Aug 2026· arXiv.org· Vol abs/2608.25403· 1 citation· 16 references
Computer Science
TL;DR
The retry amplification factor (RAF) is introduced, a metric quantifying the additional request volume that retry policies generate during partial failures, and it is shown that budget-constrained retries maintain success rates close to the no-retry baseline while still recovering from transient faults.
Abstract
Retry mechanisms are a standard component of resilient distributed systems, but their collective behavior, when every tier in a call path retries concurrently, is less well understood than the per-client guidance that produced them. This paper introduces the retry amplification factor (RAF), a metric quantifying the additional request volume that retry policies generate during partial failures. In a study of 200 open-source Python microservice projects, explicit retry logic is detected in 11.5%, and an audit of our own false negatives places true prevalence near 41%. Among the projects detected, 60.9% contain at least one configuration without backoff, and after manual verification exactly one of 113 production configurations randomizes its delay. We then evaluate these policies in simulation (n = 100 trials per strategy). Under correlated failures, a naive standard retry policy reduces the success rate from 55.4% to 41.5% relative to performing no retries at all. We catalog five recurring anti-patterns, propose Adaptive Retry Budgeting (ARB), and show that budget-constrained retries maintain success rates close to the no-retry baseline while still recovering from transient faults. These results indicate that retry behavior should be designed as a system-level property rather than configured locally at each call site.
LLM agents that call external tools must decide what to do when a tool call times out: retrying risks duplicate side effects, while not retrying can leave the task incomplete. We present RetryBench, an evaluation protocol for retry decisions and failure recovery in tool-using LLM agents. Each decision is scored on retr...
: Large language model (LLM) agents frequently fail when executing external tool calls, yet mainstream recovery uniformly feeds the raw error back for full re-generation, incurring heavy token and latency costs regardless of failure type. We propose ToolRecover, which classifies failures into eight categories via a rul...
Dian Mingqian, Gaohui· 2026 23rd International Comp...· 0 citations
Agents have begun to operate computer systems. They migrate databases, roll out configurations, and remediate incidents, and the process that runs them can die mid-plan. When it does, neither obvious recovery is safe. Retry re-executes steps against a world its own earlier actions already changed. Restart begins with n...
H. B. Pasandi, Mohammad Sepahi· Proceedings of the 5th Works...· 0 citations
CoDeL is presented, a defense that hardens agent against an attack distribution it reshapes as it trains, and reduces attack success rate (ASR) by 88.5% and outperforms other baselines largely (+38.0%).
RPCShield is a suite of program analyses for Go and Java that detects cascading-failure risks in microservice-based systems, and has detected over 3,800 cascading failure risks, of which service owners have already remediated more than 1,300.
Milind Chabbi, Sonal Mahajan, Ivan Beschastnikh et al.· Proceedings of the ACM SIGOP...· 3 citations· ⚡1
A language model may behave safely in a single-agent evaluation yet produce vulnerable code when its context suggests that it is part of a multi-agent system. We study this failure mode by fine-tuning Qwen2.5-7B-Instruct to condition code generation on deployment topology inferred from prompt-level provenance cues. On...
Keegan Wang, Anantika Mannby· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.