NetBackup 7.x Backup Process Flow: Step 1-17 process explained in detail

NetBackup 7.x Backup Process Flow

1. When a PolicyClient task has its timer expire (indicating that it is due to run) an internal job task is created within nbpem that sends a Job Start to nbjm for the job which is due. nbpem provides to nbjm the parameters indicated in the backup policy and schedule that is generating the job.


2. nbjm adds the job to its job list or queue. It then communicates with bpjobd to inform it of the job, at which time the job becomes visible in the Activity Monitor in the queued state until resources are allocated for it.

3. nbjm sends a resource allocation request to the Resource Broker, nbrb, indicating the resources which are required for the backup operation and any resource consumption constraints for the job, including max jobs per policy, max jobs per client, and max jobs this client. These resource consumption constraints were provided to nbjm by nbpem when the job was initiated.


4. nbrb requests resources from the EMM service, nbemm, including storage unit, storage unit group, media, and devices or drives.


5. When physical resources are available, nbemm will allocate them and respond to nbrb, which in turn responds to nbjm. With resources allocated for the job, nbjm will notify bpjobd and the job moves to the active state.


6. nbjm is responsible for creating the files in the Images database that will house the backup information, the Header file and the Files (.f) file. nbjm initiates this activity by communicating with bpdbm (via nbproxy).


7. nbjm uses bpcompatd to communicate via PBX to start bpbrm on the media server that will write the backup image. The media server is selected based upon the destination storage unit that is selected.

8. bpbrm on the media server starts bpbkar (the client's backup and archive process) on the client system. bpbrm also starts bptm on the media server.


9. bptm initiates a connection with nbjm in order to get media and drive information for the backup job, which nbjm returns through a separate connection it initiates.


10. bptm will then initiate the mount of the media (tape) specified on the drive specified, or the mount of the disk specified. It will also spawns a bptm child process to receive the image from the client. The details of the Media Manager daemons (ltid, txxd, txxcd, and avrd) involved in the mounting of the media on the drive are not shown in this illustration, to reduce the complexity of the illustration.


11. bpbkar sends information about the backup image to bpbrm which forwards it to bpdbm on the master server. This stream of metadata is sent throughout the backup and stored in the master server's Image database.


12. When mounting and positioning of the media in the drive, or of the disk to be used, have been accomplished, the client backup process, bpbkar, will begin sending backup data to the bptm child process on the media server system. The bptm child process receives the image and stores it block by block into a shared memory segment on the media server. The parent bptm process retrieves the image from shared memory and directs it block by block to the allocated storage media.

13. When the backup has been completed bptm will notify bpbrm, which in turn will notify the Job Manager nbjm that the job has finished bptm will also notify nbjm that it is done with the media.

14. While the client and media server processes invoked to perform the backup operation (bpbrm, bptm, and bpbkar) are terminated, nbjm will update the status for the job by communicating with bpjobd. The job will be changed to Completed status, and the ending status of the job will be recorded.

15. nbjm communicates with bpdbm (using nbproxy) to complete the writing and verification of the files for the backup image in the Images database.

16. With the backup job completed nbjm will de-allocate the resources used for the backup by communicating with nbrb.

17. nbjm will notify nbpem that the job has been completed The completion status will be included in this notification. The PolicyClient task that created the job is responsible for requesting a re-try operation for this job on failure, or for computing the new 'due time' for this job on success.

Share this

Sandy