Sun Software - Generic
 
Oracle FS1-2 Flash Storage System
 
Sun Hardware - Generic
 

PLA-Support>Sun Systems>Sun_Other>Sun Collections>SN-OTH: Sun Alert
 

In this Document

Description

Occurrence

Symptoms

Workaround

Patches

History

References

Applies to:

Description

An Extent is a contiguous space in a Drive Group associated with a single Volume. There is an Extent size limitation of 2 Terabytes on the amount of data that can be moved during a data migration or rebalancing in a QoS (Quality of Service) migration. Migration of Extents beyond this limitation will panic the controller, causing the data beyond 2 Terabytes to be lost. If several panics occur, the FS1-2 may also shutdown.

Occurrence

This issue can occur on any Flash Storage System FS1-2 that meets all of the following conditions:

1. System is running software version 06.02.03-0280.01 or earlier.

2. Single Tiered LUNs larger than 2 Terabytes.

3. Free space of 2 Terabytes or more in the Storage Domain and Storage Class.

4. Data migration is triggered by any of the following:

• Excluding a Drive Group from a Storage Domain
• Adding a Drive Group to a Storage Domain with the 'Rebalance Volume Data' option
• Daily Automatic QoS Rebalancing

Symptoms

When encountered, this issue will panic the controller executing the migration. This will generate a log bundle with the following event:

    VOLUMES_HAVE_CHANGED_STATE_COLLECT_LOGS

There will also be the following entry in the Event Log:

    Volumes have changed state in a bad way

with a description of:

    One or more volumes have changed state in a bad way.
    This event is to tell us to collect logs as a result of the changes that took place.

Workaround

The immediate workaround for this issue is to disable Automatic QoS Rebalancing for the Storage Domain(s) and to avoid adding or subtracting Drive Groups to or from a Storage Domain.

For information on how to disable Automatic QoS Rebalancing, please reference the following document:

• Document 2045089.1 "FS System: How to Enable or Disable Automatic QoS (Quality of Service) Rebalancing"

Customers who meet all of the remaining conditions described above should open a Service Request to have Oracle Support verify their exposure. A current log bundle will be needed. Please reference the following document for additional information:

• Document 1943289.1 "FS System: How to Attach a Manual Log Bundle to an Existing Service Request"

For customers that cannot provide a log bundle, Oracle has added a utility, codAudit, to the "2of2" in patch 23132401 (Release 6.2.3-0280.02). It can be used to identify 'Single Tiered' LUNs with Extents larger than 2TB. Customers should contact Oracle support for assistance in running this utility as well as its interpretation.

Resolution

This issue is addressed in the following release:

• Oracle FS1-2 Flash Storage System 6.2.3.280.2 (Patch 23132401) or later

Patches

 <Patch 23132401>

History

28-Apr-2016: Document released, status Resolved
12-May-2016: Minor update, no change in content

Internal Section: Comments:

Controller Panic string:

    2016-04-15T01:32:10.116926695Z 74191176 508002000158C1A1 11 0 546 EMERG STRACESTR
    /home/build/builds/axiom/060203-028000/src/slammer/bs/cfg/bs_cfg_shared.c 4424 bs_cfgSharedCmdPreTimeout
    BS BS "PANIC: cmdSlot:0x0 cmdCode:0x35 cmdStep"

Details on the utility to identify this issue are detailed in Document 2129692.1

Questions regarding this document should be addressed to
sunalertpublication_us_grp@oracle.com and copy the
submitter/responsible Engineer listed below.

Internal Contributor/Submitter: bob.deguc@oracle.com
Internal Eng Responsible Engineer: lon.stowell@oracle.com
Oracle Knowledge Analyst: jeff.folla@oracle.com
Internal Eng Business Unit Group: Flash Storage
Internal Associated SRs:
Internal Resolution Patches: 23132401

References

This solution has no attachment