Do you get actual job failures, or are you only seeing these errors in the logs? What kind of backup are you doing (policy type and options, schedule type)? The client OS appears to be Windows. What version of NetBackup?
In Windows VSS snapshot backups, the sequence of client processes is as follows:
bpfis create ... // Takes a snapshot, creates a "state file" (NetBackup\online_util\fi_cntl\bpfis.fim.xxxx) and possibly some associated metadata files in the same folder. This process persists until the end.
One or more bpbkar32 processes to backup up from the snapshot to storage unit. (None for "instant recovery" where the snapshot is the backup.)
bpfis delete ... // Reads the state file. Transmits the bpbkar status to the bpfis create process. Deletes the snapshot and state files if appropriate.
Without seeing more logs, this looks like the "bpfis delete" is not finding the state file that it needs to know what to do.
I know of two reasons why this might occur. First, if you have instant recovery backups with a maximum number of snapshots, a previous snapshot may have to be released in order to proceed. The state file information from that prior snapshot may be missing. Second, if a "bpfis create" process fails before creating the state file, the "bpfis delete" process won't find it.
Since the first error in the log is status 71, even before the snapshot is supposedly taken, it seems to me that there is already an issue with the parent job.
What kind of policy are we talking about here? Is it possible that the creation of the snapshot already fails before this child job starts, thus having no actual snapshot for bpfis to read?
