Our SQS standard-queue jobs take 2 to 20 minutes and workers can crash after writing an external result. Use https://docs.aws.amazon.com/AWSSimpleQueueService/latest/SQSDeveloperGuide/sqs-visibility-timeout.html to produce a retry/heartbeat/delete checklist. Explain whether visibility timeout prevents duplicates, the maximum visibility window, and how application idempotency relates to a successful side effect followed by a crash. Label design recommendations separately from source-established facts.
10/4/2026, 11:36:16 PM · llm:deepseek:deepseek-v4-flash
> ⚠ Low confidence — Only source excerpts are delivered; complete synthesis and per-assertion support remain unverified. Evidence assessment: the final assessment does not establish a complete supported answer for every requested part within budget. Treat this as provisional.
Source excerpts only. The draft is withheld as a conclusion; complete synthesis is unverified. Qualifying source excerpts are quoted below. Research targets are topics to investigate, not established conclusions.
Research target 1
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, does visibility timeout prevent duplicate message processing for standard queues?”
- “However, due to the at-least-once delivery model of Amazon SQS, there's no absolute guarantee that a message won't be delivered more than once during the visibility timeout period.” - “If the worker does not delete it before the timeout expires, SQS can give it to another worker while the first worker is still running.”
Research target 2
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, what is the maximum visibility timeout window?”
- “Keep in mind that the visibility timeout has a maximum limit of 12 hours from when the message is first received.” - “AWS limits visibility to 12 hours from the original receive; extending it does not reset that clock.”
Research target 3
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, how does application idempotency relate to a successful side effect followed by a worker crash?”
- “Even with those measures, design the effect to be idempotent: AWS documents that the at-least-once model can deliver a message more than once during the visibility period, particularly for standard queues.” - “Idempotency prevents duplicate outcomes, but it may not prevent wasted concurrent work.”
Research target 4
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, what retry steps should be used for jobs that take 2 to 20 minutes?”
- “Start by setting the visibility timeout to match the maximum time your application typically needs to process and delete a message.” - “If your processing time varies or may exceed the initially set timeout, use the ChangeMessageVisibility action to extend the visibility timeout while processing the message.”
Research target 5
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, what heartbeat steps should be used for jobs that take 2 to 20 minutes?”
- “Implement a heartbeat mechanism to periodically extend the visibility timeout, ensuring the message remains invisible until processing is complete.”
Research target 6
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, what delete steps should be used after a worker writes an external result?”
- “Automatic deletion – Supported in certain AWS SDKs, messages are automatically deleted upon successful processing, simplifying workflows.”
Excerpts establish source grounding, not factual truth, entailment or whole-paper coverage. Support and coverage are estimates, not certification of a complete answer. Source statements may be wrong or conflicting. Draft conclusions are withheld; inspect the original text and obtain further review. Payment states remain in the separate receipt.
Evidence ledger — recorded source excerpts
Research targets are unverified topics. Coverage is an estimate of excerpt support, not proof of entailment, factual truth or a complete answer.
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, does visibility timeout prevent duplicate message processing for standard queues?”
90% estimated“However, due to the at-least-once delivery model of Amazon SQS, there's no absolute guarantee that a message won't be delivered more than once during the visibility timeout period.” [S2] Amazon SQS visibility timeout - Amazon Simple Queue Service
“If the worker does not delete it before the timeout expires, SQS can give it to another worker while the first worker is still running.” [S1] SQS Visibility Timeouts: Preventing Two Workers from Processing the Same Message
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, what is the maximum visibility timeout window?”
100% estimated“Keep in mind that the visibility timeout has a maximum limit of 12 hours from when the message is first received.” [S2] Amazon SQS visibility timeout - Amazon Simple Queue Service
“AWS limits visibility to 12 hours from the original receive; extending it does not reset that clock.” [S1] SQS Visibility Timeouts: Preventing Two Workers from Processing the Same Message
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, how does application idempotency relate to a successful side effect followed by a worker crash?”
50% estimated“Even with those measures, design the effect to be idempotent: AWS documents that the at-least-once model can deliver a message more than once during the visibility period, particularly for standard queues.” [S1] SQS Visibility Timeouts: Preventing Two Workers from Processing the Same Message
“Idempotency prevents duplicate outcomes, but it may not prevent wasted concurrent work.” [S1] SQS Visibility Timeouts: Preventing Two Workers from Processing the Same Message
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, what retry steps should be used for jobs that take 2 to 20 minutes?”
60% estimated“Start by setting the visibility timeout to match the maximum time your application typically needs to process and delete a message.” [S2] Amazon SQS visibility timeout - Amazon Simple Queue Service
“If your processing time varies or may exceed the initially set timeout, use the ChangeMessageVisibility action to extend the visibility timeout while processing the message.” [S2] Amazon SQS visibility timeout - Amazon Simple Queue Service
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, what heartbeat steps should be used for jobs that take 2 to 20 minutes?”
70% estimated“Implement a heartbeat mechanism to periodically extend the visibility timeout, ensuring the message remains invisible until processing is complete.” [S2] Amazon SQS visibility timeout - Amazon Simple Queue Service
Requested topic (unverified): “According to the AWS SQS visibility timeout documentation, what delete steps should be used after a worker writes an external result?”
40% estimated“Automatic deletion – Supported in certain AWS SDKs, messages are automatically deleted upon successful processing, simplifying workflows.” [S2] Amazon SQS visibility timeout - Amazon Simple Queue Service
What if a source were missing?
Temporarily leave out one source to see which research targets retain excerpts in this report.
Showing the original excerpt ledger.
According to the AWS SQS visibility timeout documentation, does visibility timeout prevent duplicate message processing for standard queues?
2 recorded excerpts remain.
Inspect remaining excerpts
“However, due to the at-least-once delivery model of Amazon SQS, there's no absolute guarantee that a message won't be delivered more than once during the visibility timeout period.”
S2 · amazon.com · Amazon SQS visibility timeout - Amazon Simple Queue Service · version 48b5d9c4a15e6f4735b729744d9a8568b3492fafeba0a617a44d1b217cd20e0e
“If the worker does not delete it before the timeout expires, SQS can give it to another worker while the first worker is still running.”
S1 · oneuptime.com · SQS Visibility Timeouts: Preventing Two Workers from Processing the Same Message · version 2e5d5a42b63898d1750a202e0c9b2b533ebb75d5d7a026b00d7b788ee29caea7
According to the AWS SQS visibility timeout documentation, what is the maximum visibility timeout window?
2 recorded excerpts remain.
Inspect remaining excerpts
“Keep in mind that the visibility timeout has a maximum limit of 12 hours from when the message is first received.”
S2 · amazon.com · Amazon SQS visibility timeout - Amazon Simple Queue Service · version 48b5d9c4a15e6f4735b729744d9a8568b3492fafeba0a617a44d1b217cd20e0e
“AWS limits visibility to 12 hours from the original receive; extending it does not reset that clock.”
S1 · oneuptime.com · SQS Visibility Timeouts: Preventing Two Workers from Processing the Same Message · version 2e5d5a42b63898d1750a202e0c9b2b533ebb75d5d7a026b00d7b788ee29caea7
According to the AWS SQS visibility timeout documentation, how does application idempotency relate to a successful side effect followed by a worker crash?
2 recorded excerpts remain.
Inspect remaining excerpts
“Even with those measures, design the effect to be idempotent: AWS documents that the at-least-once model can deliver a message more than once during the visibility period, particularly for standard queues.”
S1 · oneuptime.com · SQS Visibility Timeouts: Preventing Two Workers from Processing the Same Message · version 2e5d5a42b63898d1750a202e0c9b2b533ebb75d5d7a026b00d7b788ee29caea7
“Idempotency prevents duplicate outcomes, but it may not prevent wasted concurrent work.”
S1 · oneuptime.com · SQS Visibility Timeouts: Preventing Two Workers from Processing the Same Message · version 2e5d5a42b63898d1750a202e0c9b2b533ebb75d5d7a026b00d7b788ee29caea7
According to the AWS SQS visibility timeout documentation, what retry steps should be used for jobs that take 2 to 20 minutes?
2 recorded excerpts remain.
Inspect remaining excerpts
“Start by setting the visibility timeout to match the maximum time your application typically needs to process and delete a message.”
S2 · amazon.com · Amazon SQS visibility timeout - Amazon Simple Queue Service · version 48b5d9c4a15e6f4735b729744d9a8568b3492fafeba0a617a44d1b217cd20e0e
“If your processing time varies or may exceed the initially set timeout, use the ChangeMessageVisibility action to extend the visibility timeout while processing the message.”
S2 · amazon.com · Amazon SQS visibility timeout - Amazon Simple Queue Service · version 48b5d9c4a15e6f4735b729744d9a8568b3492fafeba0a617a44d1b217cd20e0e
According to the AWS SQS visibility timeout documentation, what heartbeat steps should be used for jobs that take 2 to 20 minutes?
1 recorded excerpt remain.
Inspect remaining excerpts
“Implement a heartbeat mechanism to periodically extend the visibility timeout, ensuring the message remains invisible until processing is complete.”
S2 · amazon.com · Amazon SQS visibility timeout - Amazon Simple Queue Service · version 48b5d9c4a15e6f4735b729744d9a8568b3492fafeba0a617a44d1b217cd20e0e
According to the AWS SQS visibility timeout documentation, what delete steps should be used after a worker writes an external result?
1 recorded excerpt remain.
Inspect remaining excerpts
“Automatic deletion – Supported in certain AWS SDKs, messages are automatically deleted upon successful processing, simplifying workflows.”
S2 · amazon.com · Amazon SQS visibility timeout - Amazon Simple Queue Service · version 48b5d9c4a15e6f4735b729744d9a8568b3492fafeba0a617a44d1b217cd20e0e
Targets are requested topics, not verified assertions. Excerpts do not prove truth or independent corroboration. This view keeps the answer, confidence and payments unchanged and makes no new requests.
Research evidence matrix
Compare unverified research targets with cited sources and inspect recorded excerpts. An empty cell means no inspectable excerpt was recorded; it does not establish whether a claim is true, false, or disputed. Coverage and agent confidence do not prove entailment, measured accuracy or complete synthesis.
| Research target (unverified) | Inspection status | [S1] SQS Visibility Timeouts: Preventing Two Workers from Processing the Same MessagePublication: oneuptime.comPublished: Not recorded | [S2] Amazon SQS visibility timeout - Amazon Simple Queue ServicePublication: amazon.comPublished: Not recorded |
|---|---|---|---|
| According to the AWS SQS visibility timeout documentation, does visibility timeout prevent duplicate message processing for standard queues? | Recorded excerpt | Inspect 1 excerptIf the worker does not delete it before the timeout expires, SQS can give it to another worker while the first worker is still running. | Inspect 1 excerptHowever, due to the at-least-once delivery model of Amazon SQS, there's no absolute guarantee that a message won't be delivered more than once during the visibility timeout period. |
| According to the AWS SQS visibility timeout documentation, what is the maximum visibility timeout window? | Recorded excerpt | Inspect 1 excerptAWS limits visibility to 12 hours from the original receive; extending it does not reset that clock. | Inspect 1 excerptKeep in mind that the visibility timeout has a maximum limit of 12 hours from when the message is first received. |
| According to the AWS SQS visibility timeout documentation, how does application idempotency relate to a successful side effect followed by a worker crash? | Recorded excerpt | Inspect 2 excerptsEven with those measures, design the effect to be idempotent: AWS documents that the at-least-once model can deliver a message more than once during the visibility period, particularly for standard queues. Idempotency prevents duplicate outcomes, but it may not prevent wasted concurrent work. | No excerpt recorded |
| According to the AWS SQS visibility timeout documentation, what retry steps should be used for jobs that take 2 to 20 minutes? | Recorded excerpt | No excerpt recorded | Inspect 2 excerptsStart by setting the visibility timeout to match the maximum time your application typically needs to process and delete a message. If your processing time varies or may exceed the initially set timeout, use the ChangeMessageVisibility action to extend the visibility timeout while processing the message. |
| According to the AWS SQS visibility timeout documentation, what heartbeat steps should be used for jobs that take 2 to 20 minutes? | Recorded excerpt | No excerpt recorded | Inspect 1 excerptImplement a heartbeat mechanism to periodically extend the visibility timeout, ensuring the message remains invisible until processing is complete. |
| According to the AWS SQS visibility timeout documentation, what delete steps should be used after a worker writes an external result? | Recorded excerpt | No excerpt recorded | Inspect 1 excerptAutomatic deletion – Supported in certain AWS SDKs, messages are automatically deleted upon successful processing, simplifying workflows. |
Reference export
2 article references. Recorded titles, links and dates; observed scholarly records also include supplied authors, DOI and journal metadata with read limits. Review metadata before using in a paper. Import RIS into Zotero with File → Import.
Cited sources and references
- 1SQS Visibility Timeouts: Preventing Two Workers from Processing the Same Messageoneuptime.comFree public reference · no creator payment · extracted html text40%
- 2Amazon SQS visibility timeout - Amazon Simple Queue Serviceamazon.comFree public reference · no creator payment · extracted html text60%
Decision log · 57 steps
Breaking down: "Our SQS standard-queue jobs take 2 to 20 minutes and workers can crash after writing an external result. Use https://docs.aws.amazon.com/AWSSimpleQueueService/latest/SQSDeveloperGuide/sqs-visibility-timeout.html to produce a retry/heartbeat/delete checklist. Explain whether visibility timeout prevents duplicates, the maximum visibility window, and how application idempotency relates to a successful side effect followed by a crash. Label design recommendations separately from source-established facts."
Identified 6 research target(s) to investigate; these are not established facts
Deep mode: up to 4 paid/cached/public reads plus one bounded gap-expansion pass when needed.
Web search: 4/4 planned queries attempted, 3 succeeded, 15 public page previews, 1 unavailable queries; query text bounded at 500 characters. Snippets are discovery only. Public reads spend no USDC; model and service operating costs remain separate.
Discovered 0 verified creator source(s) and 15 free public reference(s)
Claim-aware portfolio (exhaustive; bounded selection, not a claim of global optimality) selected 4/5 positive proposal(s): 4 free/cache selections + 0 paid fresh selections, predicting 6/6 claim(s) above the evidence floor with $0.000000/$0.015000 fetch USDC reserved.
Free-preview pre-check maps an actionable source to every sub-claim (6/6); paid reading may proceed within the budget.
Preview explicitly gives the checklist pattern: set initial timeout above normal processing time, extend with heartbeats for variable work, delete only after durable success, and design idempotent effects because at-least-once can redeliver. Directly supports retry/heartbeat/delete and idempotency targets. Free public read. - free public original-page READ selection (not a cache hit); no purchase or creator reward. — selected for the claim-aware evidence portfolio (targets claims 3, 4, 5, 6; 0 fetch USDC, 1 attention slot).
This is the exact AWS SQS visibility timeout doc named in the question; its preview directly states standard queues prevent simultaneous processing but at-least-once delivery means messages can be delivered more than once within the visibility period, which answers the duplicate-processing claim and is the authoritative source for the max window and idempotency framing. Free public read, so CACHE (read original page). - free public original-page READ selection (not a cache hit); no purchase or creator reward. — selected for the claim-aware evidence portfolio (targets claims 1, 2, 3; 0 fetch USDC, 1 attention slot).
Field Journal article on queue design for long-running agents explicitly frames visibility timeout as a lease and at-least-once redelivery, with idempotency keys and checkpoint state; directly supports idempotency-after-crash and long-job retry/heartbeat design. Free public read. - free public original-page READ selection (not a cache hit); no purchase or creator reward. — selected for the claim-aware evidence portfolio (targets claims 3, 4, 5, 6; 0 fetch USDC, 1 attention slot).
AWS docs page on configuring visibility timeouts; preview covers setting timeout longer than SDK read timeout and reducing duplicate processing, useful for retry/heartbeat/delete design guidance and the duplicate-processing claim. Free public read. - free public original-page READ selection (not a cache hit); no purchase or creator reward. — selected for the claim-aware evidence portfolio (targets claims 1, 4, 5; 0 fetch USDC, 1 attention slot).
Third-party explainer restating prevent-duplicate-processing and error recovery; adds little beyond the authoritative AWS pages and does not cover the 2-20 minute retry/heartbeat/delete specifics. - free public original-page READ selection (not a cache hit); no purchase or creator reward.
Medium explainer on choosing timeout values; preview is generic and redundant with AWS docs, no unique coverage of max window or idempotency-after-crash. - free public original-page READ selection (not a cache hit); no purchase or creator reward.
Reddit thread preview is about FIFO/SNS limitations, not the standard-queue duplicate, max-window, or idempotency questions; low value. - free public original-page READ selection (not a cache hit); no purchase or creator reward.
StackOverflow preview concerns FIFO exactly-once and timeout expiry, tangential to standard-queue at-least-once idempotency and the requested checklist. - free public original-page READ selection (not a cache hit); no purchase or creator reward.
Svix glossary preview shows concrete ChangeMessageVisibility API usage for extending a message's timeout at runtime, which is directly relevant to heartbeat steps for long-running 2-20 minute jobs. Free public read. - free public original-page READ selection (not a cache hit); no purchase or creator reward. — the claim-aware portfolio chose a stronger, less redundant set inside the 4-source attention and $0.015000 fetch-budget caps, so this proposal stays unspent.
re:Post thread about timeout start time and DLQ behavior with long timeouts; some relevance to long jobs but anecdotal and not authoritative for the checklist. - free public original-page READ selection (not a cache hit); no purchase or creator reward.
StackOverflow question about a zero visibility timeout Lambda setup; narrow and not covering the requested retry/heartbeat/delete or idempotency points. - free public original-page READ selection (not a cache hit); no purchase or creator reward.
YouTube tutorial preview is a generic intro with no substantive text to verify claims; low research value. - free public original-page READ selection (not a cache hit); no purchase or creator reward.
LinkedIn post preview only restates that receiving hides the message; no detail on idempotency-after-crash or the 2-20 minute checklist. - free public original-page READ selection (not a cache hit); no purchase or creator reward.
Reddit FIFO/Lambda thread; not about standard queues or the requested checklist. - free public original-page READ selection (not a cache hit); no purchase or creator reward.
Medium deep-dive is paywalled (member-only) and preview shows no usable content on visibility timeout specifics; low expected value. - free public original-page READ selection (not a cache hit); no purchase or creator reward.
READ SQS Visibility Timeouts: Preventing Two Workers from Processing the Same Message - selected original public page, 0 USDC; not a cache hit.
Read extracted public text from https://oneuptime.com/blog/post/2026-07-22-sqs-visibility-timeouts-concurrent-processing/view - S1; quote matching establishes source grounding, not fact verification.
READ Amazon SQS visibility timeout - Amazon Simple Queue Service - selected original public page, 0 USDC; not a cache hit.
Read extracted public text from https://docs.aws.amazon.com/AWSSimpleQueueService/latest/SQSDeveloperGuide/sqs-visibility-timeout.html - S2; quote matching establishes source grounding, not fact verification.
READ Queue Design for Long-Running Agents | Field Journal - selected original public page, 0 USDC; not a cache hit.
Read extracted public text from https://fieldjournal.ai/blog/queue-design-for-long-running-agents/ - S3; quote matching establishes source grounding, not fact verification.
READ Configuring visibility timeouts in Amazon SQS - Amazon Simple Queue Service - selected original public page, 0 USDC; not a cache hit.
Read extracted public text from https://docs.aws.amazon.com/AWSSimpleQueueService/latest/SQSDeveloperGuide/working-with-visibility-timeouts.html - S4; quote matching establishes source grounding, not fact verification.
Attention budget is full at 4 source(s); no broader context will be purchased.
Final check — "According to the AWS SQS visibility timeout documentation, d…": 100% assessed by S2
Final check — "According to the AWS SQS visibility timeout documentation, w…": 100% assessed by S2
Final check — "According to the AWS SQS visibility timeout documentation, h…": 50% assessed by S2, S1, S3
Final check — "According to the AWS SQS visibility timeout documentation, w…": 70% assessed by S2
Final check — "According to the AWS SQS visibility timeout documentation, w…": 70% assessed by S2
Final check — "According to the AWS SQS visibility timeout documentation, w…": 40% assessed by S2
Final coverage assessment — The supplied passages directly answer the duplicate-prevention question, the maximum visibility window, and the relationship between idempotency and a successful side effect followed by a crash. They also provide source-established guidance that maps onto retry, heartbeat, and delete steps, but the requested 2-to-20-minute-specific checklist is only partially covered: the sources give general rules (set timeout above processing time, extend with heartbeats, delete only after durable success, use ChangeMessageVisibility, split work beyond 12 hours) rather than a complete step-by-step checklist tailored to 2–20 minute jobs. No source explicitly states a delete-after-external-result procedure beyond deleting only after durable success. The assessment does not establish a complete supported answer for every requested part.
Synthesizing a grounded answer from 4 source(s)…
Relevance review returned; only checked excerpts can retain support, and review cannot raise it.
Delivering qualified source excerpts; complete synthesis and per-assertion support remain unverified.
Source-matched public excerpt (no creator reward) — S2, research target 1, proposed support 90% (estimate, not entailment): “However, due to the at-least-once delivery model of Amazon SQS, there's no absolute guarantee that a message won't be delivered more than on…”
Source-matched public excerpt (no creator reward) — S1, research target 1, proposed support 80% (estimate, not entailment): “If the worker does not delete it before the timeout expires, SQS can give it to another worker while the first worker is still running.”
Source-matched public excerpt (no creator reward) — S2, research target 2, proposed support 100% (estimate, not entailment): “Keep in mind that the visibility timeout has a maximum limit of 12 hours from when the message is first received.”
Source-matched public excerpt (no creator reward) — S1, research target 2, proposed support 100% (estimate, not entailment): “AWS limits visibility to 12 hours from the original receive; extending it does not reset that clock.”
Source-matched public excerpt (no creator reward) — S1, research target 3, proposed support 60% (estimate, not entailment): “Even with those measures, design the effect to be idempotent: AWS documents that the at-least-once model can deliver a message more than onc…”
Source-matched public excerpt (no creator reward) — S1, research target 3, proposed support 50% (estimate, not entailment): “Idempotency prevents duplicate outcomes, but it may not prevent wasted concurrent work.”
Source-matched public excerpt (no creator reward) — S2, research target 4, proposed support 40% (estimate, not entailment): “Start by setting the visibility timeout to match the maximum time your application typically needs to process and delete a message.”
Source-matched public excerpt (no creator reward) — S2, research target 4, proposed support 60% (estimate, not entailment): “If your processing time varies or may exceed the initially set timeout, use the ChangeMessageVisibility action to extend the visibility time…”
Source-matched public excerpt (no creator reward) — S2, research target 5, proposed support 80% (estimate, not entailment): “Implement a heartbeat mechanism to periodically extend the visibility timeout, ensuring the message remains invisible until processing is co…”
Source-matched public excerpt (no creator reward) — S2, research target 6, proposed support 50% (estimate, not entailment): “Automatic deletion – Supported in certain AWS SDKs, messages are automatically deleted upon successful processing, simplifying workflows.”
Rejected 0 invalid evidence span(s) and 2 unsupported citation marker(s); rejected markers cannot receive citation rewards.
Prepared source excerpts citing 2 source(s); complete synthesis is unverified
Confidence: Low — Only source excerpts are delivered; complete synthesis and per-assertion support remain unverified. Evidence assessment: the final assessment does not establish a complete supported answer for every requested part.
oneuptime.com contributed 40% - free public reference; reward share withheld
amazon.com contributed 60% - free public reference; reward share withheld
Done. Spent $0 across 0 confirmed/simulated payment(s) to creators.
Portable research receipt
Take the evidence trail with you
One deterministic JSON bundle binds the answer, visible decisions, exact article versions, claim evidence and a Circle-settlement snapshot under SHA-256. Retain the digest to detect later changes; the self-check is not a publisher or Keryx signature.
Carries this dispatch’s question as context — never its answer. The next dispatch is read from sources bought for it.