This paper discusses the nature of the damage which can cause unanticipated, early fleet failures in legacy military aircraft. This damage often takes a form quite unlike the notional cracking which is used in structural integrity design, but can lead to high costs, or reduced fleet availability, by requiring additional costly inspections and recovery programs in response to discovery of the ... [Show full abstract] damage. Using examples from Australian fleets the paper demonstrates that a lack of diagnostic and prognostic tools contributed to the impact of the damage on the fleet. Development of even simple prognostic and diagnostic tools could reduce the fleet impact and cost of discovered damage, and would ultimately allow the non-crack damage to be incorporated into design to achieve a much more global level of damage tolerance. The paper discusses the key differences between these non-crack damage forms and the more traditional crack-like defect which is used in current damage tolerance based structural integrity management approaches for these aircraft. These differences are associated principally with damage variability and damage location, and they challenge some aspects of our existing structural integrity design methods such as reliance on testing and analysis of supposedly ‘representative’ example of aircraft structure. The larger challenge is to fully exploit the principles of damage tolerant structural design and management, and the paper argues that to achieve this we need to maintain a move towards a broader, risk-based approach to structural integrity management. This longer-term goal will also involve a reappraisal of the nature and distribution of damage, and a fundamental shift in our crack-centric view of structural integrity, additional diagnostic and prognostic tools for such damage would be essential for developing this transition to a more global risk-based damage management approach.