Mainframe Monitoring Engineer

• Monitoring of GUI Console with all Db2M environments in the control center and monitoring of critical/alert messages to QSYSOPR to any Db2M errors. These alerts would be for catastrophic events like Db2M suspended or node failure etc.
• Backup processes.
• Run processes to execute FlashCopy for offline backup.
• Run / monitor offline backups.
• Expert labs have provided the process for taking backups via a FlashCopy for saving to tape. If an error occurs with this process then skills with PowerHA will be required to deal it with the issue.
• Collecting data via QMGTTOOLS and Job watcher as required by IBM for supporting the environment.
• Monitor job log reporting for error conditions.
• Execute joblog monitor tool and collect data over all jobs.
• Performance Use iDoctor to analyze performance issues that may arise from the activeactive. As time goes on and new major application versions are released it is important to keep Db2 mirror in mind for best practices for performance.
• Staff running performance monitoring will minimally need to know how to look at wait buckets and call stacks well to look for socket receive times that may indicate a higherthanexpected delay on the target side. They also need to be good at identifying which target job or jobs are being used by the source jobs to see the complete picture of a jobs performance. More detailed analysis will require the ability to collect and analyze PEX logical database I/O (*DBIO) traces.
• Additional monitoring support (there are SQL services and CL commands provided by Expert labs to assist with the monitoring of these features):
• All nodes active in clustering
• Cluster resource groups active ( IP takeover)
• errors on the object tracking list
• all network resource groups active ( Db2M over RoCE )

• Object out of sync/in error condition... if it is file how to trace the dependencies on the object and correctly remove all dependencies before fixing the issue. This is most likely to occur on a resync which would prevent the resync from completing.
• Tracing issues between nodes. Db2M uses jobs on both nodes to complete a transaction and knowledge of how to identify dependent jobs is necessary to debug Db2M application issues.
• Resync... how to review a resync and determine the impact if any to running operations how to prioritize objects for resync if necessary
• Performance of resync this is a key process to get back into activeactive experience on how to monitor for errors that could hold this process up and deal with the issue is very important.
• Role swap (for maintenance) A role swap can be a complicated process to ensure the Db2M nodes are swapped then the IPs switched then the node carefully taken offline for PTFs or maintenance
• Role swap unplanned this is a worstcase scenario due to a system loss... how to deal with ensuring everything has switched correctly as well as all aspects of returning Db2 Mirror to operational after the system returns.
• Manage the replication criteria list correctly as new objects are added/removed from the environment. This is very important as altering the RCL incorrectly can result in unnecessary index builds on the target or even Db2 Mirror suspending.
• Reclone process SYSBAS. The reclone of a SYSBAS for CVS will always be a soft reclone to avoid changing any of the system identity or APP in system ( like BRMS Aldon etc. ). The soft reclone will only push the changes to objects in system like user profiles etc. to the other node.
• Reclone process IASP. A reclone of the IASP is 700TB of data and must be performed via storage disk to disk copy. The process is documented by IBM Expert labs and must be done extremely carefully ensuring that each step is follow carefully to ensure a clean copy of data on the target node.
• Change configuration a change in configuration will require changes to the reclone documentation especially the PPRC scripts as well as a slightly different process in the reclone of an IASP.
• Add/remove NRGs change management IP addresses etc. There are many configuration changes that will require changes in the Db2 Mirror/Cluster setup to reflect the changes correctly.

data,cluster,disk,gui testing,sql services,event management,application,flashcopy,swap,cl commands,monitoring performance,idoctor,pex logical database i/o (*dbio) traces,powerha,ip,ibm db2,it,db2 mirror/cluster setup,db2m,sybase adaptive server,ibm,mainframe monitoring