topic Re: HSJ80 Controller Question in Operating System - OpenVMS

HSJ80 Controller Question

MJ26 — Mon, 16 Apr 2007 15:02:49 GMT

I would like an explanation to what SMART ERROR EJECT ENABLED or (DISABLED) means? I have not seen this setting before. Here is the current setup for the controller:

Controller:
HSJ80 ZG04500045 Software V85J-10, Hardware K01
Configured for dual-redundancy with ZG01605702
In dual-redundant configuration
Configured for single-cluster
Device Port SCSI address 7
Time: 16-APR-2007 16:00:14
Smart Error Eject Enabled
Host PORT_1:
Valid node name: HSJ3RA, valid CI node 3, 16 max nodes
System ID 42001003FFE4
Path A is ON
Path B is ON
MSCP allocation class 1
TMSCP allocation class 0
MAXIMUM_HOSTS = 4
CI_4K_PACKET_CAPABILITY
REMOTE_DISK_COPY_DATA
Host PORT_2:
Valid node name: HSJ3RB, valid CI node 3, 16 max nodes
System ID 42001003F9E4
Path A is ON
Path B is ON
MSCP allocation class 1
TMSCP allocation class 0
MAXIMUM_HOSTS = 4
CI_4K_PACKET_CAPABILITY
REMOTE_DISK_COPY_DATA
Cache:
256 megabyte write cache, version 0022
Cache is GOOD
Unflushed data in cache
CACHE_FLUSH_TIMER = DEFAULT (10 seconds)
Mirrored Cache:
256 megabyte write cache, version 0022
Cache is GOOD
Unflushed data in cache
Battery:
NOUPS
FULLY CHARGED
Expires: 27-MAR-2009

Any answers would be greatly appriciated. Thank you.

Re: HSJ80 Controller Question

Hoff — Mon, 16 Apr 2007 23:28:15 GMT

"SMART Error Drive Eject â In prior versions of ACS, SMART error notifications received from a device were treated as soft (recovered) errors. The notification was passed to the host and operations continued. ACS now provides a CLI switch (smart error eject state) that offers an alternative to this behavior.

With the new switch enabled, drives in a normalized and redundant set that report a SMART error are removed from that set. SMART errors reported by drives in a non-redundant or non-normal set continue to be treated as recovered errors. If the switch is disabled, all SMART errors are treated as recovered errors."

"SMART â Self-Monitoring Analysis and Reporting Technology"

I've posted up some notes around SMART resulting from on some recent papers... HLRL: Disk Storage and Failure Rates, at http://64.223.189.234/node/93

Stephen Hoffman
HoffmanLabs LLC

Re: HSJ80 Controller Question

Wim Van den Wyngaert — Tue, 17 Apr 2007 01:38:02 GMT

http://en.wikipedia.org/wiki/Self-Monitoring%2C_Analysis%2C_and_Reporting_Technology

Wim