Making big downloads resumable in Firefox
- ID
- 35149
- Status
- summarized
- Published
- 13 Oct 2026, 3:36 AM
- Fetched
- 14 Oct 2026, 9:20 AM
- Provider
- Hacker News
- Category
- dev-community
- Original URL
- https://lunarwatcher.codeberg.page/til/firefox/resumable-big-downloads.html
- Source URL
- https://hnrss.org/best
Summary
- Score
- 5.5
- Created
- 14 Oct 2026, 9:21 AM
- Tags
- Audience
- developersai_ml_learnersvibe_coders
What happened
A TIL post by Olivia reports that Firefox 28 and newer ship with network.http.response.timeout defaulting to 300 seconds, which appears to break resuming large downloads; setting it to 0 in about:config reportedly makes big downloads resumable and more stable. The author says Firefox seems to ignore existing .part files without this change, notes the issue has been reported since around 2005, and offers wget -c -O "$filename" "$url" as a non-Firefox workaround. The author explicitly says they do not understand why it works and have not confirmed the theory against Firefox source code; the HN thread drew 277 points and 132 comments.
Why it matters
If you regularly pull multi-GB artifacts (model weights, datasets, ISOs) through Firefox and resume keeps restarting from zero, the concrete action here is to try network.http.response.timeout = 0 in about:config, or just use wget -c, which the author found stable. Treat it as an unverified experiment, not a fix: it only helps when the server supports resumable/range downloads, and the author never confirmed the mechanism in Firefox source. If you depend on this for large AI model downloads, verify the resume actually works on your target server before trusting it.
Discussion angle
The author admits they don't know why the timeout change works and never checked the source — is this a real fix, or does it just mask Firefox ignoring .part files? Worth asking who in the room has reproduced it on a multi-GB download, and whether wget -c or a package manager is simply the better default.