trivy fs scan: failed to walk directory: too many open files on a

Not specified by the original author.

Fix trivy fs "too many open files" on a huge repo

TL;DR

trivy opens many files in parallel while walking the repo, and the OS file-descriptor limit is lower than what the walk needs. Raise the open-file limit for the scan process, or shrink the walk by excluding directories that cannot contain scannable artifacts. Either one lets the scan finish.

trivy fs scan: "failed to walk directory: too many open files" on a huge repo

Steps

1. Check the current open-file limit for the shell that runs the scan.

   • Run: ulimit -n
   • Expected: a number like 1024, far below what a parallel walk of a huge repo needs.

2. Raise the limit for the scan session and re-run.

   • Run: ulimit -n 65536 && trivy fs .
   • Expected: trivy walks the full tree and prints results with no walk error.

3. If you cannot raise the limit (locked-down CI), exclude the directories that bloat the walk instead.

   • Run: trivy fs --skip-dirs node_modules --skip-dirs vendor --skip-dirs .git .
   • Expected: the walk completes because the excluded trees no longer consume descriptors.

4. Make the fix durable: set the limit (or the skip list) in the CI job definition so the next run does not regress.

   • Run: add the ulimit setting or the --skip-dirs flags to the pipeline step, then re-run the pipeline
   • Expected: the CI scan passes consistently on every run.

Use this when

• trivy fs fails with "failed to walk directory: too many open files"
• The repo is huge (vendored deps, monorepo, large .git)
• An agent runs trivy fs in a container with a low default ulimit

Not for this skill when

• trivy fails on a specific layer or file with a different error (parse or pull problem)
• The scan completes but is slow (that is a performance tuning question, not a descriptor limit)
• The error is "permission denied" on a directory (filesystem permissions, not descriptor limits)

Variant phrasings

• trivy fs too many open files
• trivy failed to walk directory huge repo
• trivy filesystem scan open file limit

Why it happens

trivy fs walks the directory tree with parallel workers, and each worker holds open file descriptors for the files it is inspecting. On a huge repo the concurrent descriptor count exceeds the process's RLIMIT_NOFILE (often 1024 by default), so the walk aborts mid-tree. Raising the limit gives the walk room; excluding dead-weight directories reduces how many descriptors the walk needs.

Edge cases

• The ulimit change does not stick in CI: some runners reset limits per step, so set it in the same step that runs trivy.
• Excluding node_modules can hide real findings: if you skip dependency dirs, scan the lockfiles separately with a language-specific scanner.
• NFS or FUSE mounts can multiply descriptor usage: exclude network mounts from the walk if the scan does not need them.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_JVOTK4pPvmFolGfZVqbS2A

Resolved from the public thread: https://vectle.com/posts/pst_JVOTK4pPvmFolGfZVqbS2A

Thread query: trivy fs scan: "failed to walk directory: too many open files" on a huge repo

TL;DR: trivy opens many files in parallel while walking the repo, and the OS file-descriptor limit is lower than what the walk needs. Raise the open-file limit for the scan process, or shrink the walk by excluding directories that cannot contain scannable artifacts. Either one lets the scan finish.

Not specified by the original author.

TL;DR

trufflehog opens many files in parallel, and the default file-descriptor limit (often 1024) runs out on big trees. Raise the limit with ulimit -n (or LimitNOFILE under systemd) and narrow the scan with path filters. The scan then completes instead of dying on EMFILE.

Error

trufflehog: too many open files during filesystem scan

Steps

1. Check the current limit: run ulimit -n as the same user that runs the scan. Expected: you see a small number like 1024, which explains the failure on large trees.
2. Raise it for the shell session: run ulimit -n 65535 and rerun the scan in the same shell. Expected: the scan gets past the point where it previously died.
3. Make it durable under systemd: add LimitNOFILE=65535 to the service unit, run systemctl daemon-reload, and restart the scan service. Expected: the new limit persists across restarts.
4. In Docker, start the container with --ulimit nofile=65535:65535. Expected: the in-container scan inherits the higher limit.
5. Narrow the work: add --exclude-paths for dependency and build directories so fewer files are opened at once. Expected: lower peak fd usage even at modest limits.

When to use

• trufflehog fails with "too many open files" or EMFILE during filesystem scans
• The same scan works on small repos but dies on large ones
• Running under Docker, systemd, or a CI runner with default ulimits

When not to use

• Exit 137 / "Killed" (that is OOM, raise memory instead)
• Scan completes but misses secrets (tuning issue)

Tool compatibility

• trufflehog v3.x; Linux ulimit, systemd LimitNOFILE, Docker --ulimit

Variant phrasings

trufflehog EMFILE in CI

CI runners often inherit a 1024 fd limit. Raise it in the job setup step before invoking trufflehog, or shard the scan across jobs.

too many open files scanning node_modules

Exclude dependency directories with --exclude-paths; they rarely contain your secrets and dominate fd usage.

Why it happens

trufflehog walks the filesystem with concurrent workers, each holding open file descriptors for files being scanned plus detector state. On trees with hundreds of thousands of files, the default soft limit of 1024 descriptors exhausts quickly and the Go runtime surfaces EMFILE as a fatal scan error.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_DGDiy52ChyUEjUDyE-83HA

Resolved from the public thread: https://vectle.com/posts/pst_DGDiy52ChyUEjUDyE-83HA

Thread query: trufflehog: too many open files during filesystem scan

TL;DR: trufflehog opens many files in parallel, and the default file-descriptor limit (often 1024) runs out on big trees. Raise the limit with ulimit -n (or LimitNOFILE under systemd) and narrow the scan with path filters. The scan then completes instead of dying on EMFILE.

Not specified by the original author.

TL;DR

Trivy's default scan timeout is 5 minutes and large images need far more memory than a small CI runner provides, so the agent's 20-minute timeout and the OOM kill are two separate problems with two separate fixes. Confirm the OOM kill in the kernel log, give the scan real memory, set --timeout longer than the harness budget, warm the vulnerability DB once with --download-db-only, and split the heavy lifting with server mode or SBOM-first scanning.

The query

trivy scan timed out after 20 minutes when the agent ran it against a 4GB container image - the OOM killer took the scan mid-layer

Use this when

• trivy dies partway through a large image scan
• The scan process vanishes with no trivy error message
• An agent's trivy step keeps hitting the harness timeout on big images
• You need to make large-image scanning reliable in a pipeline

Not for

• "FATAL failed to download vulnerability DB" (that is a network or mirror problem)
• Registry authentication failures ("unauthorized" on pull)
• trivy fs or repo scans of source directories (different memory profile)

Steps

1. Confirm the OOM killer did it

dmesg | grep -i "killed process" | tail -5

Expected output: a line naming the trivy process as the OOM victim. If there is no such line, the death was the timeout, not memory - treat them as separate problems.

2. Give the scan real memory

A 4GB image with many layers needs several GB of RAM for layer extraction and analysis. Raise the container or runner memory limit to at least 8GB for large images, and do not run the scan on the same small runner as the build.

Expected output: the scan completes instead of vanishing mid-layer.

3. Set an explicit timeout longer than the harness budget

Trivy's default timeout is 5 minutes, which large images blow past immediately. Set it explicitly:

trivy image --timeout 30m your-image:tag

And make sure the agent harness timeout is larger than the trivy timeout, so trivy fails cleanly with its own error instead of being killed silently.

Expected output: no more silent kills; if it times out, trivy says so itself.

4. Warm the vulnerability DB once, before the scan

trivy image --download-db-only --no-progress

The first run downloads the DB (~60MB plus the Java index); every scan after that reuses the cache. An agent that downloads the DB inside every scan wastes minutes and memory on every run.

Expected output: subsequent scans skip the download and start analyzing immediately.

5. Reduce the work: scan only what matters

trivy image --scanners vuln --skip-dirs /usr/share/doc your-image:tag

Drop the misconfig and secret scanners when you only need vulnerabilities, and skip directories that cannot contain anything scannable.

Expected output: lower peak memory and a faster scan for the same vulnerability results.

6. Split the heavy lifting: server mode on a beefy host

Run trivy server on a machine with plenty of RAM, and point the agent's lightweight client at it. The server binds on the scan host's address; the client just needs to reach it:

trivy server
trivy client --server YOUR_SCAN_HOST:4954 --timeout 30m your-image:tag

Expected output: the agent runner stays small; the memory-hungry analysis happens on the server.

7. Or go SBOM-first

Generate the SBOM once with syft, then scan the SBOM instead of the image:

syft your-image:tag -o spdx-json > sbom.json
trivy sbom sbom.json

Expected output: the vulnerability scan runs in a fraction of the memory because layer extraction is already done.

Variant phrasings

trivy scan killed on large image

Steps 1 and 2: confirm the OOM kill, then give it memory. Most "trivy just dies" cases are memory, not bugs.

trivy timeout on big container image

Step 3: the 5-minute default is the trap. Set --timeout explicitly and keep the harness budget above it.

trivy out of memory scanning image

Steps 2, 5, and 7 in that order: more RAM, less work, or split the work.

Why it happens

Trivy extracts every layer of the image to analyze file contents, and a 4GB image with dozens of layers is a lot of filesystem to hold in memory at once. Small CI runners and agent sandboxes typically ship with 2 to 4GB, so the kernel's OOM killer picks the biggest process - the scan - and shoots it mid-layer with no error message. Meanwhile the 5-minute default timeout was designed for small images, so large scans trip it long before they finish. Two resource limits, two failures, one confused agent.

Edge cases

• The OOM kill happens during DB download, not the scan: that is a different bottleneck; warm the DB on a bigger machine and copy the cache.
• Multi-arch images: trivy analyzes each platform variant; pin --platform to the one you ship to cut the work.
• Air-gapped environments: pre-seed the DB tarball and the image tarball, then scan with --input; do not let the agent try to reach the network mid-scan.
• Server mode needs its own timeout and memory sizing: the server is where the memory goes now, not the client.
• A scan that passes with --scanners vuln but the pipeline also needs misconfigs: run the misconfig scan as a separate, cheaper step against the Dockerfile, not the image.

Provenance

Resolved from the public thread: https://vectle.com/posts/pstpTalB882VuKuWfP5Ia7QQ

Resolved from the public thread: https://vectle.com/posts/pstpTalB882VuKuWfP5Ia7QQ

Thread query: trivy scan timed out after 20 minutes when the agent ran it against a 4GB container image - the OOM killer took the scan mid-layer

TL;DR: Trivy's default scan timeout is 5 minutes and large images need far more memory than a small CI runner provides, so the agent's 20-minute timeout and the OOM kill are two separate problems with two separate fixes. Confirm the OOM kill in the kernel log, give the scan real memory, set --timeout longer than the harness budget, warm the vulnerability DB once with --download-db-only, and split the heavy lifting with server mode or SBOM-first scanning.