Diagnostic data collection
If you need to investigate problems or issues with your system, there are three ways for diagnostic data to be gathered from it.
- Call Home
- If enabled for certain types of issues, diagnostic data will be automatically gathered and sent
to IBM for analysis. No further work is required for this first pass of diagnostics to be included
in the Call Home generated service request. Cases are opened automatically for the following event
- Diagnostic data for other system issues
- There are other types of system hardware and software issues which do not automatically
generate service requests by default. The system will still automatically gather diagnostic data
for many of these, including the following:
To display log file path, run the ap issues <issue_id> command, as in the example below:
[root@node0101 ~]# ap issues 1008 General Information ID : 1008 Date : 2018-12-18 01:24:14.456407 Close Date : None Target : hw://hadomain1.node3.nic_eth2 Target Type : nic_eth Severity : WARNING Title : Unhealthy component detected Stateful : 1 Referenced Alert ID : None Classification Group : HW Type : HW_NEEDS_ATTENTION Reason Code : 201 Processing Status State : DELIVERED Log Collection Status : COLLECTED SMTP Status : FAILED SNMP Status : UNKNOWN Call Home Status : NOT_APPLICABLE Service Request SRID : None SR Status : None Collected Logs Log File Path : /var/log/appliance/platform/management/alerts/alert_1008.tar.gz Log File Node : hadomain1.node1 Log File Checksum : de963ba04512f7f9136899d1d976771f Additional Data: Message : Status on component nic_eth at location hw://hadomain1.node3.nic_eth2 is WARNING bdf : 0002:01:00.0 creator_id : node@hw://hadomain1.node3 driver : bnx2x driver_version : bc 7.10.4 errdata1 : WARNING: Port C6-P1 within the same physical interface type on the card (fab1, bdf 0002:01:00.0) (links to fabsw01a, port 6) link state is DOWN led : off location : hw://hadomain1.node3.nic_eth2 position : C6 reason_checksum : 9b31aa9728cde678222686e436eeb43a serial : status : WARNING type : nic_eth type_desc : Quad Port 1Gbit/10Gbit NIC version : bc 7.10.4 Generated: 2019-03-21 14:03:20
- Diagnostic data outside of any specific platform-reported issue
- To provide IBM support with diagnostic data not tied to either of the above cases, the system includes a single data gathering utility, called apdiag. Data collection can be done based on targeting a specific system component or an ever growing collection of general symptoms. If you open a support ticket with IBM directly, you might be asked specifically to run apdiag with a defined set of arguments or parameters. When you execute apdiag, the responding output to the console will include path and file name of the gathered diagnostic tar-ball. For more information on the apdiag command, see Running diagnostics by using the apdiag command.
The apdiag command has a plugin for Netezza Performance Server logs. That plugin calls
nzlogcollectorinside the NPS container. You can use the following command to collect NPS-related logs:
If you want to pass options to
apdiag collect --components ips/logs
nzlogcollector, run the following command:
In this example, the -symptom crash option is passed.
apdiag collect --components ips/logs --comp-param ips/logs: "-symptom crash"