On 12/13/23 08:51, Pocket wrote:
I gave up using raid many years ago and I used the extra drives as
backups.
Wrote a script to rsync /home to the backup drives.
While external HDD enclosures can work, my favorite is mobile racks:
https://www.startech.com/en-us/hdd/drw150satbk
https://www.startech.com/en-us/hdd/hsb220sat25b
https://www.startech.com/en-us/hdd/s25slotr
On 12/13/23 10:42, Pocket wrote:
Many reasons........
No real benefit (companies excepted), and issues like you have been
posting.
I went many years without failing data HDD's, then several data drives
started dying over the course of several months. The disks were in
software RAID. I told the RAID to drop the failing disk, operated in a
degraded condition until I was ready to do the work, shutdown when I
wanted, removed the bad disk, installed a replacement disk, booted, told
the RAID to add the replacement disk, and watched the disk resilver. I
suffered zero unplanned down time. I suffered zero data damage or loss.
I was lucky that only one disk failed at a time. RAID was a huge
benefit to me.
People can have issues with RAID, just like anything else. With the
exception of the failed HDD's above, the root cause of my RAID issues
was PEBKAC. The solution was, and remains to be, learning. Thankfully,
there is Michael W. Lucas:
https://mwl.io/nonfiction/os
If the RAID controller bites the bullet you are usually toast unless
you have another RAID controller (same manufacturer and type) as a
spare.
I have zero luck replacing one companies raid controller with another
and ditto on raid built into the motherboard.
I agree that hardware RAID solutions require identical hardware spares.
That is a price you must pay if you care about the data supervised by
that controller.
I really don't need any help losing my data/files as I do a good job
of that all by myself ;)
RAID is not designed to protect against user filesystem manipulation
errors. Backups are.
zfs-auto-snapsnot(8) makes snapshots a no-brainer and recovery self-serve.
I found it is better to just have my data on several backup disks,
that way if one fails I get another disk and copy all the data to
the newly purchased disk.
How many backups do you keep on each of your several backup disks?
Do you use the rsync(1) option "--link-dest=DIR" to do file-level
deduplication?
ZFS with block-level compression and deduplication is a no-brainer:
https://lists.debian.org/debian-user/2023/03/msg00116.html
Another benefit of ZFS snapshots is that they are are atomic. (Yet
another is that they are taken quickly.) So long as your app or service
has its files in a consistent state (ideally, closed), restoring from
the snapshot should produce a set of files that work correctly.
After removing raid, I completely redesigned my network to be more
inline with the howtos and other information.
Please elaborate regarding "redesigned my network" with respect to not
needing RAID.
I have little to nothing on the client system I use daily,
everything is on networks systems and they have certain things they
do.
Please elaborate regarding "the certain things they do" with respect to
not needing RAID.
I have a "git" server that has all my setup/custom/building scripts
and all my programming and solidworks projects.
I assume your git(1) server has a repository and it is on a single disk
with rsync(1) backups. If the repository disk crashes, you replace it
with another disk, and you restore from backup, what happens to clients
who checked out projects after the backup? To clients that checked in
changed projects after the backup? Is recovery less work that replacing
a bad disk in RAID?
I have DELPHI build apps going back to about 1995.
Do you mean:
https://www.embarcadero.com/products/delphi
It all backed up to a backup server(master and slave) and also a 4TB
offline external hard drive. I have not "lost" any information
since.
Please elaborate "master and slave" with respect to not needing RAID.
I also found that DHCP and NetworkManager is your friend.
Please elaborate "DHCP and Network Manager" with respect to not needing
RAID.
David