EMC VMAX 100K, Product Manual

Page: 180 / 200

Error reporting to the mainframe host
HYPERMAX OS can detect the following error types to the mainframe host in the
VMAX storage systems:
l
Data Check — HYPERMAX OS detected an error in the bit pattern read from the
disk. Data checks are due to hardware problems when writing or reading data,
media defects, or random events.
l
System or Program Check — HYPERMAX OS rejected the command. This type of
error is indicated to the processor and is always returned to the requesting
program.
l
Overrun — HYPERMAX OS cannot receive data at the rate it is transmitted from
the host. This error indicates a timing problem. Resubmitting the I/O operation
usually corrects this error.
l
Equipment Check — HYPERMAX OS detected an error in hardware operation.
l
Environmental — HYPERMAX OS internal test detected an environmental error.
Internal environmental tests monitor, check, and report failures of the critical
hardware components. They run at the initial system power-up, upon every
software reset event, and at least once every 24 hours during regular operations.
If an environmental test detects an error condition, it sets a flag to indicate a pending
error and presents a unit check status to the host on the next I/O operation. The test
that detected the error condition is then scheduled to run more frequently. If a device-
level problem is detected, it is reported across all logical paths to the device
experiencing the error. Subsequent failures of that device are not reported until the
failure is fixed.
If a second failure is detected for a device while there is a pending error-reporting
condition in effect, HYPERMAX OS reports the pending error on the next I/O and
then the second error.
Enginuity reports error conditions to the host and to the EMC Customer Support
Center. When reporting to the host, Enginuity presents a unit check status in the
status byte to the channel whenever it detects an error condition such as a data
check, a command reject, an overrun, an equipment check, or an environmental error.
When presented with a unit check status, the host retrieves the sense data from the
VMAX array and, if logging action has been requested, places it in the Error Recording
Data Set (ERDS). The EREP (Environment Recording, Editing, and Printing) program
prints the error information. The sense data identifies the condition that caused the
interruption and indicates the type of error and its origin. The sense data format
depends on the mainframe operating system. For 2105, 2107, or 3990 controller
emulations, the sense data is returned in the SIM format.
SIM severity reporting
HYPERMAX OS supports SIM severity reporting that enables filtering of SIM severity
alerts reported to the multiple virtual storage (MVS) console.
l
All SIM severity alerts are reported by default to the EREP (Environmental Record
Editing and Printing program).
l
ACUTE, SERIOUS, and MODERATE alerts are reported by default to the MVS
console.
The following table lists the default settings for SIM severity reporting.
Mainframe Error Reporting
180 Product Guide VMAX 100K, VMAX 200K, VMAX 400K with HYPERMAX OS