Boston Linux & UNIX was originally founded in 1994 as part of The Boston Computer Society. We meet on the third Wednesday of each month, online, via Jitsi Meet.

BLU Discuss list archive


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Discuss] Cloud-backup solutions for Linux?



Most of the comments in response to my original post have addressed only a
subset of backup requirements: sync current files to offsite storage.

Some of the requirements I have include the following:

* As each file is updated, I want a new version saved, and I want history
available for each file saved in perpetuity or at least for as long as
practical (so if say an old financial record gets scrambled/overwritten and I
don't notice until tax season 2019, or a folder with 30GB of family photos
goes missing when I visit family next Christmas, I can go back and get it).

* The backup process should be continuous, set-and-forget, with no manual steps.

* Capacity and cost constraints shouldn't be a significant factor: as my
filesystems grow over the years/decades I don't want to be forced to make
substantive changes to the solution.

* The solution should perform strong integrity checks, such as periodically
calculating and comparing checksums for each stored file against the
storage-manifest (database) of files.  Ideally it would also provide a
periodic integrity-check of backup against current files (in addition to
checking its internal media).

* Restoration of a particular version of a single file or of a storage volume
should be easy to figure out.

* Bare-metal restoration of a machine should put both system software and
user-data back in place with minimal effort.

* The system should encrypt all media with keys I control.

* Monitoring of the system should inform me within a short time if the backup
process stops or slows beyond a reasonable threshold.

* If I add new files/volumes to my systems, I want to find out right away if
my backups aren't including them in their configuration.

* The backup/restore procedures must not have permissions to discard any data.

* Restore of large amounts of data (beyond about 200GB) should have an offline
mechanism to bypass network-capacity constraints.

* The vendor must be sound and trustworthy enough to survive substantial
economic disruptions.

* Support for my variant of Linux (OpenSuSE) without substantial dependencies
on system packages that require frequent distro updates.

----
Whoever invents a system that satisfies these criteria will have gone *way*
beyond mere file-sync methods.  CrashPlan has a lot of this covered but not
these:

* Bare-metal restore
* Monitoring of files beyond its own configuration
* Integrity-check of live files (and they've had issues with internal media)

Of the alternatives out there, the one that looks most promising in light of
these requirements is SpiderOak.  But no one else here has named that one so I
assume it's an obscure vendor that few of us have heard of.

-rich





BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!



Boston Linux & Unix / webmaster@blu.org