RDF System Management Manual

Managing RDF
HP NonStop RDF System Management Manual—524388-003
5-6
Handling Disk Space Problems
Handling Disk Space Problems
When creating a new image file, the receiver preallocates 16 disk extents. If there is
not enough disk space, the receiver encounters a file system error 43 when it tries to
preallocate these extents. The receiver retries the allocation every 5 seconds and
reports the problem at approximately 60-second intervals. The receiver continues
trying to preallocate the disk space indefinitely.
While the error 43 condition exists, the receiver can only:
•
Provide information for STATUS RDF commands
•
Respond to STOP RDF commands
The error 43 condition persists until enough disk space is available for an image file.
During an error 43 condition, the receiver cannot accept any more images from the
extractor.
If you free enough disk space on the image volume to clear the error 43 condition,
however, processing can resume. You can do this by moving any files that might not
be needed (be sure, however, to restore them before the receiver and updaters need
them). Alternatively, you can accomplish this goal by backing up (with the BACKUP
utility) an unopened image file, then purging that file, and finally restoring it (with the
RESTORE utility) when the first file system error 11 (file not in directory) is reported for
the file by an RDF process on the backup system.
Because the receiver cannot accept any more images from the extractor, an error 43
condition on the backup system causes the extractor to stop progressing through the
audit trail files and RDF to fall behind TMF (TMF, however, continues to generate audit
data).
If an error 43 persists on the disk when you issue the STOP RDF command, the
subsystem shuts down successfully without requiring allocation of extents for the new
file. In this case, however, before you restart the subsystem with the START RDF
command, you must make this space available for at least two files per image trail;
otherwise, the START RDF command aborts.
If the error 43 condition is cleared before it becomes necessary to stop RDF, both the
primary and backup systems continue their normal operations.
Responding to Operational Failures
RDF can recover from any of the following events, as described in detail in the next
few pages:
•
Communications line failure on the primary or backup system
•
Processor failure on the primary or backup system
•
Crash of a TMF audited volume on the primary system