A Year Consumes a Solid-State Drive: Codex Log Bug Slammed as 'Slopware'

marsbitОпубліковано о 2026-07-02Востаннє оновлено о 2026-07-02

Анотація

OpenAI's flagship AI coding tool, Codex, was found to have a critical bug causing its feedback logging system to silently and rapidly wear out users' SSDs. A developer reported that Codex was writing approximately 640 TB of data per year to a local SQLite database (`logs_2.sqlite`) through a constant cycle of inserting and immediately deleting log entries, primarily at the verbose TRACE level. While the database file itself remained around 1 GB, the underlying write-amplification from SQLite's WAL mechanism meant the physical SSD endured the full write load. This was enough to exceed the typical 600 TBW endurance rating of a consumer SSD within a year. The root cause was a hardcoded default logging level (`Level::TRACE`) in the configuration, which overrode any user attempts to reduce logging via environment variables. Analysis showed that over 96% of the logged data—including noisy WebSocket packet dumps and repeated system file events—was useless debug information. The issue, which had at least nine related bug reports in the Codex repository, remained latent because it didn't visibly consume disk space, only silently accumulated write cycles. After the report gained traction on Hacker News, OpenAI merged fixes estimated to reduce writes by about 85%. However, even post-fix, the tool would still write an estimated 96 TB annually. The incident sparked broader criticism of "slopware" in AI-assisted development tools, highlighting a lack of resource budgeting for disk, CPU, ...

A year 'eats' a 1TB SSD?

OpenAI's flagship programming tool, Codex, is burning through your solid-state drive with 640TB of writes per year.

Not long ago, a developer submitted an issue on GitHub. This now-closed GitHub issue, #28224, bears the title:

Codex's SQLite feedback log writes 640TB/year, rapidly depleting SSD lifespan.

According to the reporter's actual measurements, his primary SSD lost 37TB of write endurance after 21 days of continuous operation. At this rate, it's about 640TB per year—enough to wear out a consumer-grade drive with a 600 TBW (Total Bytes Written) rating.

As evidence, he posted two tables.

In evidence 1, the log database always appears to be only 1.2GB, seemingly like nothing happened; yet its auto-incrementing row ID has surged to 5.5 billion, while the actual retained rows are just over 500,000, a difference of ten thousand times.

The key is that SSD wear counts the total amount written, not what remains now: all those 5.5 billion rows were written to disk, and deleting them doesn't undo the writes already incurred. So you only ever see those 500,000 rows when checking the file, but the drive has already endured the write load of 5.5 billion rows.

Evidence 2 reveals the distribution of these 5.5 billion rows: over 90% are debug noise that even the developers themselves wouldn't look back at. Simply copying down each entire WebSocket data packet accounted for half of it.

The culprit is a default configuration of Level::TRACE, treating your drive's write endurance as free scratch paper.

A highly upvoted comment on Hacker News directly defined the nature of this issue:

This is one of the most egregious examples of "slopware."

This netizen also helplessly added:

This is a tragedy. The world needs someone to compete with Anthropic.

What's even more awkward is that this problem wasn't unreported.

Sporadic feedback has existed since April this year, dragging on for over two months. Only after users calculated it themselves, wrote reports, and pushed it to the top of Hacker News did it receive serious attention. Even then, this round only cut about 85% of the log writes.

Some tried to fix it themselves but found they couldn't: the desktop versions of these tools are closed-source.

There was also a classic comment in the thread: How did the review process not catch such an obvious bug? Oh right... @codex, review this.

How exactly did the 640TB get written?

What does 640TB mean?

Mainstream consumer SSDs have a rated write endurance of roughly 150 to 600 TBW, enough for the average user for over a decade or two.

Yet Codex's "record what I did" logging feature can write that much in a year.

The story begins when this user checked his drive usage. His machine, running continuously for 21 days, saw his primary SSD endure 37TB of writes.

At this speed, it's about 640TB per year.

What's more absurd is the write pattern.

Codex maintains a local SQLite database, logs_2.sqlite, specifically for feedback logs. This user monitored it for 15 seconds—36,211 rows were inserted, while the total retained row count remained 681,774 from start to finish, not a single one more.

For every row inserted, one was deleted. The row count stayed constant, but the disk was being rewritten tens of thousands of times.

This mechanism has a nickname: insert-and-prune.

Even more ridiculous is what it records: a bunch of filesystem inotify events.

ld.so.cache was logged 128,764 times, locale.alias 37,982 times, passwd 23,843 times.

The same file, by the same program, logged hundreds of thousands of times repeatedly.

The auto-incrementing ID in the logs had exceeded 5.5 billion, while only about 500,000 rows were retained.

A difference of ten thousand times.

This isn't a bug; it's more like an AI programming tool chanting to its own hard drive.

The file is only 1GB, but writes amount to 640TB

If it writes and deletes simultaneously, how big can logs_2.sqlite remain? About 1GB.

This leads to the most counterintuitive point of the whole affair: SSD lifespan depends on "write amount," not "file size." A 1GB file rewritten 640 times equals 640TB of writes for the drive.

SQLite uses a WAL (Write-Ahead Logging) mechanism. Changes are first written to a -wal file, then checkpointed back to the main database in batches. Codex performs over thirty thousand inserts and deletions every 15 seconds. Each one goes through WAL, index updates, and checkpoints—the same storage area, erased over and over.

An analogy: a 1GB notebook, where you erase and rewrite it 1,750 times a day for a year. The notebook is the same, but the paper is worn through.

This is also why this bug could remain hidden for so long: it doesn't take up space, only burns through lifespan.

Checking available disk space shows nothing unusual; the file size stays quiet. Only by reading the drive's own SMART health counters can you see the write amount silently accumulating.

Root cause: an ignored RUST_LOG line

Why were so many logs recorded?

The answer lies in a single line of configuration in the Codex source code: the SQLite feedback log sink is initialized with Targets::new().with_default(Level::TRACE).

In short, the log is set to TRACE level by default—the highest, most verbose, record-everything level.

Codex's logging framework is Rust's `tracing` ecosystem. The standard practice is to read the RUST_LOG environment variable. Users certainly tried setting RUST_LOG to info, warn, or even turning it off entirely.

No use.

with_default(Level::TRACE) hard-locks the global default to TRACE. RUST_LOG simply doesn't take effect on this path. You think you turned off logging, but it writes regardless.

The most deceptive part of this bug isn't that "you forgot to configure," but that "you configured, and it pretended not to hear."

Even more glaring is a proportion.

Breaking down the retained logs by category, TRACE accounts for 70.7%, about 732.5 MB. Adding the two mirrored telemetry logs from codex_otel (log_only and trace_safe) takes up another 25.3%.

70% of writes are TRACE noise. Combined with mirrored telemetry, 96% is pure nonsense nobody would read.

Only 4% is actually meaningful content.

This isn't the first one; at least the ninth

The reporter checked the Codex repository and found at least 9 issues of this "unbounded log growth" type.

#17320: WAL writes wildly during streaming responses, root cause identical to this one—TRACE ignoring RUST_LOG.

#24275: Desktop version logs_2.sqlite explodes.

#22444: WAL grows infinitely and doesn't free up space.

#26374: Writes 0.75GB per day, no rotation.

#27911: A 4KB goals_1.sqlite gets written at 11MB/s.

#20563: Disk writes wildly even when process is idle.

#27020: Disk at 100% activity on Windows.

The earliest trace leads back to #12969, the very PR that connected the SQLite feedback log sink at the TRACE level.

A 4KB database being written at 11MB/s is enough for a standalone article. Yet, both it and the 640TB issue are symptoms of the same product, the same telemetry system.

This shows that from the beginning, Codex's logging and telemetry systems lacked the concept of a "resource budget."

The entire field is competing over token budgets, context lengths, and model capabilities.

But almost no one asks: who manages the disk, memory, and CPU budget for an Agent that resides on a user's machine, running 24/7?

Fixed, but in a very OpenAI way

Reported on GitHub on June 14th. On June 23rd, the reporter updated: three PRs merged. According to his own Codex feedback, they reduced logs by about 85%, so he closed the issue.

First, about that 85%—it's not 100%, and it's not fully deployed yet.

Of the three fixes, #29432 and #29457 were released with version 0.142.0, cutting out per-WebSocket logging and noisy targets. The third, #29599, stops another type of bridged redundant log and will be in version 0.143.0.

Even with all three in place, the remaining ~15% still amounts to ~96TB of writes per year, merely reducing it from "burning out a drive in a year" to "burning it out in six years."

Some defended it: trace logs are stored by design for debugging, not a bug, and indeed help OpenAI track down edge cases.

But that's precisely the issue: using paying users' SSD endurance as free storage for the vendor's debugging—did the users ever agree to this?

In the programming battlefield, more than just SSDs are being burned through

Interestingly, Codex wasn't the only one singled out.

Comments quickly added: Claude Code also writes debug logs heavily to local storage. Some had to symlink the log directory to a RAM disk (tmpfs) to extend their SSD's life.

Both flagship products suffer from the same type of flaw.

Community discussion soon expanded from one bug to the broader quality issues of AI programming tools.

Some complained these agents keep the GPU maxed out and memory usage at 70GB+. Others simply coined a name for this generation of software: slopware.

The original developer's suggestion was simple: set a limit for the app, don't exceed 3GB. Just this one line took Codex 9 issues and several months to finally consider drawing.

The question is: why would a company that constantly talks about "AGI" stumble on a problem even an intern could spot?

Why could this flaw hide for so long? One comment also hit the nail on the head.

A decade ago, setting logs to TRACE would cause the program to freeze instantly, fixed the same day. Today, CPUs are fast enough, memory large enough, disks robust enough that such flaws are quietly absorbed by hardware performance. The program runs, the interface works, users feel nothing—until the SSD dies prematurely one day.

In recent years, software has been stuffed with AI-generated code. Features pile up, abstraction layers thicken, and resource consumption skyrockets, all barely propped up by hardware makers releasing faster chips every year.

Thus, an absurd cycle emerged: software gets worse, hardware gets more powerful. Users pay for new machines under the illusion that "it doesn't seem slower," when in reality, the new hardware is just barely supporting worse software.

A single small bug can't crush OpenAI, of course. But competition between Codex and Claude Code has already spread from model capabilities to the entry point of developers' workflows.

On this front, making quick changes and responding to developer needs was never a bonus—it's just the entry ticket.

References:

https://github.com/openai/codex/issues/28224

https://news.ycombinator.com/item?id=48626930

This article is from the WeChat public account "New Zhiyuan," author: ASI Revelation

Трендові криптовалюти

Пов'язані питання

QWhat is the core issue reported regarding OpenAI's Codex programming tool in the article?

AThe core issue is that Codex's SQLite feedback logging mechanism is set to TRACE level by default, causing it to perform an excessive number of database insert and delete operations. This results in an estimated 640TB of write operations to a user's solid-state drive (SSD) per year, which can quickly deplete or exceed the drive's total write endurance (TBW), potentially leading to premature drive failure.

QHow does the article describe the 'insert-and-prune' mechanism causing the excessive disk writes?

AThe 'insert-and-prune' mechanism described in the article involves Codex continuously inserting new log entries into a SQLite database (logs_2.sqlite) and immediately deleting old ones to keep the file size constant (around 1GB). However, each insert and delete operation contributes to the total data written to the SSD. Even though the file size remains small, the constant churn of data within the file leads to massive cumulative write volumes on the physical disk over time.

QAccording to the article, why was this logging bug difficult for users to detect initially?

AThe bug was difficult to detect initially because it did not consume significant disk space. The logs_2.sqlite file maintained a relatively constant size (~1GB). The massive write volume was only observable by monitoring the drive's SMART health data (which tracks total bytes written), not by checking typical metrics like available disk space or file size.

QWhat was the community's reaction on platforms like Hacker News, as mentioned in the article?

AThe community reaction on platforms like Hacker News was highly critical. A high-rated comment labeled the issue as one of the most notorious examples of 'slopware' (slop software). There was frustration over the delayed response from OpenAI and criticism that such a fundamental resource management flaw could persist in a flagship product, with some comments calling for more competition in the AI coding assistant space.

QWhat broader software development trend does the article suggest this Codex bug exemplifies?

AThe article suggests this bug exemplifies a broader trend of declining software efficiency, often referred to as 'slopware.' Modern software, increasingly bloated with AI-generated code and complex abstractions, consumes excessive system resources (CPU, memory, disk I/O). This inefficiency is often masked by the rapid performance improvements in hardware, creating a cycle where users upgrade their hardware to compensate for poorly optimized software rather than demanding better code quality.

Пов'язані матеріали

Former Nvidia AI Director Overturns Transformer, Creates 5 Trillion-Context Physical AI That Can Simulate the Entire Universe

Former NVIDIA AI Director Anima Anandkumar and her co-founder Benedikt Jenik have unveiled a groundbreaking "Physical AI" model through their startup, Accelerated Understanding. This model departs entirely from the dominant Transformer architecture, instead utilizing "Neural Operators" to directly understand and simulate physical phenomena in their full 4D spacetime (3D space + time). It achieves a staggering inference-time context window of over 5 trillion tokens—reportedly 5 million times larger than current top LLMs—enabling it to generate complete 4D trajectories of complex physical systems in a single, one-shot inference without subsampling or chunking. The model, pre-trained with 1 trillion parameters and scaled to 35 trillion in experiments, functions as a universal physics simulator capable of handling diverse domains from weather to materials science. It establishes a "simulate → improve → simulate" loop for reality validation. The founders notably turned down a lavish offer from Jeff Bezos's Project Prometheus, which included 35% equity, a $2M salary, and over $20B in committed funding. Industry speculation points to NVIDIA CEO Jensen Huang, who previously championed Anandkumar's work, as a potential key supporter behind the scenes. This development signals a paradigm shift from language/image-centric AI to physics-first intelligence.

marsbit13 хв тому

Former Nvidia AI Director Overturns Transformer, Creates 5 Trillion-Context Physical AI That Can Simulate the Entire Universe

marsbit13 хв тому

NVIDIA Earnings Report Quick Read: Quarterly Revenue on the Verge of Breaking the $100 Billion Mark, Can Still Grow 70% Next Year

NVIDIA Q2 FY2027 Earnings Report: Record Revenue Nears $100 Billion, Projects 70% Growth for Next Year NVIDIA reported exceptional Q2 FY2027 results, with revenue reaching $96.22 billion (up 106% YoY) and non-GAAP net income of $53.95 billion (up 118% YoY). Q3 revenue guidance of $108 billion signals the company's imminent entry into a "quarterly $100 billion revenue" era. The most striking projection came from the CFO, who forecasted approximately 70% revenue growth for FY2028, significantly above Wall Street's 45% expectation, and notably stated this outlook excludes any data center revenue from China. The data center segment remained the core driver, generating $89 billion (up 117% YoY), representing over 90% of total revenue. Growth continues to be primarily fueled by capital expenditures from hyperscale cloud providers like Amazon, Microsoft, Google, and Meta. A key product milestone was the confirmation that the Vera Rubin platform has entered full-scale production and begun shipments. Management stated that every 1 GW of Vera Rubin compute deployed represents a roughly $40 billion revenue opportunity, with the platform expected to contribute about 20% of data center revenue in Q3. CEO Jensen Huang noted its ramp-up is the fastest in company history, providing confidence for the strong future outlook. Management emphasized that AI demand is not slowing but accelerating and broadening into new areas like inference, enterprise AI, and robotics. The primary constraint on growth is now supply, not demand. Huang indicated that without supply limitations, the FY2028 outlook would be "much higher." Bottlenecks include components like HBM, advanced packaging, and data center power. In response, NVIDIA is expanding its role, collaborating with financial firms to mobilize over $500 billion in third-party capital for AI infrastructure and securing key resources like land and power. Crossing the $100 billion quarterly revenue mark represents a new scale for NVIDIA, while the 70% growth projection for next year suggests it may just be the starting point for the next phase.

Odaily星球日报16 хв тому

NVIDIA Earnings Report Quick Read: Quarterly Revenue on the Verge of Breaking the $100 Billion Mark, Can Still Grow 70% Next Year

Odaily星球日报16 хв тому

Inflation Has Not Improved. Will Warsh Support a Rate Hike on Friday?

U.S. inflation remained stubbornly high in July, with the PCE price index holding at a year-on-year increase of 3.7%, unchanged from June and still far above the Federal Reserve's 2% target. The core PCE index also stayed flat at 3.3%. While inflation did not worsen, the fact that it did not improve either has increased market expectations for further interest rate hikes. Futures pricing now indicates a higher probability of a rate increase in September and fully prices in one hike by year-end. The economic backdrop is mixed. Second-quarter GDP growth was revised to 1.5%, but underlying components like consumer spending and business investment were robust. However, inflation-adjusted consumer spending stalled in July, and real incomes have barely grown over the past year, eroding purchasing power. The data provides arguments for both sides of the policy debate. The "wait-and-see" camp points to the lack of acceleration in inflation and upcoming methodological changes that may lower reported figures. The "pro-hike" camp highlights that inflation remains hotter than forecasts, sticky services prices, rising diesel and chip costs, and renewed trade tensions with Canada. All eyes are now on Fed Chair Kevin Warsh's upcoming speech at Jackson Hole for clarity on his policy stance. With inflation persistently above target for over five years and midterm elections approaching where prices are a key issue, the pressure for decisive action is mounting. The speech carries significant two-way risk for markets.

marsbit1 год тому

Inflation Has Not Improved. Will Warsh Support a Rate Hike on Friday?

marsbit1 год тому

Торгівля

Спот

Популярні статті

Як купити T

Ласкаво просимо до HTX.com! Ми зробили покупку Threshold Network Token (T) простою та зручною. Дотримуйтесь нашої покрокової інструкції, щоб розпочати свою криптовалютну подорож.Крок 1: Створіть обліковий запис на HTXВикористовуйте свою електронну пошту або номер телефону, щоб зареєструвати обліковий запис на HTX безплатно. Пройдіть безпроблемну реєстрацію й отримайте доступ до всіх функцій.ЗареєструватисьКрок 2: Перейдіть до розділу Купити крипту і виберіть спосіб оплатиКредитна/дебетова картка: використовуйте вашу картку Visa або Mastercard, щоб миттєво купити Threshold Network Token (T).Баланс: використовуйте кошти з балансу вашого рахунку HTX для безперешкодної торгівлі.Треті особи: ми додали популярні способи оплати, такі як Google Pay та Apple Pay, щоб підвищити зручність.P2P: Торгуйте безпосередньо з іншими користувачами на HTX.Позабіржова торгівля (OTC): ми пропонуємо індивідуальні послуги та конкурентні обмінні курси для трейдерів.Крок 3: Зберігайте свої Threshold Network Token (T)Після придбання Threshold Network Token (T) збережіть його у своєму обліковому записі на HTX. Крім того, ви можете відправити його в інше місце за допомогою блокчейн-переказу або використовувати його для торгівлі іншими криптовалютами.Крок 4: Торгівля Threshold Network Token (T)Легко торгуйте Threshold Network Token (T) на спотовому ринку HTX. Просто увійдіть до свого облікового запису, виберіть торгову пару, укладайте угоди та спостерігайте за ними в режимі реального часу. Ми пропонуємо зручний досвід як для початківців, так і для досвідчених трейдерів.

817 переглядів усьогоОпубліковано 2024.12.10Оновлено 2026.06.02

Як купити T

Обговорення

Ласкаво просимо до спільноти HTX. Тут ви можете бути в курсі останніх подій розвитку платформи та отримати доступ до професійної ринкової інформації. Нижче представлені думки користувачів щодо ціни T (T).

活动图片