TGZ to TB2 Converter

How to Convert TGZ to TB2 Online

1. Upload Your TGZ File: Drag and drop or click "+ Add Files" to select one or more .tgz (gzip-compressed tar) archives. Batch is supported — drop a folder of nightly backups in at once.
2. Pick the TB2 Output: TB2 is selected automatically as the target. .tb2, .tbz2, .tar.bz2, and .tbz are all aliases for the same bzip2-compressed tar container — pick whichever extension your downstream tooling expects.
3. Choose Single Archive or Individual Archives (Optional): Default is Single Archive — every uploaded TGZ is unpacked and repacked into one combined .tb2. Switch to Individual Archives to produce one .tb2 per input TGZ (useful when you want to keep each backup snapshot separate).
4. Convert and Download: Click Convert. The TGZ is decompressed with gzip, the inner tarball is recompressed with bzip2, and the resulting TB2 downloads to your machine. File permissions, symbolic links, and directory structure are preserved end-to-end.

Why Convert TGZ to TB2?

TGZ (.tar.gz) wraps a tarball with the gzip Deflate algorithm — fast to write, fast to read, and the default for most Linux source tarballs and quick backups. TB2 (.tar.bz2) wraps the same tarball with bzip2's Burrows–Wheeler block-sorting algorithm, which trades CPU time for noticeably smaller files. Re-packing a TGZ as TB2 keeps the tar metadata intact while shrinking the payload — useful any time storage or bandwidth costs more than the few extra CPU seconds bzip2 takes.

• Shrink archives before long-term cold storage — In published benchmarks, bzip2 at -9 produced files ~21% smaller than gzip at -6 (16.63% vs 21.25% of original size). On a 10 GB log/source-code dataset that is roughly 460 MB saved before it ever hits S3 Glacier or a tape rotation.
• Distribute source code on bandwidth-tight mirrors — Many GNU and BSD projects (binutils, coreutils, FFmpeg historically) publish .tar.bz2 alongside .tar.gz because the smaller download saves mirror bandwidth at the cost of a slower extraction users do once.
• Email or upload over capped pipes — Gmail caps attachments at 25 MB and most messaging apps at 8–25 MB. A 28 MB TGZ that won't attach can squeeze under the limit as a 22 MB TB2 without changing what's inside.
• Repack legacy backups before archiving — Old nightly TGZ rotations from cron + tar -czf can be re-encoded in bulk to TB2 to reclaim disk on a NAS without touching the file contents, ownership, or mtimes.
• Match a downstream toolchain that expects bzip2 — Some build systems (older Yocto/BitBake recipes, certain Debian source packages, FreeBSD ports trees) ship or expect .tar.bz2. Converting from TGZ avoids editing recipes or makefiles.
• Better corruption containment than gzip — Bzip2 compresses in independent ~900 kB blocks, so a single bit-flip damages only the affected block; gzip's Deflate stream typically loses everything after the corruption point. Worth the trade for archives you'll keep for years.

TGZ vs TB2 — Format Comparison

Benchmark numbers above reflect default compression levels (gzip -6, bzip2 -9) on mixed-text datasets; binary or already-compressed payloads (JPEG, MP4, ZIP) will compress poorly under either algorithm.

When TB2 Wins and When It Doesn't

If you're rebuilding archives from scratch and not constrained to bzip2 specifically, also consider TGZ to tar.xz for higher compression, or TGZ to ZIP and TGZ to 7z for broader Windows compatibility.

Frequently Asked Questions

Are .tb2, .tbz2, .tbz, and .tar.bz2 the same thing?

Yes — all four are interchangeable names for a POSIX tar archive that has been compressed with bzip2. The bytes on disk are identical; only the filename extension changes. Older 8.3-filename systems and some FTP mirrors prefer the shorter .tb2 or .tbz, while modern Linux and macOS tooling typically uses .tar.bz2. tar -xjf and tar --bzip2 will extract any of them regardless of the extension.

How much smaller will my TB2 file actually be than the original TGZ?

For text-dominant content (source code, logs, configs, CSVs), expect roughly 15–25% size reduction at default settings — bzip2 -9 averages about 16.6% of original size versus gzip -6's 21.2% in published benchmarks. For archives that are mostly already-compressed media (photos, videos, MP3s, PDFs), the gain is usually under 2% because neither gzip nor bzip2 can re-compress entropy-coded data.

Why is TB2 slower than TGZ to extract?

Bzip2 uses the Burrows–Wheeler transform, which requires sorting rotations of a large data block (~900 kB by default) during both compression and decompression. That sort dominates runtime. On the same hardware, bzip2 typically decompresses at ~23 MB/s versus gzip's ~95 MB/s — a 4x gap that you'll feel on multi-gigabyte archives. The trade is paying that extraction cost once in exchange for storing 15–25% fewer bytes.

Is bzip2 still actively maintained, or should I be using a newer format?

Bzip2 is still maintained (Micah Snyder has been the upstream maintainer since 2021, after Federico Mena took over the long-dormant project in 2019; version 1.0.8 shipped in July 2019). It is not deprecated and remains widely shipped on every major Linux distro and macOS. That said, modern formats like .tar.xz (LZMA2) and .tar.zst (Zstandard) compress better, decompress faster, or both — many distros (Arch, Fedora packages, Conda) have already moved off bzip2 for new artifacts. Convert to TB2 when you need a bzip2 archive specifically; otherwise xz or zstd are often the better long-term call.

Will Linux, macOS, and Windows all open a TB2 file?

Yes. GNU tar (tar -xjf file.tb2) handles it on Linux and macOS out of the box. macOS Archive Utility opens .tar.bz2 and its aliases by double-click. On Windows, 7-Zip and WinRAR both read bzip2-compressed tar; the built-in Windows 11 archive support added in 2023 also handles .tar.bz2. Older Windows builds without a third-party archiver may need 7-Zip installed first.

Does the conversion preserve file permissions, symlinks, and ownership?

Yes. The conversion decompresses the gzip layer, reads the tar headers, and re-compresses with bzip2 — the tar payload itself is not modified, so POSIX modes, owner/group IDs, symbolic links, hard links, and modification times all survive intact. This matters for source-code releases and for backups you intend to restore on the same system.

Can I convert TB2 back to TGZ later if I change my mind?

Yes — the reverse direction is lossless because both formats contain the same tar payload. Use the reverse tool at TB2 to TGZ or simply bunzip2 file.tb2 | gzip > file.tgz at the command line. No file inside the archive is altered.

Why would I choose Individual Archives instead of Single Archive?

Pick Individual Archives when each input TGZ is conceptually a separate unit you want to keep distinct — for example, nightly backups (db-2025-01-01.tgz, db-2025-01-02.tgz, …) where you'd rather end up with one TB2 per night than one giant combined archive. Single Archive (the default) is better when you're consolidating loose TGZs into a single deliverable, since one larger archive usually compresses slightly better and is easier to ship.

Is there a file size limit on xconvert?

Conversions run server-side, so the cap depends on your current xconvert plan rather than your browser. Free use covers typical archive sizes; very large multi-gigabyte conversions may take several minutes because bzip2 compression is CPU-bound at roughly 7 MB/s. See xconvert's pricing page for current per-file ceilings.