| Administrator Handbook | TOC |
The installation of the Interface Monitor Plugin is
performed in the directory workspace.
Select one host in your directory and then from the contextual Menu select
Insert Task and Plugin.
You should see the following result in your workspace.
By a double click on the hammer icon you should start the Plugin Dialog Window and could configure it.
The configuration of the Interface Monitor is performed
directly in the Plugin main window.
Few parameter have to be set.
You should first select an interface frome the list.
You should set a polling interval. If the polling interval is high the line graph x scale will be bigger and thus you will have a view on a longer time period.
If you want to be warn when a load threshold is reached you have to set up the three fields.
"If usage >" parameter specifies the load
threshold in % above which you want to be warn.
"Send Event" parameter should contain the event number displayed
in the event manager (value should be unique within your set of events
and higher than 10000).
The "At level" value (between 0 and 10) specifies the gravity
of the alarm and the color of the event.
The screen below shows you example of alarm for each severity level.
The information provided by the various indicators present in the screen are all very instructive and give rich information on the interface traffics and errors.
Here is the list of these indicators.
Graph :
Incoming/Outgoing for interface index ...
Graph : Interface load in %
Graph : Packets in Error
Graph : Packets Discarded
Information
ifAdminStatus (Interface Administrative Status)
ifinUnknowProtos (Interface Inbound unknow
Protocol)
ifinUcastPkts (Interface Inbound Unicast
Packets)
ifinNUcastPkts (Interface Inbound Non-Unicast
packets)
ifOperStatus (Interface Operational Status)
ifOutQlen (Interface Output Queue lenght)
ifOutUcastPkts (Interface Out Unicast packets)
ifOutNUcastPkts (Interface Out Non-unicast
Packets) Broadcast and Multicast
Rx discard Pkts/s (Received Packets discarded)
Tx discard Pkts/s
Rx error Pkts/s
Tx error Pkts/s
Rx broad Pkts/s
Tx broad Pkts/s
This graph displays the number of octets sent and received by the interface. The in (yellow) line display the incoming octet value, the out (purple) line displays the ougoing octet value.
Both value comes from the ifOutOctets OID (The total number of octets transmitted out of the interface, including framing characters) and ifInOctects OID (The total number of octets received on the interface, including framing characters.
The y-axis displays octet values, the x-axis is the time scale.
The value displayed on the y-axis is the number of inbound or ounbond Octets counted during the polling interval. The octet value depends from the polling interval selection. if you want to know the octet/second values you should do the following conversion.
y-axis value / Polling interval = Octets per second (average)
If you want to have a speed value in bits per second the formula is :
(y-axis value / Polling interval) * 8 = Bits per second (average)
With this formula, remenber that more the polling interval is high more the value is averaged. In other words, if your traffic contains surges you could miss them.
The following table gives you examples of values for various interface types, various polling intervals and various loads.
| Number of octets per polling interval | ||||||||
|---|---|---|---|---|---|---|---|---|
| Interface type - Load | Interface Speed Mbps |
1 seconds | 5 secondes | 8 seconds | 10 seconds | 30 seconds | 60 seconds | 5 mn |
| 4 000 octets/s 50 % load on LL 64 Kbps |
64 000 | 4 000 | 20 000 | 32 000 | 40 000 | 120 000 | 240 000 | 1 200 000 |
| 8 000 octets/s 100 % load on LL 64 Kpbs |
64 000 | 8 000 | 40 000 | 64 000 | 80 000 | 240 000 | 480 000 | 2 400 000 |
| 125 000 octets/s 50 % load on E1 line |
2 000 000 | 125 000 | 625 000 | 1 000 000 | 1 250 000 | 3 750 000 | 7 500 000 | 37 500 000 |
| 250 000 octets/s 100 % load on E1 line |
2 000 000 | 250 000 | 1 250 000 | 2 000 000 | 2 500 000 | 7 500 000 | 15 000 000 | 75 000 000 |
| 625 000 octets/s = 50% load on Ethernet 100 Mbps |
10 000 000 | 625 000 | 3 125 000 | 5 000 000 | 6 250 000 | 18 750 000 | 37 500 000 | 187 500 000 |
| 1 250 000 octets/s = 100% load on Ethernet 100 Mbps |
10 000 000 | 1 250 000 | 6 250 000 | 10 000 000 | 12 500 000 | 37 500 000 | 75 000 000 | 375 000 000 |
| 6 250 000 octets/s = 50% load on Ethernet 100 Mbps |
100 000 000 | 6 250 000 | 31 250 000 | 50 000 000 | 62 500 000 | 187 500 000 | 375 000 000 | 1 875 000 000 |
| 12 500 000 octets/s = 100% load on Ethernet 100 Mbps |
100 000 000 | 12 500 000 | 62 500 000 | 100 000 000 | 125 000 000 | 375 000 000 | 750 000 000 | 3 750 000 000 |
This graph displays the percentage of load for the interface. Two lines, one for the incoming traffic, one for the outgoing traffic are represented.
The % of load is calculated by :
- reading the interface speed value in Bits/s (ifSpeed OID)
- reading the current throughput in Octets/s (ifIinOctets and ifOutOctets
OIDs)
- and by applying the following formula :
(Octets per second / Polling interval) / ((Speed of interface in Bits per second) / 8 / 100 )
The maximum reached by one of both indicator is displayed as a white line and its value is indicated on the right side (82.10 % for the Inbound load in our example).
Error packet are malformed packet at the link control layer, Ethernet for example, that could not be used by upper applications.
The maximum reached by one of both indicators is displayed as a white line and its value is indicated on the right side.
The value displayed on the y-axis is the number of inbound or outbound packets counted during the polling interval.
This table gives you the value in Pkts/s and their corresponding values for various polling interval .
| Number of packets per polling interval | |||||||
|---|---|---|---|---|---|---|---|
| 1 seconds | 5 secondes | 8 seconds | 10 seconds | 30 seconds | 60 seconds | 5 mn | |
| 10 Pkts/s | 10 | 50 | 80 | 100 | 300 | 600 | 3000 |
| 100 Pkts/ | 100 | 500 | 800 | 1 000 | 3 000 | 6 000 | 30 000 |
| 1 000 Pkts/s | 1 000 | 5 000 | 8 000 | 10 000 | 30 000 | 60 000 | 300 000 |
| 10 000 Pkts/s | 10 000 | 50 000 | 80 000 | 100 000 | 300 000 | 600 000 | 3 000 000 |
In our screen shot the maximum value recorded is 5817 packets. It gives us an average of 1164 Packet per second.
The graph displays the number of inbound and outbound packets which were chosen to be discarded even though no errors had been detected to prevent their being deliverable to a higher-layer protocol. One possible reason for discarding such a packet could be to free up buffer space.
The maximum reached by one of both indicator is displayed as a white line and its value is indicated on the right side.
The value displayed on the y-axis is the number of inbound or outbound packets counted during the polling interval.
This field displays information about the Network interface.
Description. The name given to the interface for better identification
Type. The type of interface, distinguished according to the physical/link protocol(s) immediately below the network layer in the protocol stack.
MTU (maximum Transmit Unit). The size of the largest datagram which can be sent/received on the interface, specified in octets. For interfaces that are used for transmitting network datagrams, this is the size of the largest network datagram that can be sent on the interface.
Speed. An estimate of the interface's current bandwidth in bits per second. For interfaces which do not vary in bandwidth or for those where no accurate estimation can be made, this object should contain the nominal bandwidth.
Physical address. The interface's address at the protocol layer immediately `below' the network layer in the protocol stack. For interfaces which do not have such an address (e.g., a serial line), this object should contain an octet string of zero length.
ifAdminStatus - Interface
Administrative Status
The desired state of the interface. The states could be :
The testing(3) state indicates that no operational packets can be passed.
ifinUnknowProtos - Interface
Inbound Unknown Protocols
The number of packets received via the interface which were discarded because of an unknown or unsupported protocol. Accepted protocols should be supported by the type of interface.
ifinUcastPakts - Interface Inbound Unicast Packets
The number of subnetwork-unicast packets delivered
to a higher-layer protocol.
ifinNUcastPkts - Interface
Inbound Non-Unicast Packets (Broadcast and Multicast)
The number of non-unicast (i.e., subnetwork- broadcast
or subnetwork-multicast) packets
delivered to a higher-layer protocol.
ifOperStatus - Interface Operational
Status
The current operational state of the interface.
The testing(3) state indicates that no operational
packets can be passed.
ifOutQlen - Interface Output
Queue Lenght
The length of the output packet queue (in packets)
ifOutUcastPkts - Interface
Outbound Unicast Packets
The total number of packets that higher-level protocols
requested be transmitted to a
subnetwork-unicast address, including those that were discarded or not
sent.
ifOutNUcastPkts - Interface
Outbound Non-Unicast Packets
The total number of packets that higher-level protocols
requested be transmitted to a non-
unicast (i.e., a subnetwork-broadcast or subnetwork-multicast) address,
including those
that were discarded or not sent.
The average value of received packets discarded per
second during the polling interval
Tx discard Pkts/s
The average value of transmitted packets in error per
second during the polling interval
Rx error Pkts/s
The average value of received packets in error per second
during the polling interval
Tx error Pkts/s
The average value of transmitted packets in error per
second during the polling interval
Rx broad Pkts/s
The average value of broadcast packet received per second
during the polling interval
Tx broad Pkts/s
The average value of broadacst packet transmitted per second
during the polling interval
www.loriotpro.com |
|
