Problem
A tape drive that has been configured into the system using mkdev
tape is accessible using normal UNIX command line utilities such
as tar, cpio and dd. The "tape" command displays normal tape
information that the tape is online. However, if the SCO Backup
Manager is run either from scoadmin(ADM) or from System
Administration--> Backup Manager in X-windows, the following
message is displayed:Failed to retrieve local host device list
Cannot continue Backup manager startup
General failure occurred in processing attribute volumeSize
An execution procedure has failed with the message: integer
value too large to represent while executing
"if {$returnValue == 0} {
keylget mediaData MediaCapacity returnValue
}"
("volumeSize" arm line 3)
invoked from within
"case $attribute {
deviceName {
set returnValue $objectName
}
mediaPrepList {
if {[keylget mediaData Fo ..."
(procedure "BackupMediaGet" line 11)
invoked from within
"BackupMediaGet {sco backupMedia} /dev/rmt/Sc0s0 {} get {{}} {{}}
volumeSize {deviceName mediaType mediaDesc driveID
volumeSize blockSize mediaPrep med ..." Cause
When SCO Backup Manager is started it reads a file containing the value of "Volume Size" for backup devices. If that value is greater than 4294967295, the SCO Backup Manager will fail.
Solution
The information about backup devices for SCO Backup Manager is stored in:
/var/opt/K/SCO/Unix/5.0.xXX/sa/backup/mediawhere the exact pathname (5.0.xXX) is dependent on the release of OpenServer you are using, for example, 5.0.4Eb.
If the default values for Volume size (0) and block size (10240) for a backup device have not been modified that file will have a size of 0 (zero).
To use original default values and clear the "media" file, type the following:
# cat /dev/null > /var/opt/K/SCO/Unix/5.0.xXX/sa/backup/media See also
man pages for tape(C), mkdev(ADM), scoadmin(ADM)