I use LTO for archives only. With the help of ChatGPT, I am building some shell scripts to prebuild the tars with indexes, checksums etc. I am stoked. I have been bummed by the demise of Tolis, the weirdness of LTFS and IBM ditching support on windows. I did a rocky linux install and everything works out of the gate.
If you use Linux regularly, tar -xzvf will become a common command. I currently use Veeam (free version) for my personal backups. My previous employer used TSM for their robotic tape libraries but I suspect that is probably pricey.
Just looked at Veeam. It looks amazing. So much better than BRU ever was. I am trying to get away from anything proprietary at this point. I am going to migrate a couple hundred TB of archives to LTO-8 eventually and hope to do this once - Tar seem to be the thing that will always be around.
Agreed, any UNIX like OS (and even recent Windows IIRC) supports tar. I don't see it going away anytime soon, at least as long as POSIX compatibility is a thing.
Yah. the posix formatting for paths is amazing. I got super sick of having to adjust my path lengths for LTFS. with posix I will never have an issues coming from NTFS. I am always blown away by how amazing linux is. If I could, I would go 100% linux.
Not necessarily for tape, maybe, but I’ve been using borg backup for many years now. Deduplicated, compressed, encrypted archives, easy remote use without the need for a daemon (works via ssh), archives can be mounted (also remote). It doesn’t get any better, IMHO. Plus, there’s borgbase.com for cheap cloud archive storage.
I don't use tape as a regular backup, I have mirrored servers and cloud for that. I use it only for archiving which is so much simpler than backups - no incremental, no snapshots, just data. I have projects from 10-15 years ago that I have need of from time to time so my concern is always, will it be around in 10-15 year. For me tar looks pretty good - it is the way :-)
“No incremental” is also what I love about deduplicated backup solutions like borg. If you backup the same file to the archive as yesterday, it’s only adding index data, but doesn’t use archive space. Every single backup is a full snapshot, but doesn’t take more space than an incremental backup. Deleting old backups only removes the data that isn’t also included in newer backups.
While it might not be as reliable as tar in this regard, borg is also good at recovering data from corrupted archives. But there’s a trade off there, of course.
Borg definitely looks amazing! I will read more about it. I can see this being very useful for certain projects where I rotate a couple "nearline" backup drives in case of calamity.
Read these "rules" AND introduce
yourself before your first post
Being true to what the artists intended
(opinion / entertainment piece)
Comments on Profile Post by Justin S