Contact Center Trace Levels

Trace Levels

Support personnel who debug the Unified Communication solution need not know the details of trace levels across Unified Communication solution applications. If debugging a problem requires more detailed debug information, three levels of trace setting exist that can map internally to a particular application or application component. The intent is not to revamp existing trace setting values but to map the levels to the relevant and existing trace settings for a particular component. Use the following defined trace levels with the Diagnostic Framework Portico and the Unified System CLI tools.

The following trace levels are defined for the Unified Communication solution:

Trace Level

Trace Value

Description

Default

0

Default install, should have no or minimal performance impact

Warning

1

Log more detailed (plus default level) trace messages, small performance impact

Error

2

Log more detailed (plus warning and default level) trace messages, medium performance impact

Debug

3

Log most detailed (plus error and warning and default level) trace messages, high performance impact.


Note

If you have enabled the debug trace levels, disable them after collecting the logs to avoid an unnecessary performance impact to your solution.

If the trace level does not match any of the defined levels, it displays custom (99).

Most EMSTraceMasks are based on this registry key: HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\<Instance>\<Component>\EMS\CurrentVersion\Library\Processes\<process>\EMSTraceMask

Get and Set trace level and collect trace files are supported only for the following processes.


Note

If the trace mask is the same for multiple levels, the GetTraceLevel returns the highest level. For example, GetTraceLevel returns Level 3 for Logger/baimport.

Trace–All Nodes

You can set trace levels on these processes:

Process

Level 0

(Default - Error)

Level 1

(Warning)

Level 2

(Informational)

Level 3

(Debug)

NM

0x00

0x0F

0x0F

0xFF

NMM

0x00

0x0F

0x0F

0xFF

The Diagnostic Framework does not support the Administrator Workstation.

Trace–Administration and Data Server

You can set trace levels on these processes for the Administration and Data Server:

Process

Level 0

(Default - Error)

Level 1

(Warning)

Level 2

(Informational)

Level 3

(Debug)

CONFIGLOGGER

0x00

0x0F

0xFF

0xFFF

CMSNODE

0x00

0x00

0x00

0xFFFFFFFF

CMS_JSERVER

0x00

0x00

0x00

0xFFFFFFFF

REPLICATION

0x00

0x0F

0xFF

0xFFF

RTCLIENT

0x00

0x0F

0xFF

0xFFF

RTDIST

0x00

0x0F

0xFF

0xFFF

UPDATEAW

0x00

0x0F

0xFF

0xFFF

ISEMAN

0x00

0x00

0x00

0x01

Note

The minimum and default trace level for the CMS, CMSJServer and ISE components is 2.

Trace–Router

You can set trace levels on these processes for the Router:

Process

Level 0

(Default - Error)

Level 1

(Warning)

Level 2

(Informational)

Level 3

(Debug)

Notes

Agi

0x00

0x01

0x07

0x3F

—

ccagent

0x00

0x03

0x0F

0xFF

—

dbagent

0x00

0x01

0xFF

0xFF

—

DBWORKER

0x00

0x01

0xFF

0xFF

—

mdsproc

0x00

0x07

0xFF

0xFF

—

rtr

0x00

0x0F

0xFF

0xFFF

—

ROUTER *

Route Requests

Route Requests

- Network VRU

- Trans Route

- VRU Bank

- CIC Request

- Script Select

- Call Queuing

- Agent changes

- Call Type Real Time

In case of RTRTRACE or RTRTEST

Note: If you restart the Router process, the settings return to the default level.

Trace–Logger

You can set trace levels on these processes for the Logger:

Process

Level 0

(Default - Error)

Level 1

(Warning)

Level 2

(Informational)

Level 3(Debug)

BAIMPORT

0xFF

EMSUserData = 0xFFFF

0xFF

EMSUserData = 0xFFFF

0xFF

EMSUserData = 0xFFFF

0xFF

EMSUserData = 0xFFFF

CAMPAIGN MANAGER

0xFF

EMSUserData = 0xFFFF

0xFF

EMSUserData = 0xFFFF

0xFF

EMSUserData = 0xFFFF

0xFF

EMSUserData = 0xFFFF

CONFIGLOGGER

0x00

0x0F

0xFF

0xFFF

CSFS

0x00

0x00

0x00

0xFF

CW2KFEED

0x00

0x00

0x00

0x07

DTP

0x00

0x04

0x06

0x0F

HISTLOGGER

0x00

0x0F

0xFFF

0xFFF

RECOVERY

0x00

0x0F

0xFFF

0xFFF

REPLICATION

0x00

0x0F

0xFFF

0xFFF

Trace–Peripheral Gateway

You can set trace levels on these processes for the PG:

Process

Level 0

(Default - Error)

Level 1

(Warning)

Level 2

(Informational)

Level 3

(Debug)

JTAPIGW *

JT_JTAPI_EVENT_ USED

JT_TPREQUESTS

JT_PIM_EVENT

JT_ROUTE_ MESSAGE

JT_CONNECTION *CONF*

JT_JTAPI* JT_HEX JT_ROUTE* JT_TERM* JT_LOW*

JT*

MDSPROC

0x00

0x07

0x0F

0xFF

MSGIS

0x00

0x00

0x00

0x3F

OPC **

Default, cstacerEMSTraceMask = 0x40

agent, inrcmsg, closedcalls, tpmsg, routingEMSTraceMask = 0x40 Note: Remove "default" tracing set in Default(0) level, but include cstacer.

Calls, NCT, simplified EMSTraceMask = 0x40Note:Remove "default" tracing set in Default(0) level, but include Level 1.

Missingdata, halfhour EMSTraceMask = 0x40Note: Remove "default" tracing set in Default(0) level, but include Level 2.

PGAGENT

0x00

0x03

0x0F

0xFF

EAGTPIM *

tp* precall *event csta* call_object teld_agent_state opcrequest

periph* jtapi_dialed*

autoconfig* teld* call_match_timing timer*

lock* universal* service* threadid jtapi*

VRUPIM

EMSTraceMask= 0x0

EMSUserData= 0x71f7e0

EMSTraceMask= 0x0

EMSUserData= 0x71f7e0

00000000000000000000

00000000000000000000

00000000000000000000

000000000180

EMSTraceMask= 0x0

EMSUserData= 0x71f7e0

00000000000000000000

00000000000000000000

00000000000000000000

000000000180

EMSTraceMask= 0x0EMSUserData= 0xf1fff0

0000000000000000000

0000000000000000000

0000000000000000000

000000000000180

ACMIPIM

EMSUserData = (hex) 01, 7f, 46, 00, 00, 00, 00, 00, 00, 00, 00, 01, 00, 00, 00, 00, 00, 3f, ff, ff, ff, 67, cf, d7, fd, ef, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, f0, fa

For reference, this is the default + all_peripherals

EMSUserData = (hex) f5, 7f, 46, 00, 00, 00, 00, 00, 00, 00, 00, 01, 00, 00, 00, 00, 00, 3f, ff, ff, ff, 67, cf, d7, fd, ef, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, f0, fa

For reference this is level 0 + timer events

EMSUserData = (hex) f5, 7f, c6, 00, 00, 00, 00, 00, 00, 00, 00, 01, 00, 00, 00, 00, 00, 3f, ff, ff, ff, 67, cf, d7, fd, ef, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, f0, fa

For reference this is level 1 + Monitor Item traversal

EMSUserData = (hex) f5, 7f, f6, 00, 00, 00, 00, 01, ff, ff, fe, c1, 00, 00, 00, 00, 00, 3f, ff, ff, ff, 67, cf, df, fd, ef, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, fe

For reference this is level 2 + locks + socket data

ARSPIM *

tp* precall *event csta* call_object teld_agent_state opcrequest

periph*

autoconfig* teld* call_match_timing timer*

lock* universal* service* threadid

MRPIM

EMSUserData = 0x00

Procmon:

> trace mr* /off

Note

 

Level 2 is the default level for MRPIM.

EMSUserData = 0x40

Procmon:

> trace mr* /off

> trace mr_msg_comm_session /on

EMSUserData = 0x58

Procmon:

> trace mr* /off

> trace mr_msg_comm_session /on

> trace mr_*_mr /on

EMSUserData = 0x5F

Procmon:

> trace mr* /off

> trace mr_msg_comm_session /on

> trace mr_*_mr /on

> trace mr_*_inrc /on

> trace mr_*_csta /on

Avaya Aura PIM ( Symposium)

Default

Default

Default

Tpcsta*, call*, csta*

AAS

Default

EMS_TRACE_GENERAL

EMS_TRACE_CONAPI

EMS_TRACE_SEI

EMS_TRACE_AASDRIVER

EMS_TRACE_MSL

Aspect PIM

Default

Default

Default

Tpcsta*, call*, csta*, app*, pim*, rtb*

Avaya ECSPIM

Default

Default

Default

Bri*, agent*, call*, csta*, cms*, tp*, 3pdc*, route*, cv*

Avaya TAESPIM

Default

Default

Default

agent*, call*, csta*, cms*, tp*, monitor*, route*, value*, tsapi*

CTISRVR

0x000000f0

0x000000f6

0x000000fe

0x000000ff

CTIOS SERVER NODE

0x00060A0F

0x00240A2F

0x00260A2F

0x002E0A2F

BADIALER

EMSTraceMask= 0x0000003f

EMSUserData= 0xFFFF

EMSTraceMask= 0x0000003f

EMSUserData= 0xFFFF

EMSTraceMask= 0x0000007f

EMSUserData= 0xFFFF

EMSTraceMask= 0x0000007f

EMSUserData= 0xFFFF

Trace–Web Setup

You can set trace levels on the Web Setup process.

Process

Level 0

(Default - Error)

Level 1

(Warning)

Level 2

(Informational)

Level 3

(Debug)

Web Setup

EMERGENCY

CRITICAL

WARNING

DEBUG

Websetup uses log4j.net for logging. Websetup uses a XML file (log4j.xml) through which you can set the trace levels. The XML file also contains other information you require for logging.

You can find the log4j.xml file here: <InstallDrive>:\icm\tomcat\webapps\setup\WEB-INF\classes\log4j.xml

Trace–Diagnostic Framework

You can set trace levels on the Diagnostic Framework.

Process

Level 0

(Default - Error)

Level 1

(Warning)

Level 2

(Informational)

Level 3

(Debug)

Notes

Diagnostic Framework

Info

Info

Info

Debug

When the Diagnostic Framework receives a request for its own trace level, if the trace level is at Info, Level 2 is returned. When the Diagnostic Framework receives a request to be set for Level 0, Level 1, or Level 2, the trace level is set to Info.

Trace–SADLib

You can set trace levels on SADLib.

Process

Trace Level

Trace Value

Description

SADLib

0

NO

No logs are generated.

1

WARNING

The Warning and Error logs are generated.

2

INFO

The Warning, Error, and Info logs are generated.

3

DEBUG

The Warning, Error, Info, and Debug logs are generated.

Reference Tables

Regarding JTAPI logging,

Jtapi uses EMSUserData value default ( OX49E2) to control logging, we can enable/disable JT_ bits using procmon (trace command). JT_ maps to below tables values which is masked with EMSUserData to control tracing.

Process

Level 0

(Default - Error)

JT_HEARTBEATS

1

JT_TPREQUESTS

2

JT_HEX

3

JT_JTAPI*

4-7

JT_CONNECTION

8

JT_PIM_EVENT

10

JT_ROUTE*

11-12

JT_LOW_LEVEL*

13-14

JT_CONF_XFER

15

JT_TERM_EVENT_RTP

16

JT_DEBUG

17

Regarding MRPIM logging

Similar to jtapi, mrpim default EMSUserData is 0xD8, we can enable/disable mr* bit using procmon(trace command)

Process

Level 0

(Default - Error)

mr_msg_config

1

mr_msg_comm_session

2

mr_heartbeat_messages

3

mr_msg_incoming_mr

4-7

mr_msg_outgoing_mr

8

mr_msg_incoming_inrc

10

mr_msg_outgoing_inrc

11-12

mr_msg_outgoing_csta

13-14

mr_function_call

15

mr_outgoing_opc

16

EMS Log Compression

To collect logs that span a greater period of time, EMS log files from the CTI OS Server and the following PG components are zipped:
  • Router
  • OPC-CCE
  • OPC-TDM
  • CTISVR
  • EAGTPIM
  • JTAPIGATEWAY
  • VRUPIM
  • BADIALER
  • MRPIM
  • CTI OS Server

Note
These are the only components that currently support EMS log compression.

Dumplog

Dumplog handles the compressed EMS files and can be used in the general way. Dumplog looks for gzip.exe in <Install Drive>\icm\bin to unzip compressed EMS files before dumping logs. To dump logs from compressed EMS files (with .gz extension) outside of a PG or CTI OS Server, unzip the EMS files before you use dumplog.

EMS File Compression Control

Use the EMSZipCompressionEnabled registry key in \EMS\CurrentVersion\Library\Processes\<process name> to enable or disable compression of EMS log files. Do not modify this registry key. This key takes effect only on components that support EMS file compression.

Other Registry Keys

The following two other registry keys are also available in …\EMS\CurrentVersion\Library\Processes\<node name>

  • EMSZipFormat
  • EMSZipExtension

Note
Do not modify these registry keys.

Set Router Tracing

To set the Unified ICM/CCE Router, use the Router Trace utility. This utility is a single-form Windows GUI utility that is loaded on the Unified ICM/Unified CCE server.


Note
The router process must be running before you set router tracing or attempt to retrieve router traces. This requirement applies to any tool which sets or retrieves router tracing, including, for example, RTTEST, RTRTRACE, and System CLI. (The router process must be running because the router trace settings are only stored in memory).

Note

Business Hours tracing has to be set or reset using RTR Trace utility or RTTEST utility only. Setting the trace level for Router in Diagnostic portico will not change the Business Hours setting.

All trace settings using "RTRTRACE" take effect immediately in the Router.

You can observe specific status of call routing, call type, skill group, and schedule target variables using the following RTTEST command: rttest /cust <instance>/node <node>.

You can monitor a particular variable and determine if its value is incremented or decremented by enabling the RTTEST "watch" command on router logs: rttest: watch <variable>.

The logs are written only when value is changed.

Procedure

Step 1

Connect to server through remote desktop or go to local console.

Step 2

Run command C:> \icm\bin\rtrtrace to invoke RTRTRACE from ICM\BIN

Figure 1. Router Trace Utility

Step 3

Set trace options.

When a call routing failure occurs, the basic traces should at the minimum be "Route Requests" and "Translation Route" if you use translation routing.

Also, enable the other tracing depending on the specific problems you see.

For... Enable...
Any type of VRU Network VRU Tracing
NAM-CICM (Hosted) CICR Requests
Suspected queuing issues Call Queuing
Reporting events problems Reporting Events
Agent issues Agent Changes
Business Hours problems Business Hours

How to Set OPC Tracing

To set Unified ICM/Unified CCE OPC tracing, use the OPCTEST utility. This is a command-line utility and you require remote desktop or local console access.

Command Syntax (launch): 

C:> opctest /cust <instance> /node <node>

Where <instance> is the Unified ICM/Unified CCE instance name and <node> is the desired node name (for example, /cust cust1 /node PG1A).

After you invoke the instance name, you are presented with an opctest: prompt where you can enter commands according to the syntax expected. To display all commands, enter a "?" at the opctest: prompt. However, OPCTEST is a powerful utility and if you use it incorrectly, it can have a negative effect on a production system in operation. Do not run a command against a production system unless you are absolutely certain of the impact it can introduce.

Use the following commands to alter default trace levels. Before using the commands, understand your current utilization to ensure there is sufficient capacity to accommodate the added tracing.

General Diagnostics

opctest:debug /on

Diagnosing Network Transfer Issues

opctest:debug /on
opctest:debug /NCT

Note
Dialog ID in network transfer call flows appear as a negative number in OPC logs. This is applicable for other components like Router and VRU PIM as well. This is to ensure that Router generated DialogID's do not conflict with DialogID's originated from the pre-routing client (also referred to as NIC DialogID). This pre-routing client can either be a NIC or a VRU PIM integrated with CVP. While the router-assigned DialogID can be any unique number with the most significant bit set, the router will use the expired NIC DialogID value as a troubleshooting aid.

Diagnosing Multimedia Issues

opctest:debug /on

opctest:debug /task /passthru

Diagnosing VRU PG Issues

opctest:debug /on
opctest:debug /passthru

The default is:

opctest:debug /routing /agent /closedcalls /cstacer /rcmsg /tpmsg /simplified /inrcmsg

and 

EMSTracemask = 0x40

EMSTracemask is reset in the Windows registry.

TAC directs you to alter or add additional tracing based upon the analysis of collected logs.

How to Restore Default Trace Levels

opctest:debug /on

This parameter turns on the /default tracing, modifies the EMSTracemask to 0x40, and turns off all other enabled tracing.

How to Display Trace Levels

opctest:debug /showtrace

This parameter displays current trace levels enabled on the peripheral.

How to Set Unified CCM PIM Tracing

To reset trace levels with the Unified Communications Manager Peripheral Interface Manager component (for example, "EAGTPIM"), use the ProcMon (process monitoring) utility. This is a command-line utility and you require remote desktop or local console access.

Table 1. Setting Unified CCM PIM Tracing

Command Syntax (launch)

C:> ProcMon <instance> <node> pim<pim number>

Example

C:> ProcMon acme PG1A pim1

Commands

>>>debug /on

How to Set JTAPI Gateway Tracing

To reset trace levels for the Unified Contact Center JTAPI (Java Telephony Applications Programming Interface) Gateway component (for example, "JTAPIGW"), use the ProcMon (process monitoring) utility. This is a command-line utility and you require remote desktop or local console access.

Table 2. Setting JTAPI Gateway Tracing

Command Syntax (launch)

C:> ProcMon <instance> <node> jgw<jtapigw number>

Example

C:> ProcMon acme PG1A jgw1

Commands

>>>trace* /off

>>>debug /on

How to Set JTAPI Gateway Default Tracing

The default tracing for JTAPI gateway consists of a set of tracing levels that currently exist.

To enable only the default tracing, enter the following commands in ProcMon:

trace * /off

Note: debug /on does not turn off non-default tracing so you need this first.

debug /on This enables only default tracing.

To turn off debug tracing, enter the following command in ProcMon:

debug /off This turns off only default tracing. All other tracing is not affected.

How to Set Contact Sharing Tracing

The Contact Sharing service is an optional process which runs under the main Router service (when enabled through web setup on an Contact Director deployment). Tracing levels for this process are set through a ProcMon connection as detailed in the following table.

Table 3. Setting Contact Sharing Tracing

Command Syntax (launch)

C:>ProcMon <customer name> <ra or rb> <csn> <systemname>

Example

C:\>ProcMon bos01 ra csn boston-p-cc

Commands

>>>trace <bit> /on

>>>trace <bit> /off

Note

"ra" and "rb" = Router A or B

"csn" = Contact Sharing Node

How to Set CTI Server Tracing

To reset trace levels with the Unified ICM/Unified CCE CTI Server (for example, CTI Gateway or CG), use the Microsoft Registry Editor (regedit) to modify the trace mask saved in the Windows registry.

Table 4. Setting CTI Server Tracing

Registry Key

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\<instance>\<CG#A/B>\EMS\CurrentVersion\Library\Processes\ctisvr

Example

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\acme\CG1A\EMS\CurrentVersion\Library\Processes\ctisvr

Item

EMSTraceMask

Value

F0 (hex)

This is the default value. The value of F0 provides sufficient tracing information to troubleshoot most issues.

Setting CTI Server Default Tracing

The default tracing level for CTI Server is EMSTraceMask = 0xF0. Do not enable any other tracing at the default trace level. EMSUserData should be NULL.

ProcMon debug commands:

debug /on sets the EMSTraceMask to the default value of 0xF0 and NULL out EMSUserData. No other command is needed to set default tracing.

debug /off sets EMSTraceMask to 0x00 and NULL out EMSUserData.

Setting CTI OS Tracing

Resetting trace levels with the Unified ICM/Unified CCE Cisco Computer Telephony Integration Option (CTI OS) is accomplished by altering the trace mask saved in the Windows registry. Use the Windows REGEDIT utility to change this numeric value.

Table 5. Setting CTI Server Tracing

Registry Key

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\<instance>\CTIOS\EMS\CurrentVersion\Library\Processes\ctios

Example

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\acme\CTIOS\EMS\CurrentVersion\Library\Processes\ctios

Item

EMSTraceMask

Value

60A0F (hex)

Increasing the trace levels (other than the Default 0x00060A0F) impacts the CTI OS Server performance. You must revert High Tracemask to the default trace levels after collecting the required logs.

Levels

Level 0: 0x00060A0F

Level 1: 0x00240A2F

Level 2: 0x00260A2F

Level 3: 0x002E0A2F

Setting VRU PIM Tracing

Resetting trace levels with the Unified ICM/Unified CCE VRU Peripheral Interface Manager (PIM) is accomplishing by altering the trace mask and user data values saved in the Windows registry. Use the Windows REGEDIT utility to change these numeric values.

Table 6. Setting VRU PIM Tracing

Registry Key

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\<instance>\PG<PG number><A or B>\EMS\CurrentVersion\Library\Processes\pim<pim number>

Example

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\acme\PG2A\EMS\CurrentVersion\Library\Processes\pim1

Item

EMSUserData

Value

7F F7 E0 (hex)

Item

EMSTraceMask

Value

0 (zero)

When you collect the trace logs, collect both VRU PIM trace logs and the VRU trace capture file. To obtain VRU trace capture files, run the VRUTRACE tool in the following directory:

\icm\<inst>\<pg><pg number><a or b>\vrucap

For example: \icm\acme\pg2a\vrucap

Setting VRU PIM Default Tracing

The default tracing for VRU PIM consists of a set of tracing levels that currently exist.

ProcMon debug commands:

debug /off turns off all tracing

debug /on enables default tracing only and turns off any previously enabled tracing

Setting Outbound Option Tracing

The utility tools provide centralized control to set up each component trace level. Additionally, you can manually modify the registry key values.

How to Reset CampaignManager Tracing

To reset CampaignManager trace levels, use the Microsoft Registry Editor (regedit) to modify the trace mask saved in the Windows registry.

Registry Key:

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\<instance>\LoggerA\EMS\CurrentVersion\Library\Processes\CampaignManager

Example:

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\m3pc1\LoggerA\EMS\CurrentVersion\Library\Processes\CampaignManager

How to Reset baImport Tracing

To reset baImport trace levels, use the Microsoft Registry Editor (regedit) to modify the trace mask saved in the Windows registry.

Registry Key:

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\<instance>\LoggerA\EMS\CurrentVersion\Library\Processes\baImport

Example:

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\m3pc1\LoggerA\EMS\CurrentVersion\Library\Processes\baImport

How to Reset Dialer Tracing

To reset Dialer trace levels, use the Microsoft Registry Editor (regedit) to modify the trace mask saved in the Windows registry.

Registry Key:

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\<instance>\Dialer\EMS\CurrentVersion\Library\Processes\baDialer

Example:

HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\m3pc1\Dialer\EMS\CurrentVersion\Library\Processes\baDialer

Trace File Retention Settings

You can modify several Windows registry values to adjust the trace log retention parameters, for example, increase the amount of trace data – extend the trace retention window. To modify the trace log parameters, use the Microsoft Registry Editor (regedit).

Unified ICM/Unified CCE Event Management System (EMS) tracing is stored in a binary format in a set of files in a directory on the local drive following a specific structure.
Trace log file location
[Drive]:\icm\<instance>\<node>\logfiles
Example
C:\icm\acme\pg1a\logfiles
Trace log file names
Process_YYMMDD_HHMMSS.ems
Example
opc_090713_123025.ems

This is an OPC trace log file that was created 13 July, 2009 at 12:30:25.

Under the control of the Event Management System, the following rules apply while traces are written to the trace log files:
  • If the size of this file is greater than or equal to the maximum (configured) size that a single EMS trace log file is allowed, the file is closed and a new file is created.
  • If the maximum number of trace log files for this process is greater than the maximum (configured) number of trace log files, then the oldest trace log file is deleted.
  • If the total combined size of all process trace log files is greater than or equal to the maximum (configured) total size of all process trace log files, then the oldest trace log files are deleted until the total size is less than the configured maximum size.

Registry Items

You can change the following registry item values to increase or decrease the amount of disk space allocated for a single process.

Registry key
HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\<instance>\<node>\EMS\CurrentVersion\ Library\Processes\<process>
Example
HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\acme\PG1A\EMS\CurrentVersion\Library\ Processes\opc
Items
EMSLogFileMax

The maximum size, in bytes, of a single trace log file for this process.

EMSLogFileCountMax

The maximum number of trace log files permitted for this process.

EMSAllLogFilesMax

The total space allowed for all trace log files (combined size) for this process.


Note

EMSLogFileMax multiplied by EMSLogFileCountMax may be greater than EMSAllLogFilesMax and it often is by default; this is to ensure trace log files created by frequent process restarts (where a number of small trace log files are created) are not lost when the max count is exceeded but very little disk space is used. EMSAllLogFilesMax is used to guarantee that under any circumstances, the maximum amount of disk space allocated is never exceeded.

The default values of these items are evaluated with every release of the Unified ICM/Unified CCE to determine the optimal limits based on disk usage of the application and typical disk capacity of servers available at the time of release. In nearly all cases, the default values are increased over time as disk drive sizes increase.

Router Full Dump Enabled by Default

In Release 10.0(1), router full dump is enabled by default. This change to the default setting is intended to provide more information to help troubleshoot the issue if a critical process crashes.

The registry key that sets this default is EMSGenerateSmallMemoryDump, which is located here: HKLM\SOFTWARE\Cisco Systems, Inc.\ICM\<instanceID>\RouterA\EMS\CurrentVersion\Library\Processes\rtr.

The value of 0x20000032 for this registry key indicates a router full dump will be generated. To generate a mini-dump rather than a full dump, change the last digit to 1 so that the value is 0x20000031.

The middle six digits of the registry key value determine the number of dumps that are kept. If, for example, these digits are all zeros (0x20000002), CCE by default keeps only the last five generated dumps for any component. For the CCE router, however, only the three latest dumps are kept (0x20000032) by default, since the size of the full dump can be up to 2GB. Keep disk space usage in mind when selecting the number of dumps you want to retain.

If the CCE Router crashes, provide the full dump (the .mdmp file generated under the icm\<instanceid>ra\logfiles directory), the PDB file (for example, router.pdb) from the icm\bin directory, and the associated executable (router.exe) from the icm\bin directory.


Note

The registry key value is set to the default value of 0x20000032 on a new install or upgrade, or whenever you run Web Setup.