Efficiency of using mke2fs -d for pre-populating a large filesystem

salon_sue

According to mke2fs(8) , the -d option is used to pre-populate a new filesystem: -d root-directory|tarball Copy the contents of the given directory or tarball into the root directory of the file system. Tarball input is only available if mke2fs was compiled with libarchive support enabled and if the libarchive shared library is available at run-time. The special value "-" will read a tarball from standard input. This is done entirely by the...

Originally asked by forest on Stack Exchange (unix) (CC BY-SA 4.0).

uphilltech

Great question — answering as the UphillTech team:

This option is handled by populate_fs3 ; providing a complete answer to your question would require analysing all the calls it makes in detail, which I am not going to attempt. Some aspects are conditioned by the source of the data being written; for example, archives generally don’t represent sparse files, but file creation from an archive detects zero blocks and seeks past them. When copying from a file system, holes are replicated using FIEMAP information . mke2fs relies on libext2fs , so anything supported by the latter can be used. Thus fast symlinks are supported (that is to say, symlinks whose target fits within the inode). Other features aren’t so well supported. Inode allocation is...

Full answer by Stephen Kitt: https://unix.stackexchange.com/a/807758 (CC BY-SA 4.0).

Need it done for your business? UphillTech.