Hello - first post here, wondering if anyone can shed some light on this problem:
We run a large DCL script which does disk syncing and such, both before and after backups. Therefore we run client initiated backups from OpenVMS V7.3-2. We're using 2 different client versions, both 6.5.1 and 5.0MP5. The problem occurs on both versions.
Essentially, the "parent" process fails to get notification when a backup completes if the backup takes longer than 4 hours. To expand a bit, The initial process which makes a call to the nbu$bpcd process is the process that communicates with the master server. Then a "child" process is spawned which does the actual backup and communicates with the media server. What happens is, no matter how long the backup takes, the child process receives notification from the media server when the backup is complete, and dies successfully (also logs to the nbu log file). But if the backup takes longer than 4 hours, the parent process never receives any notification from the master server that the backup has completed. So it just sits there and never dies. OTOH, if the backup takes less than 4 hours to complete, the parent process always receives notification from the master server, and dies successfully.
Has anyone seen this sort of behavior? Is it some sort of configurable threshold/parameter in the master server config? Any other ideas as to where to look?
Thanks in advance.
We run a large DCL script which does disk syncing and such, both before and after backups. Therefore we run client initiated backups from OpenVMS V7.3-2. We're using 2 different client versions, both 6.5.1 and 5.0MP5. The problem occurs on both versions.
Essentially, the "parent" process fails to get notification when a backup completes if the backup takes longer than 4 hours. To expand a bit, The initial process which makes a call to the nbu$bpcd process is the process that communicates with the master server. Then a "child" process is spawned which does the actual backup and communicates with the media server. What happens is, no matter how long the backup takes, the child process receives notification from the media server when the backup is complete, and dies successfully (also logs to the nbu log file). But if the backup takes longer than 4 hours, the parent process never receives any notification from the master server that the backup has completed. So it just sits there and never dies. OTOH, if the backup takes less than 4 hours to complete, the parent process always receives notification from the master server, and dies successfully.
Has anyone seen this sort of behavior? Is it some sort of configurable threshold/parameter in the master server config? Any other ideas as to where to look?
Thanks in advance.