System status

NAM Server ► Tools ► Diagnostics ► System status

This report lists vital system health indicators for the report server. Use it to find performance issues.

Report columns

  • Status
    A quick overview of the status: red for problems, orange for warnings, and green for normal operation.
  • Shown as health check status
    Indicates whether an alert icon is shown on all reports if a problem occurs with the particular system health indicator.
    A "Yes" value indicates that an orange or red alert icon appears in the upper left-hand side corner of all the traffic monitoring reports, and that the data displayed on the reports may be affected by a system-wide problem.
  • Module name
    Displays the major operational component.
  • Object
    Displays sub-component names.
  • Each indicator is provided with an explanatory comment.

Interpreting a system problem

The module and object names, as well as additional informational messages given for the various system health indicators, are self explanatory, though note that a particular problem may be caused by a variety of reasons. Use the following information to help find and rectify some of the more common problems.

NAM Probe SSL status

The status of the SSL engine initialization for the configured NAM Probes. Shows a list of the observed types of return codes for SSL initialization and the last observed return code. The address of the NAM Probe and the time interval to which the status applies is shown in the Object column.

The following SSL initialization return codes can be reported:

  • ERR_LIB_NULL : unsupported engine - the SSL engine specified in the configuration is not supported.
  • ERR_LIB_OPEN : error opening engine library
  • ERR_LIB_FUNC : error in engine function look-up
  • ERR_LIB_INIT : error initializing card
  • EERR_LIB_AUTH : module authorization error
  • ERR_LIB_OK : SSL engine initialization successful

In case of an unsupported engine, check whether your NAM Probe configuration specifies the correct SSL engine. Which engines are supported depends on the software installed on the NAM Probe. If other errors occur, report them to Customer Support.

NAM Probe errors

The percentage of packets with errors. The address of the NAM Probe and the time interval to which the status applies is shown in the Object column. The status column turns yellow if the percentage is above 0.1% and red if it reaches 0.5%.

If thresholds are exceeded, details to diagnose the problem can be found on the detailed NAM Probe statistical graphs.

NAM Probe interface utilization

Average and peak interface utilization. The address of the NAM Probe and the time interval to which the status applies is shown in the Object column.

Use these figures to assess if the interfaces are overloaded. If average utilization approaches 100%, may mean that at peak times the interface is overloaded. Peak figures of 100% confirm that.

The status column turns yellow if the percentage is above 60% and red if it reaches 80%.

If thresholds are exceeded, details to diagnose the problem can be found on the detailed NAM Probe statistical graphs.

Connection with NAM Probes

Apart from a physical connection problem, this may also indicate configuration problems. For example, it can be caused by an improper port speed and duplex settings, either on the switch or on the NAM Probe. By default, the NAM Probe's communication port is set to the auto negotiation mode. The current NAM Probe communication port settings can be checked by executing the command ethtool eth0 at the operating system prompt on the NAM Probe. If auto negotiation is used and if it fails, it is possible to force the NAM Probe's communication NIC to work in 100 Mbps Full Duplex mode. For more information, see RTM configuration tool (rtminst).

Data processing

Problems with data processing can be caused by missing data files coming from NAM Probes. This can be caused by communication problems – described above.

Database info

Provides basic information on the report server database, such as free space, data space, and reserved space.

License

Provides information on purchased module licenses and their expiration dates.

Memory

This indicator shows the distribution of memory problems, also referred to as memory alerts, since last midnight. The percentage figure for each alert level shows the percentage of time the memory alert was at the given level.

Note

If a given memory alert occurs, the report server may automatically suspend some functions to conserve memory. See the list below for details.

Default thresholds for low memory alert levels

Alert level 1

20% or less of free memory: affected functionality. The user data is not retained in cache and latest hour cache is not used. This may increase the time required to generate reports.

Alert level 2

15% or less of free memory: affected functionality. The report data for new (today's) objects are not calculated in cache. This means that no new monitoring objects (such as users, servers, operations) will be shown on reports for today, because today's reports are calculated from cache. If midnight occurs, the new objects will then be shown for reports for yesterday.

Alert level 3

10% or less of free memory: affected functionality. No new site, server or client definitions created.

Automatic system reboot

If system memory decreases below 5%, the report server will re-start automatically.

Module version information

Lists all the module names with their version numbers.

Processing speed

Low processing speed can be caused by a variety of problems. If problems with processing speed occur, you can, for example, consider the following:

  • Increase the length of the individual monitoring interval. The default length of the monitoring interval is 5 minutes. Increasing this value will reduce the number of chunks of data that need to be transferred and processed. Even though this will cause an increase in the volume of recorded data for each interval, it results in an overall increase of the data processing speed. A disadvantage of this solution is that the data monitoring granularity will also increase.
  • Consider aggregating monitored users. Note that even though you may be monitoring a small number of servers, there may be a very large number of users.
  • Find out if there were any data transmission problems (retransmissions) between NAM Probes and the server. These problems can result in data processing problems.
  • Consider reducing the number of monitored objects such as servers, users, sites, and operations.

Operation data limit (ADS only)

This section displays two lines of information on the number of processed operations:

  • The current operation count and the operation limit, as defined in the server configuration.
  • Information on whether the operation limit has been exceeded in the past. The operation limit per day is determined by the server scalability settings.

Repository (NAM Server only)

Provides information whether the BULK INSERT method of inserting data into a database is used or not. On the local server, in most cases it works (status: true), but on the remote server it requires additional configuration. If the current status of the BULK INSERT is false, it might be a threat to report server performance.

System

Shows basic information on the system such as the report server IP, OS version or the exact time when the server was started.

The names of system components and subcomponents, as shown in Module name and Object, correspond to the module and sub-module names used in the LOW_OPER_4_SYS_MOD alert that can be used to alert you to system problems.