beautypg.com

Windows server, Oracle solaris, Windows server oracle solaris – HP 4000.6000.8000 Enterprise Virtual Arrays User Manual

Page 159

background image

Table 57 HP-UX failure scenarios (continued)

Failure effect

Fault stimulus

Short term: Data transfer stops. Possible I/O errors.
Long term: Job hangs, cannot umount disk, fsck failed, disk corrupted,
need mkfs disk.

Server path failure

Short term: Data transfer stops. Possible I/O errors.
Long term: Job hangs, replace cable, I/O continues. Without cable
replacement job must be aborted; disk seems error free.

Storage path failure

Windows Server

Table 58 Windows Server failure scenarios

Failure effect

Fault stimulus

OS runs a command called chkdsk when rebooting. Data lost, data that
finished copying survived.

Server failure (host power-cycled)

Write delay, server hangs until I/O is cancelled or cold reboot.

Switch failure (SAN switch disabled)

Write delay, server hangs or reboots. One controller failed, other
controller and shelves critical, shelves offline. Volume not accessible.
Server cold reboot, data lost. Check disk when rebooting.

Controller failure

Controller momentarily in failed state, server keeps copying. All data
copied, no interruption. Event error warning error detected during paging
operation.

Controller restart

Write delay, volume inaccessible. Host hangs and restarts.

Server path failure

Write delay, volume disappears, server still running. When cables
plugged back in, controller recovers, server finds volume, data loss.

Storage path failure

Oracle Solaris

Table 59 Oracle Solaris failure scenarios

Failure effect

Fault stimulus

Check disk when rebooting. Data loss, data that finished copying survived.

Server failure (host power-cycled)

Short term: Data transfer stops. Possible I/O errors.
Long term: Repeated error messages on console, no access to CDE.
System reboot causes loss of data on disk. Must newfs disk.

Switch failure (SAN switch disabled)

Short term: Data transfer stops. Possible I/O errors.
Long term: Repeated error messages on console, no access to CDE.
System reboot causes loss of data on disk. Must newfs disk.

Controller failure

Short term: Data transfer stops. Possible I/O errors.
Long term: Repeated error messages on console, no access to CDE.
System reboot causes loss of data on disk. Must newfs disk.

Controller restart

Short term: Data transfer stops. Possible I/O errors.
Long term: Repeated error messages on console, no access to CDE.
System reboot causes loss of data on disk. Must newfs disk.

Server path failure

Short term: Job hung, data lost.
Long term: Repeated error messages on console, no access to CDE.
System reboot causes loss of data on disk. Must newfs disk.

Storage path failure

Failure scenarios

159