From mboxrd@z Thu Jan 1 00:00:00 1970 From: swedebugia Subject: bug#34768: guix-daemon tmpfs out of space on parabola Date: Thu, 07 Mar 2019 15:20:07 +0100 Message-ID: <0E607D15-3AB6-410C-A609-AEB092844EF6@riseup.net> References: <22f759c1-a41c-d3b8-08e7-fe4cda616894@riseup.net> <8736o08494.fsf@nckx> Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="----JJ6QDKAQV0NQ5ODE4KU1U8V1GH1N3N" Content-Transfer-Encoding: 7bit Return-path: Received: from eggs.gnu.org ([209.51.188.92]:59750) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1h2KuZ-0006hu-EV for bug-guix@gnu.org; Fri, 08 Mar 2019 14:12:04 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1h2KuY-0002GD-K6 for bug-guix@gnu.org; Fri, 08 Mar 2019 14:12:03 -0500 Received: from debbugs.gnu.org ([209.51.188.43]:51363) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1h2KuY-0002G2-BG for bug-guix@gnu.org; Fri, 08 Mar 2019 14:12:02 -0500 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1h2KuY-0001JV-45 for bug-guix@gnu.org; Fri, 08 Mar 2019 14:12:02 -0500 Sender: "Debbugs-submit" Resent-Message-ID: In-Reply-To: <8736o08494.fsf@nckx> List-Id: Bug reports for GNU Guix List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: bug-guix-bounces+gcggb-bug-guix=m.gmane.org@gnu.org Sender: "bug-Guix" To: Tobias Geerinckx-Rice Cc: 34768@debbugs.gnu.org ------JJ6QDKAQV0NQ5ODE4KU1U8V1GH1N3N Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Tobias Geerinckx-Rice skre= v: (6 mars 2019 15:31:13 CET) >Hullo, > >swedebugia wrote: >> I'm trying to build New Moon (unbranded palemoon browser) in=20 >> parabola=2E >> The tmpfs i 2G which is way too little it seems=2E >> >> I read the man page of guix-daemon and the manual and could not=20 >> find a >> way to point guix-daemon to another tmpdir than /tmp=2E > >Well, TMPDIR=3D should Just Work :-) > >You'll have to set it to the environment of guix-daemon (and=20 >restart the daemon) for it to have any effect=2E How that's done=20 >depends on your service manager/init system=2E > >I guess a =E2=80=98set-tmpdir=E2=80=99 RPC could be added to the protocol= =2E I=20 >haven't thought through the security implications, and IMO it's=20 >just papering over the real bug, which is=E2=80=A6 > >> Is this a bug? > >Not sure what exactly you're referring to, but yes, I do think=20 >that using /tmp instead of /var/tmp (or any non-tmpfs) is a bug by=20 >modern(?) conventions=2E It's guaranteed to break eventually on=20 >almost every non-Guix System, and when it doesn't it's an abuse of=20 >RAM and swap space=2E > >Kind regards, > >T G-R Hi =F0=9F=98=83 Thanks for the details=2E I agree that it is a bug=2E I solved it by masking tmp=2Emount in systemd according to the arch wiki a= nd thus disabled tmpfs for the whole thing=2E --=20 Sent from my k-9 mail for Android=2E ------JJ6QDKAQV0NQ5ODE4KU1U8V1GH1N3N Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable
Tobias Geerinckx-Rice &= lt;somebody@not-sent-or-endorsed-by=2Etobias=2Egr> skrev: (6 mars 2019 1= 5:31:13 CET)
Hullo,

swedebugia wrote:
 I'm trying to build New Moon (unbranded pa=
lemoon browser) in 
 parabola=2E
 The tmpfs i 2G which is way too lit=
tle it seems=2E

 I read the man page of guix-daemon and the manual a=
nd could not 
 find a
 way to point guix-daemon to another tmpdir tha=
n /tmp=2E

Well, TMPDIR=3D should Just Work :-)

Y=
ou'll have to set it to the environment of guix-daemon (and 
restart the=
 daemon) for it to have any effect=2E  How that's done 
depends on your =
service manager/init system=2E

I guess a =E2=80=98set-tmpdir=E2=80=
=99 RPC could be added to the protocol=2E  I 
haven't thought through th=
e security implications, and IMO it's 
just papering over the real bug, =
which is=E2=80=A6

I=
s this a bug?

Not sure what exactly you're referring to=
, but yes, I do think 
that using /tmp instead of /var/tmp (or any non-t=
mpfs) is a bug by 
modern(?) conventions=2E  It's guaranteed to break ev=
entually on 
almost every non-Guix System, and when it doesn't it's an a=
buse of 
RAM and swap space=2E

Kind regards,

T G-R

Hi =F0=9F=98=83

Thanks for th= e details=2E I agree that it is a bug=2E

I solved it by masking tmp= =2Emount in systemd according to the arch wiki and thus disabled tmpfs for = the whole thing=2E
--
Sent from my k-9 mail for Android=2E
------JJ6QDKAQV0NQ5ODE4KU1U8V1GH1N3N--