Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.os.linux.misc > #75070 > unrolled thread

Floppies - Actual Question

Started byc186282 <c186282@nnada.net>
First post2025-09-24 00:02 -0400
Last post2025-09-28 01:00 -0700
Articles 20 on this page of 180 — 23 participants

Back to article view | Back to comp.os.linux.misc


Contents

  Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-24 00:02 -0400
    Re: Floppies - Actual Question Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-09-24 05:06 +0000
      Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-24 05:43 +0000
        Re: Floppies - Actual Question Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-09-27 15:21 +0000
          Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-27 14:17 -0400
            Re: Floppies - Actual Question Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-09-27 22:31 +0000
              Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-28 00:26 +0000
                Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-28 01:21 +0000
                Re: Floppies - Actual Question Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-09-28 10:21 +0000
                  Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-28 19:27 +0000
                    Re: Floppies - Actual Question Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-10-04 09:37 +0000
                      Re: Floppies - Actual Question Bobbie Sellers <bliss-sf4ever@dslextreme.com> - 2025-10-04 12:04 -0700
                  Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-09-29 14:12 +0200
                    Re: Floppies - Actual Question Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-10-04 09:43 +0000
                      Re: Floppies - Actual Question Rich <rich@example.invalid> - 2025-10-04 15:22 +0000
                        Re: Floppies - Actual Question Richard Kettlewell <invalid@invalid.invalid> - 2025-10-04 17:38 +0100
                        Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-04 20:34 +0200
                          Re: Floppies - Actual Question Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-10-04 20:53 +0000
                        Re: Floppies - Actual Question Pancho <Pancho.Jones@protonmail.com> - 2025-10-04 22:56 +0100
                          Re: Floppies - Actual Question Richard Kettlewell <invalid@invalid.invalid> - 2025-10-05 09:51 +0100
                            Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-06 02:17 -0400
                            Re: Floppies - Actual Question Pancho <Pancho.Jones@protonmail.com> - 2025-10-07 09:56 +0100
                              Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-08 01:55 -0400
                                Re: Floppies - Actual Question Pancho <Pancho.Jones@protonmail.com> - 2025-10-08 12:58 +0100
                                  Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 02:36 -0400
                              Re: Floppies - Actual Question John Ames <commodorejohn@gmail.com> - 2025-10-08 09:02 -0700
                                Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-08 17:35 +0100
                                  Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-10 09:29 +0200
                                    Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-10 09:58 +0100
                                      Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-10 13:45 +0200
                                        Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-10 14:02 +0100
                                          Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-10 21:17 +0200
                                            Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-10-11 00:55 +0000
                                          Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-11 02:35 -0400
                                Partitioning (was Re: Floppies - Actual Question "David W. Hodgins" <dwhodgins@nomail.afraid.org> - 2025-10-08 14:54 -0400
                                  Re: Partitioning (was Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-10-08 21:15 +0000
                                  Re: Partitioning (was Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-09 12:35 +0100
                                Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-10-08 21:13 +0000
                                  Re: Floppies - Actual Question John Ames <commodorejohn@gmail.com> - 2025-10-08 14:29 -0700
                                    Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-09 12:37 +0100
                                      Re: Floppies - Actual Question Rich <rich@example.invalid> - 2025-10-09 15:07 +0000
                                        inodes, customizing partitions "Carlos E.R." <robin_listas@es.invalid> - 2025-10-10 09:45 +0200
                                          Re: inodes, customizing partitions c186282 <c186282@nnada.net> - 2025-10-11 00:57 -0400
                                            Re: inodes, customizing partitions "Carlos E.R." <robin_listas@es.invalid> - 2025-10-11 15:28 +0200
                                              Re: inodes, customizing partitions Nuno Silva <nunojsilva@invalid.invalid> - 2025-10-11 23:46 +0100
                                                Re: inodes, customizing partitions "Carlos E.R." <robin_listas@es.invalid> - 2025-10-12 02:42 +0200
                                                  Re: inodes, customizing partitions c186282 <c186282@nnada.net> - 2025-10-13 03:13 -0400
                                                    Re: inodes, customizing partitions "Carlos E.R." <robin_listas@es.invalid> - 2025-10-13 10:53 +0200
                                                Re: inodes, customizing partitions c186282 <c186282@nnada.net> - 2025-10-11 22:49 -0400
                                              Re: inodes, customizing partitions c186282 <c186282@nnada.net> - 2025-10-11 21:51 -0400
                                              Re: inodes, customizing partitions c186282 <c186282@nnada.net> - 2025-10-13 03:10 -0400
                                      Re: Floppies - Actual Question John Ames <commodorejohn@gmail.com> - 2025-10-09 09:00 -0700
                                        Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 21:24 -0400
                                          Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-10 09:52 +0100
                                            Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-11 01:45 -0400
                                              Re: Floppies - Actual Question Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-10-11 17:48 +0000
                                                Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-11 22:11 -0400
                                  Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 02:56 -0400
                                Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 02:50 -0400
                          Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-10-05 20:25 +0000
                      Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-04 20:35 +0200
              Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-28 09:55 +0100
              Re: Floppies - Actual Question Richard Kettlewell <invalid@invalid.invalid> - 2025-09-28 11:08 +0100
                Re: Floppies - Actual Question Stéphane CARPENTIER <sc@fiat-linux.fr> - 2025-09-28 12:19 +0000
                  Re: Floppies - Actual Question Richard Kettlewell <invalid@invalid.invalid> - 2025-09-28 15:07 +0100
                    Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-28 15:36 +0100
                      Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-28 23:35 +0000
                        Re: Floppies - Actual Question Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-09-29 00:28 +0000
                          Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-28 21:08 -0400
                          Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-29 05:31 +0000
                            Re: Floppies - Actual Question Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-09-29 16:32 +0000
                              Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-29 18:36 +0100
                              Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-29 19:30 +0000
                          Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-29 10:12 +0100
                            Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-29 19:33 +0000
                          Re: Floppies - Actual Question candycanearter07 <candycanearter07@candycanearter07.nomail.afraid> - 2025-09-29 19:50 +0000
                  Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-28 15:31 +0100
                Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-09-29 14:15 +0200
                  Re: Floppies - Actual Question Bobbie Sellers <bliss-sf4ever@dslextreme.com> - 2025-09-29 08:18 -0700
                    Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-29 19:27 +0000
            Re: Floppies - Actual Question Rich <rich@example.invalid> - 2025-09-28 00:31 +0000
              Re: Floppies - Actual Question Nuno Silva <nunojsilva@invalid.invalid> - 2025-09-28 11:19 +0100
                Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-28 15:22 +0100
                  Re: Floppies - Actual Question Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-09-28 18:07 +0000
                    Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-28 19:18 +0000
                    Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-28 23:55 +0100
            Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-28 09:53 +0100
            Re: Floppies - Actual Question Peter 'Shaggy' Haywood <phaywood@alphalink.com.au> - 2025-10-02 18:28 +1000
              Re: Floppies - Actual Question Tauno Voipio <tauno.voipio@notused.fi.invalid> - 2025-10-02 21:44 +0300
                Re: Floppies - Actual Question Nuno Silva <nunojsilva@invalid.invalid> - 2025-10-03 00:10 +0100
                  Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-03 00:35 -0400
                Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-03 00:32 -0400
              Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-03 00:21 -0400
                Re: Floppies - Actual Question Peter 'Shaggy' Haywood <phaywood@alphalink.com.au> - 2025-10-04 12:55 +1000
    Re: Floppies - Actual Question Andy Burns <usenet@andyburns.uk> - 2025-09-24 06:49 +0100
      Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-24 11:45 +0100
        Re: Floppies - Actual Question Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-09-24 16:53 +0000
          Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-24 14:35 -0400
          Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-24 15:05 -0400
            Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-24 21:32 -0400
          Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-24 20:38 +0000
            Re: Floppies - Actual Question rbowman <bowman@montana.com> - 2025-09-25 00:04 +0000
              Re: Floppies - Actual Question Robert Riches <spamtrap42@jacob21819.net> - 2025-09-25 00:44 +0000
                Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-24 20:58 -0400
                  Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-09-25 13:07 +0200
                    Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-25 12:48 +0100
                    Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-25 15:52 -0400
                    Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-25 22:09 +0000
                Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-25 09:44 +0100
                Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-09-25 13:04 +0200
                  Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-25 12:48 +0100
                    Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-09-25 22:15 +0200
                      Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-25 21:06 -0400
                  Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-09-25 22:10 +0000
                    Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-25 21:09 -0400
                    Re: Floppies - Actual Question Lawrence D’Oliveiro <ldo@nz.invalid> - 2025-10-06 22:08 +0000
                    Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-06 23:30 -0400
                      Re: Floppies - Actual Question Rich <rich@example.invalid> - 2025-10-07 15:16 +0000
                        Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-08 02:34 +0200
                          Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-08 01:05 -0400
                            Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-08 10:58 +0200
                              Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-08 11:06 +0100
                              Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-08 14:14 +0200
                                Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 02:44 -0400
                                  Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-09 21:10 +0200
                                    Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 21:36 -0400
                                      Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-10 09:54 +0100
                                        Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-11 01:57 -0400
                                      Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-10 13:49 +0200
                                        Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-11 02:29 -0400
                                          Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-11 14:32 +0200
                                Re: Floppies - Actual Question Joerg Walther <joerg.walther@magenta.de> - 2025-10-09 12:50 +0200
                                  Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-10-09 21:11 +0200
                                    Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 21:37 -0400
                                    Re: Floppies - Actual Question Joerg Walther <joerg.walther@magenta.de> - 2025-10-10 11:19 +0200
                                      Phone wallet "Carlos E.R." <robin_listas@es.invalid> - 2025-10-10 13:51 +0200
                                        Re: Phone wallet c186282 <c186282@nnada.net> - 2025-10-11 02:30 -0400
                                          Re: Phone wallet "Carlos E.R." <robin_listas@es.invalid> - 2025-10-11 14:32 +0200
                                            Re: Phone wallet c186282 <c186282@nnada.net> - 2025-10-11 21:35 -0400
                                              Re: Phone wallet "Carlos E.R." <robin_listas@es.invalid> - 2025-10-12 14:48 +0200
                                  Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 21:17 -0400
                                    Re: Floppies - Actual Question Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2025-10-10 06:51 +0000
                                      Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-10 03:29 -0400
                                      Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-10 09:55 +0100
                                        Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-11 02:10 -0400
                              Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-08 13:18 +0100
                                Re: Floppies - Actual Question Robert Riches <spamtrap42@jacob21819.net> - 2025-10-09 03:09 +0000
                                  Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 02:58 -0400
                                  Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-09 12:45 +0100
                                    Re: Floppies - Actual Question Robert Riches <spamtrap42@jacob21819.net> - 2025-10-10 02:53 +0000
                                      Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 23:44 -0400
                                Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 02:45 -0400
                              Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-09 02:38 -0400
                              Re: Floppies - Actual Question Rich <rich@example.invalid> - 2025-10-09 15:16 +0000
                                Faraday cages (was: Re: Floppies - Actual Question) Nuno Silva <nunojsilva@invalid.invalid> - 2025-10-09 17:02 +0100
                                  Re: Faraday cages Rich <rich@example.invalid> - 2025-10-09 18:48 +0000
                                  Re: Faraday cages The Natural Philosopher <tnp@invalid.invalid> - 2025-10-10 09:49 +0100
                                Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-10-10 09:47 +0100
                                  Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-11 01:05 -0400
                          Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-08 01:52 -0400
                        Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-10-08 01:38 -0400
              Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-24 20:45 -0400
                Re: Floppies - Actual Question rbowman <bowman@montana.com> - 2025-09-25 05:27 +0000
                  Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-25 15:42 -0400
                    Re: Floppies - Actual Question rbowman <bowman@montana.com> - 2025-09-26 04:59 +0000
                      Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-26 01:36 -0400
                        Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-27 09:12 +0100
                          Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-27 13:58 -0400
                        Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-27 03:11 -0400
                          Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-09-29 14:22 +0200
                            Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-29 23:03 -0400
                              Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-09-30 09:32 +0200
                        Re: Floppies - Actual Question Rich <rich@example.invalid> - 2025-09-26 15:19 +0000
            Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-24 20:23 -0400
              Re: Floppies - Actual Question The Natural Philosopher <tnp@invalid.invalid> - 2025-09-25 09:45 +0100
    Re: Floppies - Actual Question Marco Moock <mm@dorfdsl.de> - 2025-09-24 08:41 +0200
    Re: Floppies - Actual Question "Carlos E.R." <robin_listas@es.invalid> - 2025-09-24 12:04 +0200
      Re: Floppies - Actual Question c186282 <c186282@nnada.net> - 2025-09-24 14:34 -0400
    Re: Floppies - Actual Question John McCue <jmclnx@SPAMisBADgmail.com> - 2025-09-24 17:11 +0000
    Re: Floppies - Actual Question Daniel <me@sc1f1dan.com> - 2025-09-28 01:00 -0700

Page 3 of 9 — ← Prev page 1 2 [3] 4 5 6 7 8 9  Next page →


#75841

FromRich <rich@example.invalid>
Date2025-10-09 15:07 +0000
Message-ID<10c8j2q$2tcv3$1@dont-email.me>
In reply to#75837
The Natural Philosopher <tnp@invalid.invalid> wrote:
> On 08/10/2025 22:29, John Ames wrote:
>> On Wed, 8 Oct 2025 21:13:01 -0000 (UTC)
>> Lawrence D’Oliveiro <ldo@nz.invalid> wrote:
>> 
>>>> Having a separate partition with a (vaguely) well-organized
>>>> directory structure ...
>>>
>>> You can have the latter without the former. Remember, *nix systems
>>> maintain a single directory hierarchy, regardless of how the storage
>>> may actually be partitioned into separate volumes.
>> 
>> Certainly, but keeping it separate allows for some fine-tuning (e.g.
>> picking a larger inode size for a partition where small files will be
>> the exception rather than the rule.)
>> 
> Crikey, how long has THAT been an issue.

It /effectively/ ceased being an issue somewhere around about the 
latter ext3 or early ext4 filesystem days.  Except for *very* unsusal 
usage patterns (i.e., Usenet NNTP spools stored as files directly in 
the filesystem) it is no longer something anyone thinks about anymore.  
Yes, there's a "limit", but much like with "64-bit memory space" the 
limit is so large it is all but near impossible to actually hit it 
anymore.

A quick check here with 'df -i' shows all but one mounted filesystem in 
the single digit percentage usage of inodes.  The one 'worst' case (a 
4T disk for media encoding/transcoding and other misc usages) is only 
consuming 17% of its max inodes.

> Like the time you had to define how many TCP/IP buffers or virtual 
> ttys you had...  Suddenly there was enough memory and disk space to 
> not give a shit

Yep.

[toc] | [prev] | [next] | [standalone]


#75878 — inodes, customizing partitions

From"Carlos E.R." <robin_listas@es.invalid>
Date2025-10-10 09:45 +0200
Subjectinodes, customizing partitions
Message-ID<45qmrlx6oh.ln2@Telcontar.valinor>
In reply to#75841
On 2025-10-09 17:07, Rich wrote:
> The Natural Philosopher <tnp@invalid.invalid> wrote:
>> On 08/10/2025 22:29, John Ames wrote:
>>> On Wed, 8 Oct 2025 21:13:01 -0000 (UTC)
>>> Lawrence D’Oliveiro <ldo@nz.invalid> wrote:
>>>
>>>>> Having a separate partition with a (vaguely) well-organized
>>>>> directory structure ...
>>>>
>>>> You can have the latter without the former. Remember, *nix systems
>>>> maintain a single directory hierarchy, regardless of how the storage
>>>> may actually be partitioned into separate volumes.
>>>
>>> Certainly, but keeping it separate allows for some fine-tuning (e.g.
>>> picking a larger inode size for a partition where small files will be
>>> the exception rather than the rule.)
>>>
>> Crikey, how long has THAT been an issue.
> 
> It /effectively/ ceased being an issue somewhere around about the
> latter ext3 or early ext4 filesystem days.  Except for *very* unsusal
> usage patterns (i.e., Usenet NNTP spools stored as files directly in
> the filesystem) it is no longer something anyone thinks about anymore.
> Yes, there's a "limit", but much like with "64-bit memory space" the
> limit is so large it is all but near impossible to actually hit it
> anymore.
> 
> A quick check here with 'df -i' shows all but one mounted filesystem in
> the single digit percentage usage of inodes.  The one 'worst' case (a
> 4T disk for media encoding/transcoding and other misc usages) is only
> consuming 17% of its max inodes.

Telcontar:~ # df -i
Filesystem                         Inodes   IUsed      IFree IUse% Mounted on
/dev/nvme0n1p5                    9830400 2135223    7695177   22% /
...
/dev/sdc9                        16777216 1632362   15144854   10% /data/WasReiserfs		(holds the news spool)
...
/dev/sdb10                       15295056  943965   14351091    7% /data/storage_b
/dev/sdc10                        2621440  131132    2490308    6% /usr/local
...



-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [next] | [standalone]


#75918 — Re: inodes, customizing partitions

Fromc186282 <c186282@nnada.net>
Date2025-10-11 00:57 -0400
SubjectRe: inodes, customizing partitions
Message-ID<Q7mdneCN_8orfnT1nZ2dnZfqnPGdnZ2d@giganews.com>
In reply to#75878
On 10/10/25 03:45, Carlos E.R. wrote:
> On 2025-10-09 17:07, Rich wrote:
>> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>> On 08/10/2025 22:29, John Ames wrote:
>>>> On Wed, 8 Oct 2025 21:13:01 -0000 (UTC)
>>>> Lawrence D’Oliveiro <ldo@nz.invalid> wrote:
>>>>
>>>>>> Having a separate partition with a (vaguely) well-organized
>>>>>> directory structure ...
>>>>>
>>>>> You can have the latter without the former. Remember, *nix systems
>>>>> maintain a single directory hierarchy, regardless of how the storage
>>>>> may actually be partitioned into separate volumes.
>>>>
>>>> Certainly, but keeping it separate allows for some fine-tuning (e.g.
>>>> picking a larger inode size for a partition where small files will be
>>>> the exception rather than the rule.)
>>>>
>>> Crikey, how long has THAT been an issue.
>>
>> It /effectively/ ceased being an issue somewhere around about the
>> latter ext3 or early ext4 filesystem days.  Except for *very* unsusal
>> usage patterns (i.e., Usenet NNTP spools stored as files directly in
>> the filesystem) it is no longer something anyone thinks about anymore.
>> Yes, there's a "limit", but much like with "64-bit memory space" the
>> limit is so large it is all but near impossible to actually hit it
>> anymore.
>>
>> A quick check here with 'df -i' shows all but one mounted filesystem in
>> the single digit percentage usage of inodes.  The one 'worst' case (a
>> 4T disk for media encoding/transcoding and other misc usages) is only
>> consuming 17% of its max inodes.
> 
> Telcontar:~ # df -i
> Filesystem                         Inodes   IUsed      IFree IUse% 
> Mounted on
> /dev/nvme0n1p5                    9830400 2135223    7695177   22% /
> ...
> /dev/sdc9                        16777216 1632362   15144854   10% / 
> data/WasReiserfs        (holds the news spool)
> ...
> /dev/sdb10                       15295056  943965   14351091    7% / 
> data/storage_b
> /dev/sdc10                        2621440  131132    2490308    6% /usr/ 
> local
> ...

   We tend to think about inodes less and less these
   days as both CPUs got more powerful and disks got
   much larger and faster.

   However sometimes you DO need to think about them,
   especially for certain environments, usually servers.
   Are you going to be storing a few BIG files or a
   zillion TINY files ? Adjustments to block sizes
   and inode params may be advised.

   Office environ ... bad old days you'd get mass quantities
   of .txt/com/bin files to back up. As such small block
   sizes and more inodes were wise. Today it's all Excel,
   Word and PDFs, relatively huge. Larger, 4k/8k, block
   sizes are more economical but the VOLUME hasn't decreased
   so you may still need extra inodes compared to a 'home'
   system.

   Linux distro defaults are just "guesses", eval
   your particular need and adjust accordingly. Did lots
   of backup servers, some with up to 12 big SATA drives
   and home-made aux cooling. Always wanted to maximize
   the capacity/capability. Bigger disks were bigger money.

   Today, a lot backup to 'the cloud'. It has its uses but
   beware Vlad's boyz - it COULD all go away rather suddenly.
   A combo of local AND cloud backups is always the best tact.

   Also pre-encrypted every backup file before sending it
   to cloud ... don't even trust local corps anymore, too
   much $$$ incentive to 'mine'.

[toc] | [prev] | [next] | [standalone]


#75945 — Re: inodes, customizing partitions

From"Carlos E.R." <robin_listas@es.invalid>
Date2025-10-11 15:28 +0200
SubjectRe: inodes, customizing partitions
Message-ID<hk2qrlxifk.ln2@Telcontar.valinor>
In reply to#75918
On 2025-10-11 06:57, c186282 wrote:
> On 10/10/25 03:45, Carlos E.R. wrote:
>> On 2025-10-09 17:07, Rich wrote:
>>> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>>> On 08/10/2025 22:29, John Ames wrote:
>>>>> On Wed, 8 Oct 2025 21:13:01 -0000 (UTC)
>>>>> Lawrence D’Oliveiro <ldo@nz.invalid> wrote:

...

>>> A quick check here with 'df -i' shows all but one mounted filesystem in
>>> the single digit percentage usage of inodes.  The one 'worst' case (a
>>> 4T disk for media encoding/transcoding and other misc usages) is only
>>> consuming 17% of its max inodes.
>>
>> Telcontar:~ # df -i
>> Filesystem                         Inodes   IUsed      IFree IUse% 
>> Mounted on
>> /dev/nvme0n1p5                    9830400 2135223    7695177   22% /
>> ...
>> /dev/sdc9                        16777216 1632362   15144854   10% / 
>> data/WasReiserfs        (holds the news spool)
>> ...
>> /dev/sdb10                       15295056  943965   14351091    7% / 
>> data/storage_b
>> /dev/sdc10                        2621440  131132    2490308    6% / 
>> usr/ local
>> ...
> 
>    We tend to think about inodes less and less these
>    days as both CPUs got more powerful and disks got
>    much larger and faster.
> 
>    However sometimes you DO need to think about them,
>    especially for certain environments, usually servers.
>    Are you going to be storing a few BIG files or a
>    zillion TINY files ? Adjustments to block sizes
>    and inode params may be advised.

Well, you can see that my root filesystem has a high percent of used 
inodes, more than the /data/WasReiserfs which holds the news spool and 
is customized for many small files.

I'd like to know the inode percent per directory in a partition. Chatgpt 
suggests

du --inodes -d 1

-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [next] | [standalone]


#75960 — Re: inodes, customizing partitions

FromNuno Silva <nunojsilva@invalid.invalid>
Date2025-10-11 23:46 +0100
SubjectRe: inodes, customizing partitions
Message-ID<10cemnn$13dfg$2@dont-email.me>
In reply to#75945
On 2025-10-11, Carlos E.R. wrote:

> I'd like to know the inode percent per directory in a
> partition. Chatgpt suggests
>
> du --inodes -d 1

Is there some implementation of the du utility that provides a
percentage with that invocation?

(I guess an actually intelligent AI would have attempted to precisely
clarify what "percent per directory" is expected to mean before
suggesting any code or command?)

-- 
Nuno Silva

[toc] | [prev] | [next] | [standalone]


#75964 — Re: inodes, customizing partitions

From"Carlos E.R." <robin_listas@es.invalid>
Date2025-10-12 02:42 +0200
SubjectRe: inodes, customizing partitions
Message-ID<25arrlx2l1.ln2@Telcontar.valinor>
In reply to#75960
On 2025-10-12 00:46, Nuno Silva wrote:
> On 2025-10-11, Carlos E.R. wrote:
> 
>> I'd like to know the inode percent per directory in a
>> partition. Chatgpt suggests
>>
>> du --inodes -d 1
> 
> Is there some implementation of the du utility that provides a
> percentage with that invocation?

I don't know.

> 
> (I guess an actually intelligent AI would have attempted to precisely
> clarify what "percent per directory" is expected to mean before
> suggesting any code or command?)
> 

It suggested more methods, I took the simplest :-)

To check **inode usage per directory** in Linux — that is, to see **how many inodes (files/directories)** each subdirectory is using — you can use a combination of tools like `find`, `awk`, and `du`.

Here are the most useful and efficient ways:

---

## ✅ **1. Using `find` to count inodes per subdirectory**

This gives you the number of inodes (i.e. files + dirs) in each top-level subdirectory:

```bash
for dir in *; do
   [ -d "$dir" ] && echo -n "$dir: " && find "$dir" | wc -l
done
```

> This recursively counts all files and directories inside each folder — i.e. inode usage.

---

## ✅ **2. Using `find` + `awk` for sorted output**

List inode counts for subdirectories in current dir, sorted by usage:

```bash
find . -xdev -type f -printf '%h\n' | sort | uniq -c | sort -nr
```

* `-xdev`: restricts to current filesystem (avoid crossing mount points)
* `-type f`: only counts files (you can remove this to include dirs too)
* `%h`: prints parent directory (for grouping)

---

## ✅ **3. Using `du` (but for inodes) — with `--inodes`**

If your version of `du` supports it (on newer GNU coreutils), use:

```bash
du -a --inodes
```

Or for top-level dirs:

```bash
du --inodes -d 1
```

> This shows inode usage (number of files/directories) like `du` shows disk usage.

---

## ✅ **4. Get total inode usage of filesystem**

To check **total inodes used/available** on a filesystem (not per directory):

```bash
df -i
```

Sample output:

```
Filesystem     Inodes  IUsed  IFree IUse% Mounted on
/dev/sda1      524288  10234 514054    2% /
```

---

## 🧠 Pro Tip: What’s an inode?

Each **file**, **directory**, **symlink**, etc. uses **one inode**. On filesystems like ext4, the number of inodes is fixed at creation time. You can run out of inodes even if disk space is free — especially if you create millions of tiny files (like maildirs, caches, etc.).

---

## ✅ Summary

| Goal                   | Command                      |        |         |           |
| ---------------------- | ---------------------------- | ------ | ------- | --------- |
| Inode usage per subdir | `find "$dir"                 | wc -l` |         |           |
| Sorted inode usage     | `find . -xdev -printf '%h\n' | sort   | uniq -c | sort -nr` |
| Inode-aware du         | `du --inodes -d 1`           |        |         |           |
| Total inodes used/free | `df -i`                      |        |         |           |

Let me know if you want a script to alert when inode usage gets too high.



--------

The invocation:

find . -xdev -type f -printf '%h\n' | sort | uniq -c | sort -nr | less

found out where the inodes are going to in my root:

   76119 ./tmp/imapsync_cache/legolas-e.valinor/cer/telcontar.valinor/cer/alpine_r_gmx/alpine_r_gmx
   62998 ./tmp/imapsync_cache/legolas-e.valinor/cer/telcontar.valinor/cer/alpine_r_tl/alpine_r_tl
   49326 ./tmp/imapsync_cache/legolas.valinor/cer/telcontar.valinor/cer/alpine_r_gmx/alpine_r_gmx

caches of mail folders, created by "imapsync", seems one zero sized file per email. /tmp is in root (/).






-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [next] | [standalone]


#76029 — Re: inodes, customizing partitions

Fromc186282 <c186282@nnada.net>
Date2025-10-13 03:13 -0400
SubjectRe: inodes, customizing partitions
Message-ID<pY-dnaHGLdwNO3H1nZ2dnZfqn_GdnZ2d@giganews.com>
In reply to#75964
On 10/11/25 20:42, Carlos E.R. wrote:
> On 2025-10-12 00:46, Nuno Silva wrote:
>> On 2025-10-11, Carlos E.R. wrote:
>>
>>> I'd like to know the inode percent per directory in a
>>> partition. Chatgpt suggests
>>>
>>> du --inodes -d 1
>>
>> Is there some implementation of the du utility that provides a
>> percentage with that invocation?
> 
> I don't know.
> 
>>
>> (I guess an actually intelligent AI would have attempted to precisely
>> clarify what "percent per directory" is expected to mean before
>> suggesting any code or command?)
>>
> 
> It suggested more methods, I took the simplest :-)
> 
> To check **inode usage per directory** in Linux — that is, to see **how 
> many inodes (files/directories)** each subdirectory is using — you can 
> use a combination of tools like `find`, `awk`, and `du`.
> 
> Here are the most useful and efficient ways:
> 
> ---
> 
> ## ✅ **1. Using `find` to count inodes per subdirectory**
> 
> This gives you the number of inodes (i.e. files + dirs) in each top- 
> level subdirectory:
> 
> ```bash
> for dir in *; do
>    [ -d "$dir" ] && echo -n "$dir: " && find "$dir" | wc -l
> done
> ```
> 
>> This recursively counts all files and directories inside each folder — 
>> i.e. inode usage.
> 
> ---
> 
> ## ✅ **2. Using `find` + `awk` for sorted output**
> 
> List inode counts for subdirectories in current dir, sorted by usage:
> 
> ```bash
> find . -xdev -type f -printf '%h\n' | sort | uniq -c | sort -nr
> ```
> 
> * `-xdev`: restricts to current filesystem (avoid crossing mount points)
> * `-type f`: only counts files (you can remove this to include dirs too)
> * `%h`: prints parent directory (for grouping)
> 
> ---
> 
> ## ✅ **3. Using `du` (but for inodes) — with `--inodes`**
> 
> If your version of `du` supports it (on newer GNU coreutils), use:
> 
> ```bash
> du -a --inodes
> ```
> 
> Or for top-level dirs:
> 
> ```bash
> du --inodes -d 1
> ```
> 
>> This shows inode usage (number of files/directories) like `du` shows 
>> disk usage.
> 
> ---
> 
> ## ✅ **4. Get total inode usage of filesystem**
> 
> To check **total inodes used/available** on a filesystem (not per 
> directory):
> 
> ```bash
> df -i
> ```
> 
> Sample output:
> 
> ```
> Filesystem     Inodes  IUsed  IFree IUse% Mounted on
> /dev/sda1      524288  10234 514054    2% /
> ```
> 
> ---
> 
> ## 🧠 Pro Tip: What’s an inode?
> 
> Each **file**, **directory**, **symlink**, etc. uses **one inode**. On 
> filesystems like ext4, the number of inodes is fixed at creation time. 
> You can run out of inodes even if disk space is free — especially if you 
> create millions of tiny files (like maildirs, caches, etc.).
> 
> ---
> 
> ## ✅ Summary
> 
> | Goal                   | Command                      |        | 
>           |           |
> | ---------------------- | ---------------------------- | ------ | 
> ------- | --------- |
> | Inode usage per subdir | `find "$dir"                 | wc -l` | 
>           |           |
> | Sorted inode usage     | `find . -xdev -printf '%h\n' | sort   | uniq 
> -c | sort -nr` |
> | Inode-aware du         | `du --inodes -d 1`           |        | 
>           |           |
> | Total inodes used/free | `df -i`                      |        | 
>           |           |
> 
> Let me know if you want a script to alert when inode usage gets too high.
> 
> 
> 
> --------
> 
> The invocation:
> 
> find . -xdev -type f -printf '%h\n' | sort | uniq -c | sort -nr | less
> 
> found out where the inodes are going to in my root:
> 
>    76119 ./tmp/imapsync_cache/legolas-e.valinor/cer/telcontar.valinor/ 
> cer/alpine_r_gmx/alpine_r_gmx
>    62998 ./tmp/imapsync_cache/legolas-e.valinor/cer/telcontar.valinor/ 
> cer/alpine_r_tl/alpine_r_tl
>    49326 ./tmp/imapsync_cache/legolas.valinor/cer/telcontar.valinor/cer/ 
> alpine_r_gmx/alpine_r_gmx
> 
> caches of mail folders, created by "imapsync", seems one zero sized file 
> per email. /tmp is in root (/).

   'df -i" will tell you exactly what you need.

[toc] | [prev] | [next] | [standalone]


#76038 — Re: inodes, customizing partitions

From"Carlos E.R." <robin_listas@es.invalid>
Date2025-10-13 10:53 +0200
SubjectRe: inodes, customizing partitions
Message-ID<0arurlxelj.ln2@Telcontar.valinor>
In reply to#76029
On 2025-10-13 09:13, c186282 wrote:
> On 10/11/25 20:42, Carlos E.R. wrote:
>> On 2025-10-12 00:46, Nuno Silva wrote:
>>> On 2025-10-11, Carlos E.R. wrote:

...

> 
>    'df -i" will tell you exactly what you need.

No, it doesn't. That's the issue.

Telcontar:~ # df -i
Filesystem                         Inodes   IUsed      IFree IUse% Mounted on
/dev/nvme0n1p5                    9830400 2135223    7695177   22% /


I need to know where exactly the inodes are going.



Telcontar:/ # du --inodes -d 1
1	./lost+found
729	./boot
3	./%{_rundir}
2	./.config
10	./.razor
6263	./CopiaSeguridadParcial
62	./aeat
147	./bin

here it stops for a long time. It ignores ctrl-C, by the way.


Telcontar:/ # find . -xdev -type f -printf '%h\n' | sort | uniq -c | sort -nr | less
  76119 ./tmp/imapsync_cache/legolas-e.valinor/cer/telcontar.valinor/cer/alpine_r_gmx/alpine_r_gmx
   62998 ./tmp/imapsync_cache/legolas-e.valinor/cer/telcontar.valinor/cer/alpine_r_tl/alpine_r_tl
   49326 ./tmp/imapsync_cache/legolas.valinor/cer/telcontar.valinor/cer/alpine_r_gmx/alpine_r_gmx
  

results are fast, and I have the data that I need: what directories are loaded with inodes in the root filesystem, no, the root partition. du --inodes is still running.


This one is still running in another terminal:

Telcontar:/ # du --inodes -d 1
1	./lost+found
729	./boot
3	./%{_rundir}
2	./.config
10	./.razor
6263	./CopiaSeguridadParcial
62	./aeat
147	./bin
3902694	./data
1570	./dev
6623	./etc
6291	./etc_13.1
182164	./home
35368	./home1
1406028	./home_aux
17485	./lib
202	./lib64
1	./.cache
2	./media


long time here

3266859	./mnt
16	./new
84459	./opt
48	./other
du: cannot access './proc/12757/task/12757/fd/3': No such file or directory
du: cannot access './proc/12757/task/12757/fdinfo/3': No such file or directory
...
1941858	./proc
55126	./root
du: cannot access './run/user/1000/doc': Permission denied
du: cannot access './run/user/1000/gvfs': Permission denied
3385	./run
298	./sbin
1	./selinux
6440	./srv
1	./subdomain
100488	./sys
5	./tftpboot
591728	./tmp
1430987	./usr
28623	./var
28	./windows
5	./transfernotebook
13076004	.
Telcontar:/ #


Well, after a long time (an hour?) it prints the information, but hard to interpret.

This is exactly what I need:

find . -xdev -type f -printf '%h\n' | sort | uniq -c | sort -nr | less


-- 
Cheers, Carlos.
ES🇪🇸, EU🇪🇺;

[toc] | [prev] | [next] | [standalone]


#75980 — Re: inodes, customizing partitions

Fromc186282 <c186282@nnada.net>
Date2025-10-11 22:49 -0400
SubjectRe: inodes, customizing partitions
Message-ID<W-WcnRNsBfS0inb1nZ2dnZfqnPhj4p2d@giganews.com>
In reply to#75960
On 10/11/25 18:46, Nuno Silva wrote:
> On 2025-10-11, Carlos E.R. wrote:
> 
>> I'd like to know the inode percent per directory in a
>> partition. Chatgpt suggests
>>
>> du --inodes -d 1
> 
> Is there some implementation of the du utility that provides a
> percentage with that invocation?

   No. Use "df -i"

   You CAN specify "/" too if you want
   less noise.

> (I guess an actually intelligent AI would have attempted to precisely
> clarify what "percent per directory" is expected to mean before
> suggesting any code or command?)

   Hmm ... language subtleties ... LLMs don't
   really love those - will probably try to
   bullshit you  :-)

[toc] | [prev] | [next] | [standalone]


#75972 — Re: inodes, customizing partitions

Fromc186282 <c186282@nnada.net>
Date2025-10-11 21:51 -0400
SubjectRe: inodes, customizing partitions
Message-ID<W-WcnRtsBfQNlHb1nZ2dnZfqnPhj4p2d@giganews.com>
In reply to#75945
On 10/11/25 09:28, Carlos E.R. wrote:
> On 2025-10-11 06:57, c186282 wrote:
>> On 10/10/25 03:45, Carlos E.R. wrote:
>>> On 2025-10-09 17:07, Rich wrote:
>>>> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>>>> On 08/10/2025 22:29, John Ames wrote:
>>>>>> On Wed, 8 Oct 2025 21:13:01 -0000 (UTC)
>>>>>> Lawrence D’Oliveiro <ldo@nz.invalid> wrote:
> 
> ...
> 
>>>> A quick check here with 'df -i' shows all but one mounted filesystem in
>>>> the single digit percentage usage of inodes.  The one 'worst' case (a
>>>> 4T disk for media encoding/transcoding and other misc usages) is only
>>>> consuming 17% of its max inodes.
>>>
>>> Telcontar:~ # df -i
>>> Filesystem                         Inodes   IUsed      IFree IUse% 
>>> Mounted on
>>> /dev/nvme0n1p5                    9830400 2135223    7695177   22% /
>>> ...
>>> /dev/sdc9                        16777216 1632362   15144854   10% / 
>>> data/WasReiserfs        (holds the news spool)
>>> ...
>>> /dev/sdb10                       15295056  943965   14351091    7% / 
>>> data/storage_b
>>> /dev/sdc10                        2621440  131132    2490308    6% / 
>>> usr/ local
>>> ...
>>
>>    We tend to think about inodes less and less these
>>    days as both CPUs got more powerful and disks got
>>    much larger and faster.
>>
>>    However sometimes you DO need to think about them,
>>    especially for certain environments, usually servers.
>>    Are you going to be storing a few BIG files or a
>>    zillion TINY files ? Adjustments to block sizes
>>    and inode params may be advised.
> 
> Well, you can see that my root filesystem has a high percent of used 
> inodes, more than the /data/WasReiserfs which holds the news spool and 
> is customized for many small files.
> 
> I'd like to know the inode percent per directory in a partition. Chatgpt 
> suggests
> 
> du --inodes -d 1

   That shows how many are used, but not a percentage of
   what's available.

   "df -i" is far better at that.

   Modern distros, I doubt you'll ever run out of inodes
   under almost any circumstance. But, as said, if you
   store vast quantities of tiny files - some DBs might
   keep those, maybe some audio mixing platforms too - you
   may want to look into formatting with smaller blocks and
   and more inodes. Exceptional cases do exist. One size
   never fits all.

[toc] | [prev] | [next] | [standalone]


#76028 — Re: inodes, customizing partitions

Fromc186282 <c186282@nnada.net>
Date2025-10-13 03:10 -0400
SubjectRe: inodes, customizing partitions
Message-ID<1NucnWl3DZpWOHH1nZ2dnZfqnPGdnZ2d@giganews.com>
In reply to#75945
On 10/11/25 09:28, Carlos E.R. wrote:
> On 2025-10-11 06:57, c186282 wrote:
>> On 10/10/25 03:45, Carlos E.R. wrote:
>>> On 2025-10-09 17:07, Rich wrote:
>>>> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>>>> On 08/10/2025 22:29, John Ames wrote:
>>>>>> On Wed, 8 Oct 2025 21:13:01 -0000 (UTC)
>>>>>> Lawrence D’Oliveiro <ldo@nz.invalid> wrote:
> 
> ...
> 
>>>> A quick check here with 'df -i' shows all but one mounted filesystem in
>>>> the single digit percentage usage of inodes.  The one 'worst' case (a
>>>> 4T disk for media encoding/transcoding and other misc usages) is only
>>>> consuming 17% of its max inodes.
>>>
>>> Telcontar:~ # df -i
>>> Filesystem                         Inodes   IUsed      IFree IUse% 
>>> Mounted on
>>> /dev/nvme0n1p5                    9830400 2135223    7695177   22% /
>>> ...
>>> /dev/sdc9                        16777216 1632362   15144854   10% / 
>>> data/WasReiserfs        (holds the news spool)
>>> ...
>>> /dev/sdb10                       15295056  943965   14351091    7% / 
>>> data/storage_b
>>> /dev/sdc10                        2621440  131132    2490308    6% / 
>>> usr/ local
>>> ...
>>
>>    We tend to think about inodes less and less these
>>    days as both CPUs got more powerful and disks got
>>    much larger and faster.
>>
>>    However sometimes you DO need to think about them,
>>    especially for certain environments, usually servers.
>>    Are you going to be storing a few BIG files or a
>>    zillion TINY files ? Adjustments to block sizes
>>    and inode params may be advised.
> 
> Well, you can see that my root filesystem has a high percent of used 
> inodes, more than the /data/WasReiserfs which holds the news spool and 
> is customized for many small files.
> 
> I'd like to know the inode percent per directory in a partition. Chatgpt 
> suggests
> 
> du --inodes -d 1
> 

   'df -i" is better.

[toc] | [prev] | [next] | [standalone]


#75843

FromJohn Ames <commodorejohn@gmail.com>
Date2025-10-09 09:00 -0700
Message-ID<20251009090008.0000560f@gmail.com>
In reply to#75837
On Thu, 9 Oct 2025 12:37:30 +0100
The Natural Philosopher <tnp@invalid.invalid> wrote:

> > Certainly, but keeping it separate allows for some fine-tuning (e.g.
> > picking a larger inode size for a partition where small files will
> > be the exception rather than the rule.)
> >   
> Crikey, how long has THAT been an issue. Like the time you had to
> define how many TCP/IP buffers or virtual ttys you had... Suddenly
> there was enough memory and disk space to not give a shit

Everybody can feel free to make their own judgements - but in my own
experience, "more than enough disk space" always manages to grind its
way down to "not as much space as I thought" eventually. Parkinson's
Law is an insidious force...

[toc] | [prev] | [next] | [standalone]


#75857

Fromc186282 <c186282@nnada.net>
Date2025-10-09 21:24 -0400
Message-ID<JRydnaPwbsbI_XX1nZ2dnZfqn_ednZ2d@giganews.com>
In reply to#75843
On 10/9/25 12:00, John Ames wrote:
> On Thu, 9 Oct 2025 12:37:30 +0100
> The Natural Philosopher <tnp@invalid.invalid> wrote:
> 
>>> Certainly, but keeping it separate allows for some fine-tuning (e.g.
>>> picking a larger inode size for a partition where small files will
>>> be the exception rather than the rule.)
>>>    
>> Crikey, how long has THAT been an issue. Like the time you had to
>> define how many TCP/IP buffers or virtual ttys you had... Suddenly
>> there was enough memory and disk space to not give a shit
> 
> Everybody can feel free to make their own judgements - but in my own
> experience, "more than enough disk space" always manages to grind its
> way down to "not as much space as I thought" eventually. Parkinson's
> Law is an insidious force...


   Yep, there's NEVER "enough" space  :-)

   Kind of like building a bigger closet, it too
   will be packed to overflowing within a year.

   A lot of people work with 8+ K 120fps video now,
   that'll fill drive and memory up really quick.

[toc] | [prev] | [next] | [standalone]


#75881

FromThe Natural Philosopher <tnp@invalid.invalid>
Date2025-10-10 09:52 +0100
Message-ID<10cahg9$3u22g$4@dont-email.me>
In reply to#75857
On 10/10/2025 02:24, c186282 wrote:
> On 10/9/25 12:00, John Ames wrote:
>> On Thu, 9 Oct 2025 12:37:30 +0100
>> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>
>>>> Certainly, but keeping it separate allows for some fine-tuning (e.g.
>>>> picking a larger inode size for a partition where small files will
>>>> be the exception rather than the rule.)
>>> Crikey, how long has THAT been an issue. Like the time you had to
>>> define how many TCP/IP buffers or virtual ttys you had... Suddenly
>>> there was enough memory and disk space to not give a shit
>>
>> Everybody can feel free to make their own judgements - but in my own
>> experience, "more than enough disk space" always manages to grind its
>> way down to "not as much space as I thought" eventually. Parkinson's
>> Law is an insidious force...
> 
> 
>    Yep, there's NEVER "enough" space  :-)
> 
Actually I seem to have arrived at a limit for 'personal' use.

Running with a couple of terabytes for media and 24GB or RAM I simply 
have now 'more than enough'


>    Kind of like building a bigger closet, it too
>    will be packed to overflowing within a year.
> 
Then throw those old clothes that no longer fit away. Accept you will 
never have a 30" waist ever again..

>    A lot of people work with 8+ K 120fps video now,
>    that'll fill drive and memory up really quick.
> 

-- 
Karl Marx said religion is the opium of the people.
But Marxism is the crack cocaine.

[toc] | [prev] | [next] | [standalone]


#75920

Fromc186282 <c186282@nnada.net>
Date2025-10-11 01:45 -0400
Message-ID<g-ecnYyNGMZrc3T1nZ2dnZfqnPudnZ2d@giganews.com>
In reply to#75881
n
On 10/10/25 04:52, The Natural Philosopher wrote:
> On 10/10/2025 02:24, c186282 wrote:
>> On 10/9/25 12:00, John Ames wrote:
>>> On Thu, 9 Oct 2025 12:37:30 +0100
>>> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>>
>>>>> Certainly, but keeping it separate allows for some fine-tuning (e.g.
>>>>> picking a larger inode size for a partition where small files will
>>>>> be the exception rather than the rule.)
>>>> Crikey, how long has THAT been an issue. Like the time you had to
>>>> define how many TCP/IP buffers or virtual ttys you had... Suddenly
>>>> there was enough memory and disk space to not give a shit
>>>
>>> Everybody can feel free to make their own judgements - but in my own
>>> experience, "more than enough disk space" always manages to grind its
>>> way down to "not as much space as I thought" eventually. Parkinson's
>>> Law is an insidious force...
>>
>>
>>    Yep, there's NEVER "enough" space  :-)
>>
> Actually I seem to have arrived at a limit for 'personal' use.
> 
> Running with a couple of terabytes for media and 24GB or RAM I simply 
> have now 'more than enough'
> 
> 
>>    Kind of like building a bigger closet, it too
>>    will be packed to overflowing within a year.
>>
> Then throw those old clothes that no longer fit away. Accept you will 
> never have a 30" waist ever again..
> 
>>    A lot of people work with 8+ K 120fps video now,
>>    that'll fill drive and memory up really quick.

   "Personal" - yea - enough IS enough for all practical
   purposes. I keep a gB and a few extra gB on my cheapo
   'NAS'. More than enough.

   Biz however ....

   Did a lot of backup servers - 50+ employees shit twice
   a day. An amazing amount of redundant shit on their
   disks/parts ... but you had to back up ALL of it Just
   In Case.

   And a few employees were nefarious for somehow
   disappearing almost everything on their machines.
   This included the one who did payroll ......

   Don't remember ... does Win have a /dev/null
   equiv device ? SEEMS like those employees
   did a cut/paste to one of them somehow :-)

[toc] | [prev] | [next] | [standalone]


#75948

FromCharlie Gibbs <cgibbs@kltpzyxm.invalid>
Date2025-10-11 17:48 +0000
Message-ID<ILwGQ.254740$7Ika.58130@fx17.iad>
In reply to#75920
On 2025-10-11, c186282 <c186282@nnada.net> wrote:

>    Don't remember ... does Win have a /dev/null
>    equiv device ? SEEMS like those employees
>    did a cut/paste to one of them somehow :-)

Nah, they just accidentally drop the icon while moving
across the screen toward the icon where they intended
to drop it.  One little slip while trying to keep constant
pressure on that mouse button and >poof<.  And then it's
time to start searching the entire file system.

I don't do drag-and-drop anymore.  It's too dangerous.

-- 
/~\  Charlie Gibbs                  |  Growth for the sake of
\ /  <cgibbs@kltpzyxm.invalid>      |  growth is the ideology
 X   I'm really at ac.dekanfrus     |  of the cancer cell.
/ \  if you read it the right way.  |    -- Edward Abbey

[toc] | [prev] | [next] | [standalone]


#75974

Fromc186282 <c186282@nnada.net>
Date2025-10-11 22:11 -0400
Message-ID<W-WcnRVsBfTMk3b1nZ2dnZfqnPhj4p2d@giganews.com>
In reply to#75948
On 10/11/25 13:48, Charlie Gibbs wrote:
> On 2025-10-11, c186282 <c186282@nnada.net> wrote:
> 
>>     Don't remember ... does Win have a /dev/null
>>     equiv device ? SEEMS like those employees
>>     did a cut/paste to one of them somehow :-)
> 
> Nah, they just accidentally drop the icon while moving
> across the screen toward the icon where they intended
> to drop it.  One little slip while trying to keep constant
> pressure on that mouse button and >poof<.  And then it's
> time to start searching the entire file system.

   Saw that often enough, TOO often. Took to adding
   tag files to the more important user dirs - like
   the payroll stuff ! - so I could do an easy huge
   search for them.

   However sometimes ... the giant dirs just DISAPPEARED.
   Not even in the recycle bin.

> I don't do drag-and-drop anymore.  It's too dangerous.

   I've stuck with deliberate copy/cut-paste. Far harder
   to get it wrong. If space permits I'll just copy, be
   SURE it's at the dest, then go back and delete the
   original. Uses a few extra seconds but you KNOW
   what's happening.

[toc] | [prev] | [next] | [standalone]


#75827

Fromc186282 <c186282@nnada.net>
Date2025-10-09 02:56 -0400
Message-ID<DIednYoSz4A4wXr1nZ2dnZfqnPWdnZ2d@giganews.com>
In reply to#75813
On 10/8/25 17:13, Lawrence D’Oliveiro wrote:
> On Wed, 8 Oct 2025 09:02:09 -0700, John Ames wrote:
> 
>> Having a separate partition with a (vaguely) well-organized
>> directory structure ...
> 
> You can have the latter without the former. Remember, *nix systems
> maintain a single directory hierarchy, regardless of how the storage
> may actually be partitioned into separate volumes.

   But it's not "there" until you MOUNT it.

   Major OS/config updates ... do them and then
   set up remounting the data partitions.

   This is the proper setup for systems you're
   getting paid to maintain.

   Oh, rig periodic backups of the /home and
   /etc files to those 'other' parts. Gives
   you a clear path to getting things working
   again quickly.

[toc] | [prev] | [next] | [standalone]


#75825

Fromc186282 <c186282@nnada.net>
Date2025-10-09 02:50 -0400
Message-ID<DIednYgSz4Dcxnr1nZ2dnZfqnPWdnZ2d@giganews.com>
In reply to#75800
On 10/8/25 12:02, John Ames wrote:
> On Tue, 7 Oct 2025 09:56:08 +0100
> Pancho <Pancho.Jones@protonmail.com> wrote:
> 
>> I'm questioning if it is best to follow that intention, or to instead
>> create my own data area.
> 
> Personally, I've long been in the habit of making a separate partition
> for large-scale local storage (do have a home fileserver, but that only
> gets you anywhere when you're on the home network.) /home is fine for
> config files, one-off working documents, and unsorted downloads, but it
> quickly gets cluttered if I store *everything* there. Having a separate
> partition with a (vaguely) well-organized directory structure for media/
> documents/larger-scale project files makes it a lot easier to find
> stuff later, without having to resort to some kind of indexing/search
> utility.

   As discussed earlier in the thread, what is "best"
   depends on the particular application.

   I have found it useful in the past to put all 'data'
   on a completely different partition. You can then
   do all kinds of damage to the OS/home stuff but not
   risk all the vital data. Re-mount that later.

   For 'home' systems I suppose it's OK to put everything
   in easy reach, but for places where you're getting a
   paycheck, not so much.

[toc] | [prev] | [next] | [standalone]


#75726

FromLawrence D’Oliveiro <ldo@nz.invalid>
Date2025-10-05 20:25 +0000
Message-ID<10buk7s$3o9qk$2@dont-email.me>
In reply to#75716
On Sat, 4 Oct 2025 22:56:22 +0100, Pancho wrote:

> I've never trusted the idea as /home/user being for user controlled
> data, distinct from the installed system. This folder standardly
> contains user preferences/config for the OS and installed applications.

No, the only data that might be put there that the user didn’t explicitly 
ask for would be per-user prefs for apps that they use.

This is why “dotfiles” were implemented, so that these prefs/config files 
would not clutter up normal directory listings. But dotfiles have led to 
their own clutter, which why you now have the XDG spec to limit them to a 
small number of “dot-directories”, and let the apps have their own 
subdirectories within that, where they can put whatever stuff they like.

[toc] | [prev] | [next] | [standalone]


Page 3 of 9 — ← Prev page 1 2 [3] 4 5 6 7 8 9  Next page →

Back to top | Article view | comp.os.linux.misc


csiph-web