>> So you have backup of the failed machine's disk stored somewhere else?
No, I have disc from this machine, looks not damaged (random
files). Only problem that OS does not boot beyond "emergency mode".
I would say your second sentence contradicts your first.
Nope ;). There was 1 500GB disc, with Fedora24 and Postgres 9.5. Then
copied "sector by sector" (and resized partitions, volumes, fs) to
1TB one. This was my "working" disc.
To be clear the 1TB disk is working where and with what OS?
1TB DISC is connected to OrangePi4 (as I have copied disc "sector by
sector" from 500GB to 1TB, some months ago, I can't connect both of them
to one machine -> same UUIDs, LVM does crazy). And then I copy single
files using rsync.
For test, now I have put it (500GB) in another laptop, then upgraded
F24 to F30 (step by step, once I had to do manually "pg_upgrade",
ARAIR this was at PG10->PG11 step).
For future reference you can could have gone from 9.5 -> 11 without
the intermediate upgrades.
That is not true for Fedora OS.... Tested both ways, only "step by step"
looks working (but is time consuming).
Also where were you installing Postgres from the Fedora or Postgres
repos?
From Fedora's, during OS upgrade.
Now I did my crazy move, removed battery and power while "suspend to
RAM". OS (F24! @ 1TB) don't start anymore (beyond "emergency").
I will reserve comment on that move. Still in future it would be
prudent to do a pg_dump and stash the backup somewhere off machine
before you do that sort of experiment.
I use now this upgraded disc (500GB), copying files for daily work
from 1TB one attached to OPi4.
What I understand from the above is you are copying files from the 1TB
disk to the 500GB disc to work with on the 500GB disk/machine. Am I
correct?
Yes.
Will try, if I can achieve PG9.5 run on that "non working" one. If
so, and dumpall is OK, then files on this are mainly intact (except
for NetworkManager)...
In that case you will be using 9.5 pg_dump(all) to do the dump,
Assuming it works it is not guaranteed that it will restore forward to
Postgres 11.
This will require installing development packages on your new machine.
Ok, should be easy :)
But what PREFIX should I use during configure step?
"/usr/lib/postgresql/"? And what about location of config files?
The default for a source compile install is
/usr/local/pgsql(https://www.postgresql.org/docs/11/install-procedure.html).
I would stick with that if for no other reason then helping
identifying your package and source clusters.
At 9.5 "era" there ware single directories (/bin /lib etc) for single
machine, now as I can see, there are "single" but one for each PG
version installed, with dispatcher(s) located in PATH (/usr/(local)/bin.
So I want to replicate this doing my PG upgrade.
(I can see PG10 somewhere in /usr/pgsql/postgresql-10/ - bin, lib,
share, postgresql-9.6 is empty).
Where do you see the above, old machine or new machine?
500GB disc, after system upgrade.
So those where the intermediate updates as you hopped through the
OS/Postgres versions.
Laurent
