Well, in fact this is not exactly the case :)
Download zpaqfranz for free. Zpaq compatible archiver for Win, Linux, Free/OpenBSD, Solaris, MacOS. Advanced archiver (zpaq fork) with deduplication and versioning. Never prune your backups: keeps a forever-to-ever copy (even thousands of versions) of your files, conceptually similar to Mac time...
sourceforge.net
In the "files" (for version 56.4) there is a pre-compiled FreeBSD binary (otherwise on TrueNAS it is mandatory to install clang++ or g++, a little pain in the ass)
Go ahead and post away. There are a lot of tools out there and if your tool is worth while, then someone will say it. If it's not then you will likely hear that too. The fact that you put so many warnings that it's not snake oil has me a bit concerned so I would caution anyone who may use whatever you have, do it at their own risk, at least until it's proven to be safe software. I'm not sure if this topic belongs here so I'll wait to see what you post.
I learnt the hard way that if you propose something for free to someone (on an opensource forum) you always risk finding users who think you want to sell something
Sad but true
"For free" I mean opensource
In other cases everything is much simpler, indeed you get help and they insert the software into the OpenBSD codebase without drama
Anyway maybe some videos is better than 1.000 words
A bit of more details: we want to update a backup of a folder (actually a dataset, tank/d) frequently.
For example, every hour.
The "folder" is not very big, ~130.000 files for ~70GB (a bunch of .DOC, XLS, PDF etc)
Rsync to a local folder (~22 s)
http://www.francocorbelli.it/zpaqfranz/rsync_backup.avi
The filesystem scan is fast, but still takes tens of seconds
Rsync to NAS (~130 s)
http://www.francocorbelli.it/zpaqfranz/rsync_nas.avi
Scanning for syncing to the NAS is slow (high latency), while transferring changed data is (rather) fast
Backup update from file system to local archive (~ 85 s)
http://www.francocorbelli.it/zpaqfranz/zpaq_backup.avi
Scanning the filesystem takes time and also there is the deduplication stage. Once the data has been calculated, updating the backup is quick, but the preparation phase takes time
Backup upgrade via zfs (~2 s)
http://www.francocorbelli.it/zpaqfranz/zfs_backup.avi
zfs does NOT scan the filesystem, and sends only the data "really needed" to update the backup. So 99% of the "heavy" work is done by zfs
SUGGESTION: first try on some kind of VM without relevant data, just in case
Maybe it's better if I explain again
It is opensource C++ software, so
not some kind of Python program or script
Something - broadly speaking - similar to rar or 7z
Since there is no easy way to install the clang++ compiler on TrueNAS, it is not straightforward to compile from source.
I then put a compiled binary (
and this is where the element of caution is) on sourceforge
So my advice is
- If you have FreeBSD (or Windows, or Linux, or OpenBSD, or Mac, or Solaris), download the source and compile yourself (so 100% sure). It is done in a way that doesn't require makefiles or complex things
- With TrueNAS install the compiler [actually I don't know how to do it, I would say that you have to change the FreeBSD configuration files with the related repositories, but I'm sure it is "doable"], download the source and compile it (100% sure)
- If you have a virtual/test TrueNAS (no important data) you can also try the binary from sourceforge, if you think you can trust :)
I hope this is clearer: a binary
objectively poses a security risk, whoever prepared it
Clearly those who have been following me on technical forums for years do not have these doubts, but it is good to be careful
