RDF System Management Manual for H-Series RVUs (RDF 1.8)

5 Managing RDF
You manage the RDF environment by monitoring various things using RDFCOM STATUS
commands, the EMS log, and the ASAP product.
In managing RDF, you must sometimes react to nonroutine events and conditions that affect the
RDF operating environment, by performing a variety of special tasks and activities. Although
most of this work is not required on a regular basis, the need for it does arise on occasion.
Typically, this work involves using RDFCOM, TMFCOM (the interactive interface to TMF),
SQLCI (the NonStop SQL/MP interactive interface), TACL (the interactive interface to the NonStop
operating system), or FUP.
This chapter, which is directed to both system managers and operators, discusses the following
topics:
• “Recovering From File System Errors” (page 121)
• “Handling Disk Space Problems” (page 124)
• “Responding to Operational Failures” (page 124)
• “Stopping RDF” (page 129)
• “Restarting RDF” (page 133)
• “Carrying Out a Planned Switchover” (page 133)
• “Takeover Operations” (page 135)
• “Reading the Backup Database” (page 140)
• “Access to Backup Databases in a Consistent State” (page 140)
• “RDF and NonStop SQL/MP DDL Operations” (page 141)
• “RDF and NonStop SQL/MX Operations” (page 143)
• “Backing Up Image Trail Files” (page 143)
• “Making Online Dumps With Updaters Running” (page 144)
• “Doing FUP RELOAD Operations With Updaters Running” (page 144)
• “Exception File Optimization” (page 145)
• “Switching Disks on Updater UPDATEVOLUMES” (page 145)
• “Online Remirroring of Updater SUBVOLUMES” (page 145)
Recovering From File System Errors
Errors detected by the file system generate error numbers that appear within RDF messages
reported in the EMS event log. Of particular importance are the file-system errors that can cause
RDF event messages 700 (modify operation), 705 (open operation), or 739 (create operation). As
an example, file-system error 59 appears in the following RDF event message 705:
10:59 \RDF05 $WU02 705 File open error 59 on $DATA07.QD004378.RFILE02
Tables 5-1, 5-2, and 5-3 list the file-system error numbers and appropriate recovery actions. In
general, when any of these errors occurs, one of the following actions results:
• The event is reported in the EMS event log, with the error causing the process that received
it to restart from a previously set restart point.
• The event is reported in the EMS event log, and the requested operation is skipped.
For information about restart points, see “Updater Processes” (page 52).
To analyze an error, see the appropriate table in this discussion, reading about any corrective
action specific to RDF. Then, for further information about the message (its cause, effect, and
general recovery procedures), see the file-system information in the Guardian Procedure Error and
Messages Manual.
Some errors might require you to resynchronize certain files; see the EMS event log for further
information. Any error that cannot be explained should be reported to your service provider.
Recovering From File System Errors 121