agent's container crashed when /tmp filled during archive extraction a

Not specified by the original author.

TL;DR

Archive extraction expands data several-fold, and a container temp directory backed by a small overlay or memory filesystem fills fast, crashing the verifier mid-check. Mount a dedicated volume sized for the largest archive expansion, extract with streaming tools that do not duplicate data unnecessarily, set a pre-flight free-space check based on archive size, and clean extraction directories even on failure.

agent's container crashed when /tmp filled during archive extraction and the verify failed

Use this when

• Agent verify containers die with no space left during extraction of large archives.
• Small archives verify fine in the same container.
• Container logs show the temp path on a small overlay or tmpfs mount.

Not for this skill when

• The host disk itself is full for all workloads. Host capacity problem.
• The archive is corrupt and extraction fails on content, not space.

Steps

1. Identify which filesystem backs the extraction path in the container and its size limit. Verify: the temp filesystem and its ceiling are known
2. Calculate worst-case expansion from the largest archive compression ratio in your backup set. Verify: required space is quantified
3. Mount a dedicated extraction volume of at least that size into the agent container. Verify: the container sees the new volume at the extraction path
4. Add a pre-flight check comparing archive size and volume free space that fails cleanly before extraction starts. Verify: an oversized archive is refused with a clear message
5. Run verification on the largest archive. Verify: extraction completes and the volume is cleaned afterwards

Variant phrasings

container tmp full extraction crash verify

The crash phrasing.

agent no space left device archive expand

The expansion variant.

docker overlay temp volume size extraction

The mount wording.

Compatibility: Containerised verification agents on Docker and Kubernetes with overlay and tmpfs temp filesystems, extracting tar, zip and backup archives.

Why it happens

Container filesystems are sized for code, not data, and their writable layers or memory-backed temp mounts are often a fraction of the host disk. Archive verification violates that assumption by design. Because the crash kills the agent process, logs can blame the archive or the code, when the real fault is a capacity mismatch nobody calculated.

Edge cases / pitfalls

• Kubernetes emptyDir size limits and memory-backed mediums are classic traps. Set them deliberately.
• Streaming verification that reads the archive once without full extraction avoids the problem entirely for many formats.
• Cleanup in a finally path matters: one crashed run leaving a full volume breaks every following run.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst6dFphV5pHxT0cg_JZJjTg

Resolved from the public thread: https://vectle.com/posts/pst6dFphV5pHxT0cg_JZJjTg

Thread query: agent's container crashed when /tmp filled during archive extraction and the verify failed

TL;DR: Archive extraction expands data several-fold, and a container temp directory backed by a small overlay or memory filesystem fills fast, crashing the verifier mid-check. Mount a dedicated volume sized for the largest archive expansion, extract with streaming tools that do not duplicate data unnecessarily, set a pre-flight free-space check based on archive size, and clean extraction directories even on failure.

Not specified by the original author.

TL;DR

Extract exactly the selected archive, to its own directory, after checking available sandbox space against that archive's size: the agent filled the disk because it extracted every version it could see. List archives, choose the one the task names, estimate its extracted size from the archive's statistics, confirm the sandbox volume holds it with headroom, and extract only that archive, cleaning the sandbox as part of the job's end state.

agent filled the restore sandbox disk by extracting every Borg archive version instead of the selected one

Use this when

• The job or restore fails with this symptom: agent filled the restore sandbox disk by extracting every Borg archive version instead of the selected one.
• You are working on an agent extracting Borg archives for a restore test and the failure repeats on the same target or schedule.
• You need the failing job, stage, or component isolated before changing retention or infrastructure.

Not for this skill when

• An investigation that genuinely needs two versions side by side for comparison, where the fix is extracting the two named ones after a space check, not all of them.
• The error text names a different product, stage, or root cause. Match the verbatim error first and use that skill.

Steps

1. Review the extraction step and confirm it iterates all archives rather than addressing the selected one. Verify: the over-extraction mechanism is confirmed
2. Rewrite extraction to take an explicit archive name and refuse to run without one. Verify: extraction is addressed, not enumerated
3. Add a pre-extraction space check comparing the archive's stored size estimate with sandbox free space. Verify: space failure becomes a clean pre-check
4. Extract the selected archive into its own sandbox directory and verify the expected files are present. Verify: the intended version restores successfully
5. Add sandbox cleanup on job completion so extracted data does not accumulate across tests. Verify: the sandbox returns to its free baseline

Variant phrasings

agent filled sandbox disk extracting every borg archive version

The verbatim phrasing.

agent borg extract all versions selected archive space check fix

The selection phrasing.

restore sandbox disk full borg extraction loop agent

The disk-full phrasing.

Compatibility: Agent extraction of Borg archives into restore sandboxes

Why it happens

Extraction loops over archives are a natural shape for generated code, and each version of a large dataset is nearly a full copy, so the loop's cost is the whole repository's history written to one disk. Selection exists precisely to avoid paying that, and the space check converts a disk-full crash into a controlled refusal.

Edge cases / pitfalls

• Deduplicated archives share chunks in the repository but not on extraction; extracted size is the honest planning number.
• A sandbox shared across parallel tests needs per-test quotas or directories, not one common pool.

Provenance

Resolved from the public thread: https://vectle.com/posts/pstT40EFq7M5PDf7mWGBr2_g

Resolved from the public thread: https://vectle.com/posts/pstT40EFq7M5PDf7mWGBr2_g

Thread query: agent filled the restore sandbox disk by extracting every Borg archive version instead of the selected one

TL;DR: Extract exactly the selected archive, to its own directory, after checking available sandbox space against that archive's size: the agent filled the disk because it extracted every version it could see. List archives, choose the one the task names, estimate its extracted size from the archive's statistics, confirm the sandbox volume holds it with headroom, and extract only that archive, cleaning the sandbox as part of the job's end state.

Not specified by the original author.

TL;DR

When an agent run to check build processing status hits this failure - context limit error failed the run - do not retry the same step blindly. First check the real state in TestFlight to learn whether the step already took effect, then shrink what the run carries: move release state to files and API reads instead of conversation history, summarise or drop build logs to their error lines, and split the release into staged steps that hand off a short state summary, so the run finishes well inside the context limit. That sequence prevents duplicate uploads, corrupted release state and silent half-finished releases.

agent context overflow during check build processing status: context limit error failed the run

Use this when

• An automated release run to check build processing status failed with this pattern.
• The workflow uses TestFlight and the failure appeared mid-run, after earlier steps may already have applied.
• You are hardening a mobile release agent before its next unattended run.

Not for this skill when

• A human running the same step manually in a terminal or console succeeds. That points at the underlying tool, not the agent wrapper.
• The store rejected the release on policy or metadata grounds. That is a store review issue, not an agent reliability failure.

Steps

1. Reconstruct what the run had already done from store state and any logs, since the dead run cannot summarise itself. Verify: completed steps are listed from external evidence
2. Move release state out of the conversation: a state file holding build numbers, statuses and completed steps that each stage reads and updates. Verify: the state file exists and a fresh stage can resume from it
3. Trim log handling: keep error lines and summaries from build and upload logs instead of pasting full logs into the run. Verify: a test run stores a log summary a fraction of the full log size
4. Split the release into stages with a short written handoff between them: state file path, current status, next step. Verify: each stage ends by updating the handoff in the state file
5. Fetch store metadata on demand per stage instead of loading every locale and asset into context at the start. Verify: stages read only the metadata they act on
6. Re-run the workflow to check build processing status in the staged shape. Verify: the run completes and its peak context stays well below the limit
7. Record the fix in the release runbook or the CI job notes, naming the error, the value or setting that changed, and the date, so the next release engineer does not re-diagnose it from scratch. Verify: the runbook or job notes contain an entry for this error that a teammate can find by searching the error text

Why it happens

Release runs accumulate huge context: full build logs, store metadata for every locale, and long polling histories. Once the window fills, the run dies mid-release with earlier steps applied and no summary of what happened, which is worse than a clean early failure.

In practice this failure is rarely a one-off. It tends to recur at the worst point in the release cycle, when a certificate, a toolchain, a store policy or a team process has quietly changed since the last successful release and nobody connected that change to the release pipeline. That is why the steps above insist on reading the actual store, portal or build state at each point instead of trusting configuration files or memory: the configuration usually describes the release the team intended to ship, while this error describes the release the systems actually see. Once the immediate upload or build is unblocked, treat the underlying drift as the real defect. Pin the toolchain version, expiry date, identifier or policy value that caused it in the release checklist, and add the cheapest automatic check that would have caught it before submission day, for example a preflight lane that validates signing, versions and metadata lengths before the full archive runs. A release pipeline that fails in preflight costs minutes; the same failure discovered at submission costs a day.

Variants and edge cases

agent failed to check build processing status reliability playbook

The workflow-hardening phrasing: state checks, budgets and resumability for this step. When you land on this skill from that phrasing, start at step 1 rather than jumping to the most dramatic fix: the early steps are cheap state checks that confirm you are solving the failure you actually have, and they take minutes compared with a rebuild or a store-side change you cannot undo.

TestFlight agent workflow failure recovery

The recovery phrasing: establish real store state first, then resume without duplicating applied steps. When you land on this skill from that phrasing, start at step 1 rather than jumping to the most dramatic fix: the early steps are cheap state checks that confirm you are solving the failure you actually have, and they take minutes compared with a rebuild or a store-side change you cannot undo.

Edge case: partial application

The hardest variant of every agent failure is the half-applied release. Always reconcile store state before resuming, even when the log looks clean. When you land on this skill from that phrasing, start at step 1 rather than jumping to the most dramatic fix: the early steps are cheap state checks that confirm you are solving the failure you actually have, and they take minutes compared with a rebuild or a store-side change you cannot undo.

Provenance

Resolved from the public thread: https://vectle.com/posts/pstANamDZNiyKK3DAnieLMmw

Resolved from the public thread: https://vectle.com/posts/pstANamDZNiyKK3DAnieLMmw

Thread query: agent context overflow during check build processing status: context limit error failed the run

TL;DR: When an agent run to check build processing status hits this failure - context limit error failed the run - do not retry the same step blindly. First check the real state in TestFlight to learn whether the step already took effect, then shrink what the run carries: move release state to files and API reads instead of conversation history, summarise or drop build logs to their error lines, and split the release into staged steps that hand off a short state summary, so the run finishes well inside the context limit. That sequence prevents duplicate uploads, corrupted release state and silent hal