On Wed, 2019-10-16 at 17:35 +0000, sixpack13 wrote:
I wouldn't do that. At this time, /var/tmp is expected to be on
permanent storage, some things might expect data to remain available.
Some things might generate more temp files than your RAM (especially if
you burn DVDs or Blurays). If you need to use a temporary file to
debug a problem, you lose it during crashes. Better to take a more
considered way to clean up *old* files.
When they made /tmp use tmpfs, that caused a plethora of such problems.
what are this "plethora of such problems" extactly ? [*]
why *IS* - by this danger's - /tmp per default a tmpfs ?
are you thinking the $Fedorian's are complete brain damaged ?
crying for more bug reports caused by an wrong/dangerous setup ?
I don't seem to suffer the problems I see commonly mentioned on
this
list (filling up hard drives, the wrong kernel booting, continually
having to manually reconfigure grub, etc), the kind of things that are
supposed to take care of themselves, automatically. Probably because I
haven't painted myself into a corner by doing oddball things to my
system in the first place.
I don't seem to suffer the problem too *even* if I do like I do and suggest to others
!
maybe you think I'm complete brain damaged and suggest things to others that I
don't run over month and - for my taste - proof it that way ?
I guess with an ~ 8 GB /var/tmp the OP is regarding DVD burning (4.5 GB) on a save side.
It also seems to me that he is not very customized (wording ?) with debugging and such
things.
so for my taste you brought up a lot of "might's".
My orientation regarding suggestions to others to prevent trouble are on facts.
fact *is* that the OP collected over years (!) ~4.5 GB temp. data reaching back - IIRC -
to 2015 or was it 2013 ?
fact *is* he ran into trouble caused by the lack of space during upgrade caused by the
lack of cleanage over years (!) under / or even better /var/tmp, which isn't a separte
partition.
IIRC, even the cleanage of /var/tmp would have been the successfull run of the upgrade in
the first take.
so *I* was thinking of an "automatism" to do the cleanage for him clearly having
your "might's" in *my* back of the head by running /var/tmp as tmpfs (~8 (!)
GB) since month/years *without* any trouble (so far I might say).
[*]
to be honest, here could be one of the plethora:
https://bugzilla.redhat.com/show_bug.cgi?id=1659178
*but* please don't overread "size=2%" in my mount command there which caused
the bug/RFC.
was "size=2%" a suggestion to the OP.
NO ?
guess why !
Okay ?