Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Multi part question/problem?
#1
My Linux system is an Dell Optiplex 990 I5 with 8Gb of ram.

I recently replaced the system after a 'periphial failure' post message. It was easier to pick up another 990 and diagnose the failed hardware later just to get back up and running, which I have done.

I added a 250Gb SSD, sda,  and while I regularly run BU, I would like to move the home directory to a conventional hard drive, sdb,  to minimized the writes to the SSD.

My first problem concerns moving my home directory to a second drive. I have found some directions to do so, but that bit is fuzzy right now.

If I can successively move the home directory, that seems to leave a second problem regarding backups. Will BU be smart enough to restore the backup to sdb without intervention if that was it's location on the system that has been backed up?


I do like to use BU when doing an upgrade by blasting the system drive back to ground zero and then installing the new OS before restoring my home directory. I have done this a number of times, but I'm not sure what will happen with the new proposed configuration.

A second reason for moving the home directory off the SSD regards their failure mode, which is sudden and pretty much completely unrecoverable to mere mortals. Having all of what makes the system me on a mechanical drive seem less risky. I have not had a massive number of hard drive failures, but my experience has been that one notices small problems before the whole system heads south...thus giving you a little grace to BU the data in addition to the scheduled backups. A SSD is either their or it's not...  no warning signs.

As an example of 'soft' hard drive failure, one of my Nas4free boxes has a hard time booting from the OS drive. Once Nas4free is running it requires no hard drive for the system at all, it was  minor annoyance 'cause the NAS would and has run for about 1-1/2 years between reboots. That was caused by a UPS failure. I had to smack the HD to get it to boot. Crude but it did work...

Everything has since migrated to a new Nas4Free system that boots from a thumbdrive and now has been running for about 2 years without interruption. BTW, Joe discussed the ZFS file system on one of his latest videos. The BSD OS that runs Nas4Free is using ZFS and mine has been up for 2 years... big learning curve for setup for someone like me that is not terribly hard core, but I did figure it out. It has been so stable that it is does go down I'll have to learn it all over again :>(

Thanks
paul
W9PCS

Resistance Is Not Futile!
It's voltage divided by current



Reply
#2
(11-16-2019, 06:33 PM)Paul Wrote: My first problem concerns moving my home directory to a second drive. I have found some directions to do so, but that bit is fuzzy right now.

See my reply in this thread.


(11-16-2019, 06:33 PM)Paul Wrote: If I can successively move the home directory, that seems to leave a second problem regarding backups. Will BU be smart enough to restore the backup to sdb without intervention if that was it's location on the system that has been backed up?

Funnily enough, software on Linux deals with this not by being smart but by simply not caring. Whether my home directory is on 'sdb1', 'hd0' or 'nvme0n1p1', the path to it is always '/home/leon' (obviously your users account is named differently), so for all software that interacts with files (and not with block devices) there will be no difference (BU uses 'rsync', which copies files, not raw data, so you should be fine).


(11-16-2019, 06:33 PM)Paul Wrote: A second reason for moving the home directory off the SSD regards their failure mode, which is sudden and pretty much completely unrecoverable to mere mortals. Having all of what makes the system me on a mechanical drive seem less risky. I have not had a massive number of hard drive failures, but my experience has been that one notices small problems before the whole system heads south...thus giving you a little grace to BU the data in addition to the scheduled backups. A SSD is either their or it's not...  no warning signs.

There is something called SMART, which lets you see a drives health. That is considerably more reliable than waiting for observable errors, on both HDDs and SSDs.


(11-16-2019, 06:33 PM)Paul Wrote: [...] big learning curve for setup for someone like me that is not terribly hard core, but I did figure it out. It has been so stable that it is does go down I'll have to learn it all over again :>(

Generally it is a good idea to keep an exact 'log' of what you do and what you learn, for exactly this reason.
My website - My git repos

"Things are only impossible until they’re not." - Captain Jean-Luc Picard
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)