shows RV hosts on the selected service groups.
Hosts Headline Legend
Name | Description |
---|---|
totalNodes | The total number of nodes that are currently up. |
nodesInError | The number of nodes in an error condition. |
nodesInWarning | The number of nodes in a warning condition. |
overallPacketTransmissionRate | The overall rate of transmission for the packets. |
overallTransmissionRate | The overall rate of transmission for the service groups (sampled over a 90 second period). |
overallRetransmissionRate | The ratio between packets sent and packets re-requested (good indicator of overall Network health). |
|
The total number of packets retransmitted over a 24-hour period. |
totalKBitTransmissionRate | The total rate at which the host is sending (kilobits per second). |
Hosts Table Legend
Name | Description |
---|---|
group:Host | The name of the host (if it can be resolved) and the group it belongs to. |
address | IP address of host. |
status | The status of the RV daemon on the specific host. |
problemDescription | The error description as published by RV on the failed host. |
timeOfError | Time at which the problem was seen. |
upTime | Time when the node's RV daemon (rvd) came up. |
downTime | Time when the node's RV daemon was killed or stopped abnormally (nodes down in excess of expiryHours are removed from display). |
totalRetransmissionRate | (Retransmitted messages / retransmitted messages + sent messages) * 100 |
totalRerequestRate | (Rerequested messages / rerequested messages + received messages) * 100 |
kbitTransmissionRate | Kilobits per second transmitted. |
kbitReceiveRate | Kilobits per second received. |
outboundPacketLossRate | Number of outbound packets lost per second. (Data is available for RVD version 7.5 or later) |
inboundPacketLossRate | Number of inbound packets lost per second. (Data is available for RVD version 7.5 or later) |
clientsConnectedToService | Number of clients connected to the service. (Data is available for RVD version 8.x or later) |
suppressedInboundRetransmissionRequests | Inbound data retransmission requests suppressed (that is, requests not sent) by RXC at the receiver (in packets). (Data is available for RVD version 8.x or later) |
suppressedOutboundRetransmissionRequests | Outbound data retransmission requests suppressed (that is, requests ignored) by RXC at the sender (in packets). (Data is available for RVD version 8.x or later) |
communicationID | Communication ID. For internal use only. (Data is available for RVD version 8.x or later) |
serviceCommunicationID | Service communication ID. For internal use only. (Data is available for RVD version 8.x or later) |
averageMsgSendRate_(msg/sec) | The send rate of the node averaged over a 5 advisory broadcasts. |
os | The name of the Operating System installed on the node. |
version | The version of the RV daemon running on the node. |
serialNumber | Serial Number broadcast in advisory. |
Top-Problem-Receivers View Copied
Shows the top ten RV problem receivers on the selected service groups. Problem receivers are ranked by their retransmission rate.
Top-Problem-Receivers Headline Legend
Name | Description |
---|---|
overallRerequestRate | The overall rate at which the host is sending re-request messages. |
Top-Problem-Receivers Table Legend
Name | Description |
---|---|
group:Host | The name of the host and the group that it belongs to. |
address | The IP address of the host. |
rerequestRate | The rate at which this host is sending re-request messages. (Rerequested messages / received messages) * 100 |
averageRerequestRate | rerequestRate averaged over 5 advisory broadcasts. |
totalRerequestRate | (Rerequested messages / rerequested messages + received messages) * 100 |
kbitReceiveRate | Kilobits per second received |
byteReceiveRate | Bytes per second received |
balance | The percentage of the total re-request messages currently attributable to this host. |
Top-Problem-Senders View Copied
Shows the top ten RV problem senders from the selected service groups. Problem senders are ranked by their retransmission rate.
High retransmission rate may be caused by:
- A sender flooding the network with excessive updates that the receiving nodes can not keep up with.
- A sender having intermittent network card or cable problems.
- A receiver may just be too slow to keep up.
- The network may be generally overloaded.
The Top-Problem-Senders and Top-Problem-Receiver views help to determine the cause of retransmissions.
One top problem sender and many top problem receivers indicates a problem with the sender. Many top problem senders and one top problem receiver indicates a problem with the receiver. Many of each indicates a network that is generally overloaded.
Top-Problem-Senders Table Legend
Name | Description |
---|---|
group:Host | The name of the host and the group that it belongs to. |
address | The IP address of the host. |
retransmissionRate | The rate at which the host is responding to re-request messages (Retransmitted messages / sent messages) * 100 |
averageRetransmissionRate | RetransmissionRate averaged over 5 advisory broadcasts. |
totalRetransmissionRate | (Retransmitted messages / retransmitted messages + sent messages) * 100 |
kbitTransmissionRate | Kilobits per second transmitted |
byteTransmissionRate | Bytes per second transmitted |
balance | The percentage of total re-request reply messages currently attributable to this host. |
Menu Options Copied
The hosts view has an additional menu option when an error has been detected. The Accept Error command allows the error to be marked as ok. The original error text is preserved and “(accepted)” is appended to reflect the change.
Plug-in Configuration Copied
The following parameters can be configured for this plug-in:
groups Copied
This section contains a set of service groups that will be monitored by this plug-in.
Mandatory: Yes
groups > group > name Copied
The alias given to this service group for display purposes.
Mandatory: Yes
groups > group > command Copied
This section defines the RV listen parameters that will be used to monitor this service group.
Note
NetProbe uses the defined parameters, but does not actually run the tibrvlisten command. It is possible to verify that the correct parameters have been entered here by trying with tibrvlisten on the command line.
More details on the specifics of the listen parameters can be found in the TIBCO rendezvous documentation.
Mandatory: Yes
groups > group > command > service Copied
The Rendezvous service parameter. This is the UDP or PGM port number that will be used to subscribe.
Mandatory: No
Default: A service defined on the machine called rendezvous. If this does not exist then the following hard defaults exist: TRDP 7500, PGM 7550.
groups > group > command > network Copied
The Rendezvous network parameter. This is formed of three parts colon separated. The interface name and a set of comma separated multicast addresses.
e.g.
"<ifname>;<mcgroup1>,<mcgroup2>" or "eth1;239.10.10.1,239.10.10.2"
Note
The send address part of this configuration is not relevant as the plug-in only subscribes.
Mandatory: Yes
groups > group > command > daemon Copied
The hostname and address of an RVD to connect to and subscribe from. Specify using the format <hostname>:<portnumber>
. For example:
myhost:7500
If using a local daemon the host part can be omitted.
Mandatory: No
Default: Will default to the local daemon port 7500.
maximumNumberOfNodes Copied
The maximum number of nodes declared to the Geneos licence daemon.
The number of NODES declared is counted against the <RVNODE>
licence token. Monitoring will continue to work if the number of licensed tokens is exceeded, but the extra hosts will be shown with ‘-no licence’ in the Name field and without any values.
Mandatory: No
Default: 100
messageCountResetTime Copied
The time of day at which the retransmittedPackets variable will be reset.
Mandatory: No
Default: 01:00 am.
problemResetInterval Copied
The number of seconds to display a problem for any given rvd before resetting it.
Mandatory: No
Default: 60
hostLookup Copied
Whether to resolve ip addresses to hostnames for each monitored RVD.
Mandatory: No
Default: True
rvdDownTimeout Copied
The number of seconds after which to remove an RVD which after it has moved into a down state. Setting to 0 means that the RVD will never be removed from the view.
Mandatory: No
Default: 0
messageFile Copied
If specified, points to a flat file that contains RV advisory messages rather than subscribing across the network. Used for debugging\testing purposes.
Mandatory: No