SShortSingh.
TopTechnologyAI & MLStartupsProgrammingScienceBusinessFinanceCrypto & Web3World
Back to feed
Programming

Two-Week Daily Script Crash Traced to One Missing Multiplication

0
·1 views

A developer running over 160 automated social media jobs on macOS discovered that an Instagram engagement script was being killed by SIGKILL every morning for two weeks. The root cause was never a code bug — both the action caps (101 total actions) and the random delay settings (20–60 seconds) were individually correct, but their combined runtime of roughly 4,940 seconds far exceeded the 2,400-second execution window. Because SIGKILL bypasses Python's finally blocks, Playwright browser sessions were also left uncleaned after each forced termination, compounding the damage. The oversight highlights a class of configuration bug that cannot be caught by reading code alone — only by computing the product of interdependent settings. The developer, whose automation setup underpins around ¥1.2 million in monthly revenue, noted the fix required a single arithmetic check that had simply never been performed.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

Developer's LLM agent hit 182s p95 latency, forcing a pivot from real-time to background use

A developer building Porchlight, a membership win-back tool, recorded 67 timed interactions with a live LLM agent and found a median response time of 29.2 seconds alongside a 95th-percentile latency of 182 seconds. The stark gap between the two metrics effectively ruled out any real-time, user-facing deployment of the agent. Porchlight is designed to conduct exit interviews with cancelled members, match their stated reasons for leaving against creator announcements, and draft personalised win-back messages. The developer also benchmarked the LLM against a simple keyword-matching baseline, finding the agent resolved 11 departures that the baseline completely missed across 54 test cases. The latency data, gathered before the product design was finalised, became the deciding factor in how the tool could realistically be deployed.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

David Parnas's 1994 Paper 'Software Aging' Still Offers Relevant Lessons for Developers

David Parnas's 1994 paper 'Software Aging' examines why software inevitably degrades over time and how it can either age gracefully or poorly depending on design and maintenance practices. Parnas identifies two primary causes of poor software aging: 'Lack of Movement,' where software fails to evolve with changing needs, and 'Ignorant Surgery,' where modifications made without understanding the original design introduce bugs and complexity. The paper draws humorous parallels between software and human aging to illustrate how unmaintained or poorly updated software becomes obsolete and less desirable. Real-world examples such as the decline of Internet Explorer and the rise — and subsequent bloating — of Google Chrome reflect Parnas's observations about compounding complexity over time. The paper also addresses the financial and technical costs of updating aging software, underscoring the importance of maintainability as a core design principle.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Why API Testing Is Critical to Reliable, High-Quality Software Development

APIs form the invisible backbone of modern applications, handling authentication, data exchange, and payments without any visible interface. When APIs fail, the resulting bugs often surface in confusing ways — such as stalled loaders or silently dropped orders — making them harder to trace. API testing directly validates endpoint behavior, checking status codes, data structure, response times, and error handling without relying on a user interface. Teams benefit from multiple testing layers, including functional, security, load, and contract tests, each run at different stages of development. Shifting testing left — writing API tests as soon as endpoints exist — helps catch issues early, before they compound further down the development pipeline.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

AI assistant startup Instinct raises $350M at $2.5B valuation while still in private beta

Instinct, an AI assistant startup operating as Spear Street Technology and founded by 23-year-old Noah Shinn, has raised $350 million in total funding at a $2.5 billion valuation, with its Series B of $250 million co-led by Index Ventures and Benchmark. The company, roughly one year old, remains in private beta as of August 26, 2026, and has disclosed neither user numbers nor revenue. The product allows users to connect their apps and devices to an AI assistant they can interact with via text and calls to perform tasks like booking tickets, buying groceries, and cancelling subscriptions. Analysts note that the core technical differentiator is not the AI reasoning itself — widely available through frontier models — but the authorised access Instinct holds to users' third-party apps and accounts via OAuth permissions. This deep integration has drawn privacy concerns from users over broad permissions, while investors appear to view the same access as a durable competitive advantage that is difficult to replicate at scale.

0 comments Read more at DEV Community
Two-Week Daily Script Crash Traced to One Missing Multiplication · ShortSingh