2017 – 2022 · Montpellier, France · Hybrid
Third-line support (network monitoring applications) — telecoms
Astek — Leading French telecommunications and digital services provider
The service centre carrying this client's six network monitoring applications was deteriorating in 2017: satisfaction falling, the activity running at a loss. The services company moved the centre from Toulouse to Montpellier and set up a recovery team there. Six people on the technical side — two on third-line support for the IP network monitoring application, a third on the other applications, two developers and a business analyst — split across two project managers, one on run, the other on build. Version development stayed in Toulouse, for the Montpellier team to take over.
The handover came with no organised knowledge transfer: project managers were thin on availability, outgoing teams were held to their own fixed-price targets, and what remained was a few documents, application access and autonomy. More than sixty escalated tickets were waiting, some blocked for months — reference data wrongly integrated into SNMP collection, SNMP table changes on the equipment, network incidents, SQL views refreshed at the wrong moment relative to the queries consuming them, performance problems. The initial mandate: take the quick wins and send back to their originators the tickets that had no business being there.
Third line receives unresolved tickets escalated from first and second line, along with production faults arising from changes. Beyond that, keeping the applications running: daily import checks, reporting, alert handling, and directing the operator in production. Service hours from 8 a.m. to 6 p.m., no on-call rota, under a response commitment. Knowledge transfer only came afterwards — vendor tool training, then a proper handover on the VoIP monitoring application — the understanding of the solution having first been built through fixing it.
This was where the workings of a fixed-price services company became clear: every action is sold against a unit-of-work grid, some only on a quote approved by the client — a change of frame from an in-house post where optimisation was decided alone. With the current flow absorbed and the stock cleared, the situation returned to stability at the end of 2017. The role continued until 2022, shrinking on the IP application as the development share grew, and widening to the complex problems of the whole scope.
Achievements
- 20 older tickets handled a month, on top of the current flow
Clearing the backlog inherited from the service centre transfer
More than sixty escalated tickets, some blocked for months, taken over with no handover from the outgoing team. With the current flow absorbed and the stock cleared at this rate, the situation returned to stability at the end of 2017.
- 6 Netrac-based monitoring applications
Keeping the application estate running
Daily import checks, reporting, alert handling and directing the operator in production, on service hours from 8 a.m. to 6 p.m. under a response commitment.
Diagnosing SNMP collection faults
Reference data wrongly integrated, SNMP table changes on the equipment, network incidents, SQL views refreshed at the wrong moment relative to the queries consuming them, performance problems: the causes of the backlog were spread across the network, the database and the application.
Corrective work beyond the initial scope
Called on from the first weeks beyond the application in charge: SQL scripts on the VoIP monitoring application, fixing a Perl script on another, taking over a failing Talend import.
Skills applied
- Application maintenance and operations management
- Application support and production support
- Business Objects (BO)
- Fixed-price estimating
- Linux
- Netrac
- One Automation
- Oracle
- Service level agreement (SLA)
- SNMP
- Web services