Votos - 0, Puntuación media: 0
(
)
)
|
Otros manuales para este modelo:
Resumen del manual
Rolling Reload of NonStop Kernel Processors
2/18/04
(3) reload each processor with a full recovery by each process, and (4) perform system
and application recovery after all processors are reloaded. The following text describes
how to determine the amount of time needed for each of these steps.
(1) Determine how much time is required for backup processes to fully takeover.
The time needed for backup processes to fully takeover after a halt varies
depending on processor performance, system load, processor loading, and the
subsystems configured for a specific processor. Use an amount of time based on
your operational experience, or, if no predetermined value is available, expect this
time to be on the order of 2 minutes for a lightly loaded system.
(2) Determine how much time is required to apply the change. Consult the
applicable documentation.
(3) Determine how much time is required to complete a processor reload. The
time needed to fully reload a processor after a halt varies depending on processor
performance, system load, processor loading, and the subsystems configured for a
specific processor. Reload time is normally significantly longer than the time
needed for the command-line RELOAD program “finished” response. Use an
amount of time based on your operational experience, or, if no predetermined
value is available, expect this time to be on the order of 3 minutes for a lightly
loaded system. This time should include the time needed for each process to fully
recover.
Determine how much time is required to rebalance the system. System rebalancing
time depends primarily on the application and system configuration. Additional work
might be required for systems involved in network transactions or RDF configurations,
for example, allowing the RDF processing to catch up sending transactions to a backup
system. You may also want to review the COMM access lists in order to anticipate any
network routing issues. Accurately determining time durations might require some testing
on a development system. For the best results, verify the amount of time for each activity
and in total by rehearsing the rolling reload on a development system.
2.6 Create OBEY command files
In many cases, you can use OBEY command files to help verify that it is safe to continue
with the next rolling reload step. After reading Section 3, Rolling reload preparation,
determine if a set of OBEY command files can simplify and reduce typing errors during
your rolling reload. If so, prepare the OBEY command files, testing them if possible on a
development system. You cannot automate the entire procedure because some of the
steps require manual intervention. For example, you must check the HP NonStop Open
System Management (OSM) or Compaq TSM Service Application display to determine
that there are no unexpected events. The OBEY command files must allow for the
manual checking required, so they should not simply perform a set of operations, delay
for some minutes, and repeat for the next processor. In particular, a single OBEY file
should not contain both a halt and a reload since manual checking is required before the
reload.
Hewlett-Packard Development Company L.P.
3 of 7
CSSI Website
...