About log collection by job ID

A new command line interface and API option of gathering relevant logs by specifying a job ID, then upload the gathered logs is included with NetBackup. With the specified job ID, logs within the job execution time frame are gathered from the primary server, media server, and clients if reachable. Legacy logs and try file logs may include logs outside of job execution time frame as those logs do not honor the time duration filter.

A valid job ID must be present in the Activity monitor. By default, a job ID is removed one week after the job is completed. The nblogadm utility cannot gather the logs of a job ID if bpdbjobs or the Activity monitor cannot retrieve the job details of the specified job ID.

The gathered logs include NetBackup product and NetBackup support utility (nbsu) logs. The log gathering supports one record ID at a time, no concurrent log gathering from multiple record IDs.

To avoid filling up the file system on primary server, media server, and client during log gathering, Veritas recommends that you use the KEEP_LOGS_SIZE_GB option. Veritas recommends that you specify the size of NetBackup logs that are retained before you gather the logs. See the NetBackup Administrator's Guide, Volume I for more information.

To avoid the gathered logs filling up the file system on a primary server, a predefined 10GB free space watermark is used. NetBackup uses this watermark to check and prevent the start of log gathering when the available disk space is less than the watermark. Additionally, the gathering process is stopped during logs gathering when the available space on a primary server falls under the watermark. To reduce the free space watermark to 5GB, set the HIGH_WATERMARK_TRB_LOG_RECORDS = 5 in bp.conf file.

To gather logs with higher verbosity level, manually enable logging and configure the desired logging level as documented in the NetBackup Logging Reference Guide. Then restart the job and start a log gathering task.

The gathered logs are stored on the primary server in the directory shown. All contents in the directory are uploaded after you start a log uploading task. Confirm that only the intended files are present in the directory:

You can gather logs from distributed workloads where multiple clients are involved. You must manually gather logs for every job in the Activity monitor where the specific client or node is displayed in the Client column. You must then consolidate all logs. Examples of distributed workloads include Oracle RAC and MSSQL Availability Groups.

The gathered logs can be uploaded to the Veritas Technical Support organization with the command line interface and the API options. See https://www.veritas.com/support/en_US/article.100038665 for more details.

The password that is provided to upload the logs is stored in the form of credential object in the NetBackup credential management pane. It is removed after logs are uploaded. The name of the credential object may be shown briefly during upload, but not the password itself.