VSM - Info on MIMs, SIMs, Statesaves Event logs and DLF

Asset ID:	1-71-1294345.1
Update Date:	2011-04-18
Keywords:

Solution Type Technical Instruction Sure

Solution 1294345.1 : VSM - Info on MIMs, SIMs, Statesaves Event logs and DLF

Applies to:

Sun StorageTek VSM System - Version: 4 to 5C - Release: 4.0 to 5.0
Information in this document applies to any platform.

Goal

Explain what MIMs, SIMs, Event logs, Statesaves and DLF are and how they are generated.

Solution

Overview
The Virtual Tape Storage Subsystem(VTSS) is connected to the Service Delivery Platform (SDP),a network Ethernet device that communicates with an Oracle-Sun Remote Resolution Center (RRC) maintenance server. The machine-initiated maintenance (MIM) function automatically reports whether the VTSS is inoperable, degraded, subject to potential failure, or ready to offload event log data.

Once a problem is identified, Predictive Subsystem Analysis (PSA) begins the fault isolation process, fences defective FRUs, and compiles failure reports and other related information into a Composite Failure Event (CFE) database to assist in identifying and fixing problems. It then relays a System Information Message (SIM) to the host system to alert the customer. This SIM message is an SLS message for VSM and an alert (service,moderate or serious) for SVA.

Machine-Initiated Maintenance (MIM)
A machine-detected error or fault condition occurs, it initiates a call-home sequence from the supported machine to the associated remote support center server (RRC). Messages about actual or potential unit problems are sent as they occur; critical messages are repeated at 24-hour intervals until unit is serviced. MIM message categories are numbered from 00-07. Error event and log data is available to the support representative for analysis.

The following explains two main types of MIM alert messages generated by the VTSS:

Unit Problem — invoked as needed (asynchronously) if a device is inoperable (Category 00), running with degraded performance (Category 01), or subject to potential failure (Category 02). Messages include the fault symptom code (FSC) assigned to the failure and the content of the host SIM.
Event Log Download — invoked at customer-defined intervals to download event logs, FRU replacement information, and other threshold data (Category 05).

When a call home fails, because the device could not send a MIM in a reasonable number of retries, it will generate FSC7351, indicating the MIM call failed. The possible causes of FSC7351 are then investigated by support. Also see doc id 1290862.1 for more info on processing FCS7351.

NOTE: The ServiceTek Plus program, which generated MIMs via modem, has been terminated, devices still attached to modems will see SIMs with FSC7351, instead of the actual FSC.

Service Information Messages (SIMs)
When a threshold value for related failure events is exceeded, a PSA rule is invoked to
isolate the CFE. A suspect FRU list (SFL) is compiled based on the general domain of the
failure as defined by related fault symptom codes (FSCs), and from failure report pathnode
information. Fault isolation begins when one of these thresholds is met:

• Persistence Threshold – maximum allowable number of error recovery attempts, beyond
which VTSS operation is affected
• Severity Threshold – maximum allowable level of problem severity, beyond which VTSS
operation is affected.

NOTE: A MIM is always attempted, but issuance to an Oracle-Sun remote server is customer-controlled. A SIM is sent to the mainframe host console only if the isolated FRU can cause a performance degradation in the VTSS.

All SIMs are reported to the host system console, but operators determine which SIMs are
displayed at the console by having the CSE set a display level at the Access Control
screen as follows:
• Level 0 – displays ‘Service’, ‘Moderate’, ‘Serious’, and ‘Acute’ SIMs (all levels)
• Level 1 – displays only ‘Moderate’, ‘Serious’, and ‘Acute’ SIMs
• Level 2 – displays only ‘Serious’ and ‘Acute’ SIMs
• Level 3 – displays only ‘Acute’ SIMs.

Event Logs and Statesaves
Event log files are a record of diagnostic events, failure reports, SIMs, fence events, and
FRU replacement information.

A statesave is a snapshot of VTSS status at the moment a microcode, software, or hardware
problem occurs. Statesaves are initiated by Check0 software on the ISP and IPX
cards, and by software on array drive cards whenever a problem is detected.
Statesaves record maintenance and failure event data, and store it on the VTSS support facility ISP hard drives, while customer data remains undisturbed and fully accessible on the VTSS array drives.

The largest statesaves (11 files) are 24MB uncompressed. Compression reduces file size
by roughly two-thirds. Statesave data is automatically compressed and saved in files on
VTSS support facility ISP hard drives, which have space allocated for this purpose.

Note: Statesaves do not record or in any other way affect customer data and do not store any data
on customer array drives.

If a VTSS is linked to a Oracle-Sun remote maintenance server, log files and state data can be downloaded automatically at customer-defined intervals, or manually by a CSE. The data is transferred to a network server software simulator, where Sun engineers review the data to isolate a problem, determine a cause, and issue a fix to the field.

Download Facility (DLF)
A VTSS facility that formats event log data and connects to a RRC PC to offload event log and MIM information as directed by the VTSS PSA facility and subsystem internal timers.

PSA generated information is regularly downloaded to the Oracle-Sun remote server as part of
normal event log MIM processing to allow ongoing analysis and correction of problems

Attachments

This solution has no attachment