From: "Ludovic Courtès" <ludo@gnu.org>
To: Pierre Neidhardt <mail@ambrevar.xyz>
Cc: 39734@debbugs.gnu.org
Subject: [bug#39734] [PATCH] scripts: Emit GC hint if free space is lower than absolute and relative threshold.
Date: Sun, 23 Feb 2020 17:09:11 +0100 [thread overview]
Message-ID: <87tv3hs2u0.fsf@gnu.org> (raw)
In-Reply-To: <87pne5e581.fsf@ambrevar.xyz> (Pierre Neidhardt's message of "Sun, 23 Feb 2020 15:41:18 +0100")
Hi!
Pierre Neidhardt <mail@ambrevar.xyz> skribis:
> See the original discussion:
>
> https://lists.gnu.org/archive/html/guix-devel/2020-02/msg00303.html
Oops, I’m lagging behind and had not seen it yet, sorry about that!
> Our users fall into 2 categories:
>
> - Those with a "small" store partition, say 25 GiB.
> - Those with a "big" store partition, say 1 TiB.
>
> If we use a relative threshold, users with a big partition will get
> a hint all the time. This may mislead newcomers into thinking that they
> should run "guix gc" while they still have 50 GiB of free space.
>
> If we use an absolute threshold, users with a small partition will get a
> hint all the time.
>
> In both cases, the hints becomes irrelevant if it's triggered for the
> wrong reason.
>
> The fix is to use both thresholds and to check against the lowest one.
>
> Does that make more sense?
It does, yes.
So back to the original patch… I think we should use ‘size->number’ for
the absolute threshold.
I’m also still mildly reluctant to the addition of
‘GUIX_DISK_SPACE_WARNING_ABSOLUTE’ (one more variable and a weird
name—“absolute warning”?). An option could be to allow
‘GUIX_DISK_SPACE_WARNING’ to optionally contain both thresholds. These
would be valid values:
1GiB;10%
15 ;15% relative, and default absolute
15% ;same
1G ;1G absolute, and default relative
Thus ‘%disk-space-warning’ would hold a pair.
Does that make sense?
Apologies for the misunderstanding!
Ludo’.
next prev parent reply other threads:[~2020-02-23 16:10 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-22 10:46 [bug#39734] [PATCH] scripts: Emit GC hint if free space is lower than absolute and relative threshold Pierre Neidhardt
2020-02-22 11:12 ` Ludovic Courtès
2020-02-22 11:39 ` Pierre Neidhardt
2020-02-23 11:46 ` Ludovic Courtès
2020-02-23 11:49 ` Pierre Neidhardt
2020-02-23 14:34 ` Ludovic Courtès
2020-02-23 14:41 ` Pierre Neidhardt
2020-02-23 16:09 ` Ludovic Courtès [this message]
2020-02-23 16:36 ` Pierre Neidhardt
2020-02-23 21:05 ` Pierre Neidhardt
2020-02-23 21:11 ` Pierre Neidhardt
2020-02-24 10:21 ` Ludovic Courtès
2020-02-24 10:26 ` Pierre Neidhardt
2020-02-24 13:20 ` Pierre Neidhardt
2020-02-24 21:08 ` Ludovic Courtès
2020-02-25 10:22 ` Pierre Neidhardt
2020-02-25 10:23 ` Pierre Neidhardt
2020-02-27 23:19 ` Ludovic Courtès
2020-02-28 7:17 ` Pierre Neidhardt
2020-03-05 16:29 ` Ludovic Courtès
2020-03-06 7:32 ` Pierre Neidhardt
2020-02-26 21:06 ` bug#39734: " Ludovic Courtès
2020-02-24 13:20 ` [bug#39734] " Pierre Neidhardt
2020-02-23 21:40 ` Ludovic Courtès
2020-02-24 7:51 ` Pierre Neidhardt
2020-02-24 7:52 ` Pierre Neidhardt
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87tv3hs2u0.fsf@gnu.org \
--to=ludo@gnu.org \
--cc=39734@debbugs.gnu.org \
--cc=mail@ambrevar.xyz \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this external index
https://git.savannah.gnu.org/cgit/guix.git
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.