git: RPC failed; HTTP 400 curl 22 / the remote end hung up unexpectedly
The push exceeded a size limit somewhere in the chain: the HTTP post buffer, a reverse proxy body limit, or the server's own cap. Large binary files and long histories are the usual triggers.
Quick fix
Read the commands before running them. Anything that restarts a service, deletes data or changes permissions should be tried on a non-production system first.
# Raise the client-side buffer (500 MB)
git config --global http.postBuffer 524288000
# Push in smaller batches instead of one huge push
git push origin HEAD~500:refs/heads/main
git push origin main
# Use SSH, which is not subject to HTTP body limits
git remote set-url origin [email protected]:org/repo.git
# For large binaries, move them to Git LFSHow to diagnose Git errors
Git errors are frightening mostly because the messages describe internal state rather than what to do. The reassuring fact is that almost nothing is lost: git reflog records every position HEAD has held, so a bad rebase, a hard reset or a deleted branch is nearly always recoverable. Before running any destructive command, note the current commit hash.
If the quick fix above does not resolve it, work through these steps. They apply to this whole class of error, not just to this one message, which is usually what saves the time.
- Run
git statusand read it fully. It tells you the current state and usually names the exact command to move forward or abort. - Use
git reflogto find any commit you think you have lost, thengit reset --hard <hash>orgit branch recovery <hash>to get it back. - For a rejected push, run
git fetchthengit log --oneline HEAD..@{u}to see exactly what is on the remote that you do not have. Prefer--force-with-leaseover--force, which refuses to overwrite work you have not seen. - Abort rather than improvise when a merge or rebase goes wrong:
git merge --abort,git rebase --abort,git cherry-pick --abort. - For repository corruption, run
git fsck --full. If objects are missing, re-cloning and cherry-picking your local work is usually faster and safer than repairing in place.
Tools worth reaching for
git statusgit refloggit fsck --fullgit log --graph --oneline --allgit diff --check
Authoritative references
Primary documentation for this error, worth reading before applying any fix in production.
Related Git errors
- git cherry-pick: conflictThe commit being cherry-picked modifies lines that diverged on the target branch, so Git…
- Git LFS: missing objectsGit LFS pointer files exist but the actual large file content was not downloaded. The LFS…
- git: cannot lock refAnother Git process is holding a ref lock or a stale .lock file remains from a crashed Git…
- Git: CONFLICT during rebaseA merge conflict occurred while replaying commits during a rebase. The same lines were…
- Git: Could not apply stashApplying a stash caused conflicts because the working tree has diverged from when the stash…
- git: detected dubious ownership in repositoryGit refuses to operate on a repository owned by a different user than the one running git…
- git: error: pathspec 'x' did not match any file(s) known to gitGit could not resolve the argument as a file, branch or commit. With checkout it usually…
- Git: failed to push some refs to ... (non-fast-forward)Remote branch has commits that you don't have locally. Someone else pushed changes.
Browse other categories
- HTTP 494xx client errors, 5xx server errors, redirects, headers and protocol problems.
- JavaScript 42npm resolution, async pitfalls, hydration, memory limits and runtime type…
- Database 41Connections, deadlocks, constraints, replication and memory limits.
- AI 35Rate limits, context windows, GPU memory and model-serving failures.
- Network 35Refused connections, timeouts, resets, MTU problems and port exhaustion.
- Python 35Imports, virtual environments, encoding, concurrency and dependency conflicts.
- Kubernetes 34CrashLoopBackOff, ImagePullBackOff, OOMKilled, RBAC, scheduling and storage.
- Docker 27Daemon connectivity, disk space, image pulls, ports and architecture mismatches.
- System 26Disk space, systemd units, file descriptors, OOM killer and scheduled jobs.
- Cloud 25IAM permissions, quotas, service limits and credential failures.
- Security 25JWT validation, CSRF, OAuth grants, SELinux, SSH host keys and CSP.
- TLS 24Untrusted authorities, expiry, hostname mismatch, chains and cipher negotiation.
Something missing or wrong?
This entry is maintained by hand. If the fix is out of date, incomplete, or you have a better one, email a correction and it will be reviewed.