Hitachi NAS (HNAS) Platform Data Collection
HNAS Diagnostics are needed to investigate all HNAS related issues. The Diagnostic bundle contains vital configuration details along with historical data which are necessary for our troubleshooting and investigation.
Most commonly, collecting the Diagnostics from the SMU GUI is the preferred and easiest method. In some cases, the GUI may not be available which there is a CLI collection option for.
- How to Collect HNAS Diagnostic Data from the SMU GUI
- How to Collect HNAS Diagnostics Using Command Line Interface CLI
For certain specific problems additional data may be required to start the investigative process. These are detailed below:
Performance problems
A performance problem is where the HNAS is perceived to be responding slowly or not at all:
- What Information Do I Need to Gather to Allow GSC to Diagnose an HNAS Performance Problem
- What Information Do I Need to Gather to Allow GSC to Diagnose a Hung HNAS System
- HNAS Performance Issues Questionnaire
- How to Collect and Download a HNAS Performance Information Report or PIR
- How To Collect a PIR If HNAS Is Not Configured to Send via Email
- HNAS Data Migrator to Cloud Performance Issue
Security permissions problems
Problems with files not being accessible or being too accessible etc:
Heap memory issues
Heap memory related events/issues may include:
- Failed to allocate panics ("start of "FailedToAllocate" panic"),
- Heap fragmentation ("No big free heap blocks left"),
- Heap related enforced limits:
- Insufficient free heap:
- "There is insufficient heap: dropping new connection from"
- "There is insufficient heap: dropping additional connection request from"
- Insufficient free heap in big sized heap blocks:
- "There is insufficient heap: dropping new connection from"
- "There is insufficient heap: dropping additional connection request from"
- Insufficient free heap:
File (NDMP) replication problems
Please refer to the following knowledgebase article for the information required to investigate HNAS file (NDMP) replication problems:
HNAS resets / HNAS "boot loops"
If a customer (or Remote Ops (Hi-Track)) reports a HNAS reset then you usually just need HNAS managed server diagnostics from after the reset. If the server has been restarted since the reset then the diagnostics for the reset will no-longer be in the last_dblog.txt. In that case you will need to ask the customer to get historical logs.
A boot loop is when the HNAS seems to be continually rebooting without any intervention:
Historical logs collection
Collection of historical logs is not usually required but in some cases, a support engineer may request this to investigate further:
Collect SMU diagnostics from SMU using command line
If for some reason you cannot collect SMU diagnostics from the SMU GUI it is possible to collect them from the SMU command line as follows:
Packet capture of network traffic
We often need packet captures of network traffic in order to progress HNAS issues. The following articles provide advice and guidance on how to go about collecting those: