The Signal Packet Capture tool can potentially cause a Link Detection Group (LDG) failure due to how the tool manipulates the NIC and associated NP rules. To avoid this, Ribbon recommends to increase the LDG timer values. Additionally, you can increase the LDG's threshold value to 2 (default) so that even if an echo response is delayed and the link monitor fails, the LDG does not go into a failed state and attempt to trigger a switchover upon an initial failure.
In this section:
Call Trace Filtering
The SBC supports configuring Call Trace Filters using the Signaling Packet Capture command to establish criteria to determine call events and logging PES call trace messages (using the ".TRC" file extension) to the system trace data event log. This information is used for debugging policy and call routing issues. The following default information is collected during a call trace: The first three Call Trace filters are automatically logged unless the When call trace filter criteria is provisioned and a call meets the criteria, trace information is written to a tracing file. Call trace is enabled using the provisioning interface.A stopMatch
flag in enabled, whereby the number of filters logged may be less than three. The stopMatch
flag will halt filter profile matching once a match is found.callTraceTimer
configuration is provided along with the maxTriggerCount
configuration to limit the logging of call/error filter events in the TRC file. The SBC may be configured to stop call tracing if either of following criteria is met:maxTriggerCount
is reached or,callTraceTimer
expires
The maxTriggerCount
parameter is used to set/reset the Trigger Count, a mechanism used to limit the logging of call/error filter events in the .TRC file.
A system reboot, an application restart or a Live Software Upgrade (LSWU) causes the following call trace filters to go offline:
- callFilter
- mediaPacketCapture
- signalingPacketCapture
To rectify this, disable and then reenable the applicable filters using the following CLI command syntax:
% set global callTrace signalingPacketCapture state disable % set global callTrace signalingPacketCapture state enable % set global callTrace callFilter <filterName> state disabled % set global callTrace callFilter <filterName> state enabled % set global callTrace callFilter <filterName> mediaPacketCapture disable % set global callTrace callFilter <filterName> mediaPacketCapture enable
As a general rule, set Call Trace to ingress mode unless egress mode is mandatory.
Do not turn on the call trace filter to trace all calls unless using level4. The use of the Call Trace feature at other levels significantly impacts the SBC system's network performance. Ribbon strongly recommends configuring Call Filters, which result in small quantities of calls being traced to avoid heavy loads. Also, be sure to delete Call Trace Filters once you have obtained the necessary information.
Use the following criteria to establish a Call Trace Filter:
- Called Party Number
- Calling Party Number
- Contractor Number
- Peer IP Address
- Redirecting Number
- Transfer Capability
- Trunk Group
Use the match
keyword to specify criteria to use to capture call data upon a match. Use the key keyword to specify criteria on which to filter call data.
Call events captured are:
- Call attempt
- Lookup request sent
- Lookup request received
- Alerting received
- Cut-through received
- Call Answer received
- Cut-through complete
- Service established
- Disconnect Request received
- Call terminated
- Wait for more digits
- Extra digits received
- Initiating new attempt
- Signal messages received or sent
- Egress call setup message
- Route lookup failure
- Call blocked
- Policy Request
- Policy Response
- Policy processing of calls in different layers like pre-processor, services, pre-router, router, and post-router.
- SIP signaling events for a single call incoming/outgoing to a particular source/destination IP endpoint
- SIP Protocol Data Units (PDUs) for a single call incoming/outgoing from/to a particular source/destination IP endpoint
- SIP out-of-dialog PDUs such as 18X, BYE, and INFO, regardless of whether they are successfully parsed
The TRC event log level
must be info
for this facility to operate properly.
Use the maxTriggerCount
parameter to limit the logging of call/error filter events in the .TRC file. After configuring and enabling the callFilter
, an ingress/egress call that matches the filter criteria causes the corresponding events to be logged to the .TRC file, as long as the Trigger Count has not been reached. Once the configured Trigger Count has been reached, the matching callFilter
will not be triggered and events associated with the call trace filter will not be logged in the .TRC file. To reenable triggering, you must configure the Trigger Count to a non-zero value less than 65.
Once the maxTriggerCount
is reached or callTraceTimer
expires, the call trace stops. To restart the call trace without changing the callTrace
configuration, use following CLI command:
% request global callTrace action command start
This command is only exposed after callTrace configuration is established. If no entries are present in the callTrace, this command is not available from CLI.
When you enable To capture all packet fragments, create a Signaling Packet Capture
without any configuring any signaling filter (the default behavior), the SBC creates a default filter using all the configured signaling IP addresses and ports. The default filter collects only the first fragmented IP packet and ignores the rest. This happens because the UDP/TCP ports are also included in the default filter, but the IP packet fragments except the first fragment does not include the relevant headers. For the same reason, even if you create a filter which includes only the Local Port Number
, the fragments except the first one are not captured.Signaling Packet Capture
filter with specific Local IP Address
, or Peer IP Address
, or both, and then enable Signaling Packet Capture
.
Command Syntax
Use following CLI syntax to configure call trace filters:
The maximum number of call trace filters allowed is 100.
Global Call Trace Syntax
% set global callTrace callFilter <callFilter_name> callTraceTimer <1-360> errorFilter errorType <any | earlyAttempt | none | outOfDialog | parseError> level4TraceMessages <allMessages | rcvdInvite> maxTriggerCount <0-64> sageFingerprint <disable | enable> sageTracing <disable | enable> signalingPacketCapture devices <mgmt0 | mgmt1 | pkt0 | pkt1 | pkt2 | pkt3> <VLAN tag> filter <signaling trace filter name> signalingPacketCaptureTimer <0-360 minutes> state <disable | enable>
Call Filter Syntax
% set global callTrace callFilter <callFilter_name> key <called | calling | cddn | contractor | peerIpAddress | redirecting | transferCapability | trunkGroup> level <level1 | level2 | level3 | level4> match called <string> calling <string> cddn <string> contractor <string> peerIpAddress <IP address> redirecting <string> transferCapability <audio31Khz | speech | unrestricted | unrestrictedWithTones | video> trunkGroup <trunkGroup_name> mediaPacketCapture <disable | enable> sipRecLegsCapture <disable | enable> state <disabled|enabled> stopMatch <supported | unsupported>
The default value for vlanTag
is "0". If you want to configure VLAN and capture the packets on a specific VLAN, you must provide the devices name and the vlanTag
value.
To configure signaling state:
% set global callTrace signalingPacketCapture state <enable | disable>
Command Parameters
Global Call Trace Parameters
Parameter | Length/Range | Description |
---|---|---|
| 1-23 | The name of a call trace filter to apply to the system trace data log file. This filter may be created, configured, and deleted. See Call Filter Parameters table below for details. |
callTraceTimer | 1-360 | Duration in minutes that call trace is enabled. (default = 180) |
| N/A | Use this object to specify the system-wide call trace error filter trigger criteria.
Note
The system-wide error filter triggers a call trace based on the selected type of error. If you select to trace calls with errors as well as define call filters to trace, the error filter runs in parallel with the call filters. Both may trigger producing a double-PDU output, one for the conditions matched in the call filter, and one for the detected error. |
level4TraceMessages | N/A | Control to limit traces only to INVITE messages.
|
| 0-64 | Maximum number of calls that match the (default = 0) |
sageFingerprint | N/A | Enable this flag to generate an audio fingerprint for the incoming calls that can aid in robocalling campaign identification. The back-end systems (such as Identity Hub) use this feature to identify robocalling campaigns and classify individual calls that belong to a robocalling campaign. For more information, refer to Voice Analytics to Identify Voice Campaigns. Audio fingerprinting is enabled by default. A global configuration option is provided to disable the fingerprinting on the SBC.
|
sageTracing | N/A | Use this flag to allow a sampling of the call flows in the network using the L4 call trace. When the Sage Tracing feature is enabled, the SBC captures a small set of level 4 SIP PDU trace logs automatically in order to help debug field issues. This logging utilizes a small amount of CPU, for example, 1%. This functionality has no impact if Level 4 SIP PDU tracing is already enabled to capture allMessages for all IPs, or the trace files are disabled in the eventLog configuration.
|
| N/A | Signaling Packet Capture configuration.
|
Call Filter Parameters
Parameter | Description |
---|---|
key | Choose one or more active key components for this filter:
|
level | The trace level of this call trace entry.
Note: Level 4 Call Trace Filter conditions/behavior:
For more information on the effect of different levels with respect to the flag |
match | Use this keyword to specify criteria to use to capture call data upon a match. Options are:
Note: For |
mediaPacketCapture | Specifies whether to enable or disable media packet capture settings. These files will be stored as ".pkt"files.
Note
NICE and Media Packet Capture cannot be used simultaneously because the Splitter resource can only be configured for one feature. |
sipRecLegsCapture | Enable this flag to trace and capture packets for SIPREC legs.
For more information on this flag, refer to Call Trace to Capture SIPRec Legs. |
state | Administrative state of this filter:
|
| Use this flag to stop matching filter profiles once a match is found:
|
Command Examples
To enable call tracing based on the called number:
% set global callTrace callFilter CF-1 level level1 state enabled key called,calling match called 978% % set global callTrace callFilter CF-1 mediaPacketCapture enable
To view the recently-enabled call trace filter:
% show table global callTrace callFilter callFilter CF-1 { state enabled; level level1; key called,calling; match { called 978%; } mediaPacketCapture enable;
To enable signaling packet capture:
% set global callTrace signalingPacketCapture state enable
To enable the sageFingerprint:
set global callTrace sageFingerprint enable commit
To disable the sageFingerprint:
set global callTrace sageFingerprint disable commit
Offline Call Trace Analysis Using LX Utility
The TRC files can be analyzed offline using the LX tool, which generates a SIP ladder diagram from the TRC or DBG files. The LX tool is a Windows PC application available free of charge to all customers and partners with a license agreement, and can be downloaded as part of your software download request. For details, refer to LXDOC. LX Main User Interface: