On Mon, 7 Jul 2025 01:24:06 +
"abre...@gmail.com wrote:
> Has anyone tried installing StarOffice on a recent version of Linux?
Just gave it a go with SO 5.2 on Debian stable. It failed immediately
because 32-bit and I'm not set up for multiarch. I think the least
painful way to get it running
On Tue, 3 Jun 2025 20:54:39 +
"dbarr...@blazemonger.com" wrote:
> The best approach so far is running a legacy version of Outlook in a
> Windows10 VM to export PST files (making sure to disable "cached
> exchange mode," or the PSTs are created empty), then Linux "readpst"
> to convert to mbox,
On Mon, 2 Jun 2025 07:32:30 -0700
Kent Borg wrote:
> I've tried copying a big offline Thunderbird mailbox to a live
> Dovecot server, and it worked for the first 11,880 messages, but now
> won't do any more. I know Dovecot can handle more because this came
> from an earlier instance of Dovecot.
On Sun, 1 Jun 2025 22:36:06 +
"dbarr...@blazemonger.com" wrote:
> What's the best way to import 600 mbox files to an IMAP server?
The most consistent and reliable, albeit slow, method is to connect
your email client to both servers/backends and copy messages. You might
need to chunk up the co
On Sat, 3 May 2025 00:19:30 +
"kentb...@borg.org [via Relay]" wrote:
> Poking around in the OpenWRT GUI I saw under Network -> Routing
> there is an "IPv4 Rules" tab, in there are
> 4-rules I don't understand, I disabled them. And an IPv6 tab has
> 4-rules. I disabled them too.
Those proba
On Fri, 2 May 2025 21:45:29 +
"kentb...@borg.org [via Relay]" wrote:
> - Turn off all the firewall stuff I can find on the new box. (There
> is a lot of stuff in this new box, I think I am being tripped up by
> something "clever" in there.)
I'm inclined to agree (much as I'm quite pleased wi
That got it. Thank you.
A workaround for anyone else having gmail problems is to use Firefox
Relay.
On Wed, 9 Apr 2025 13:15:30 -0400
Jerry Feldman wrote:
> You are subscribed. Gmail is blocking us because of a dkim issue
--
\m/ (--) \m/
___
Discuss
On Fri, 14 Feb 2025 10:45:19 -0500
Rich Pieri wrote:
> Claws-Mail has started making a bonk sound whenever a toolbar button
> is clicked. I think it's something new/different in GTK but I cannot
> identify what it is. I changed gtk-enable-event-sounds from "=1" to
>
Claws-Mail has started making a bonk sound whenever a toolbar button is
clicked. I think it's something new/different in GTK but I cannot
identify what it is. I changed gtk-enable-event-sounds from "=1" to
"=0" in my .gtkrc-2.0 and relogged but that didn't seem to do anything.
Any clues?
--
\m/
On Wed, 29 Jan 2025 13:52:10 -0800
Kent Borg wrote:
> What is the best way to manage this going forward? I think I would
> like to find out about any changes to the official Debian .deb files,
> so I can download new sources and reapply this patch, if necessary.
If this patch is incorporated ups
I do use Ansible. I wouldn't say extensively, but definitely for
deploying standard configurations. That's the key point: standard
configurations. When everything you deploy is a one-off, bespoke
system, tools like Ansible can be more work than just doing the work.
For example, to add a user to a
On Fri, 10 Jan 2025 14:03:23 -0800
Kent Borg wrote:
> How do I get postfix to append some mail to a file? In the past I
> have put in /etc/aliases something like:
>
> some-string: /path/to/file
I would send it through procmail and let that handle delivery.
--
\m/ (--) \m/
There are known problems with native messaging and flatpak sandboxes, and I
found this workaround:
https://github.com/browserpass/browserpass-native/issues/93#issuecomment-1646566107
It works for me on Tumbleweed with browserpass installed with zypper
from the SUSE repositories. I cannot get it w
Kernel 6.12 as shipped with Fedora and Tumbleweed has the KVM modules
enabled by default. This will conflict with other hypervisors.
I'm hoping the next releases revert this change but I don't know if
they will.
--
\m/ (--) \m/
___
Discuss mailing list
On Fri, 13 Dec 2024 15:07:00 -0500
Jerry Feldman wrote:
> For the laptop, I'm considering either back in time or another
> backup. And Dropbox is installed on my laptop. Unlike the tower, the
> laptop will be turned off frequently, and also the backup device will
> be an SSD. So, is like a backup
On Wed, 11 Dec 2024 18:38:19 -0500
Rich Pieri wrote:
> Fair. I used to have a patch for Emacs 19.34 (last version before the
> disastrous merge of MULE) for compiling on GLIBC2 machines. I'll see
> if I can dig it up, might be useful for future reference.
Found it. Doesn'
On Wed, 11 Dec 2024 11:04:31 -0500
John Abreau wrote:
> I gave that a try, but the compilation fails with some obscure error,
> and I'm not motivated to invest a significant amount of time
> debugging a codebase I'm unfamiliar with and that compiled just fine
> on older versions of Fedora.
Fair.
On Tue, 10 Dec 2024 14:20:48 -0500
John Abreau wrote:
> After having no luck finding an repos for Fedora 39, I dug through my
> backups, and I found the rpms I had used earlier this year, and
> completed my new system build.
I have to wonder why you don't compile from source.
--
\m/ (--) \m/
_
On Sun, 1 Dec 2024 08:32:31 -0800
Kent Borg wrote:
> I immediately wonder whether I want two of them. But I like using
> such beasts not just on the open internet, but internally, to fill in
> wifi gaps, and no built in ethernet switch is a pain.
In retrospect, I can see the lack of integrated s
A week after I got the portable firewall, I discover that Netgear EOL'd
my old router. And since I'm not leaving an unsupportable device facing
the public Internet, I replaced it with a GL.iNet home router. Because
despite their custom browser UI not being fully open source, everything
underneath i
A tangent to my wireguard fun, my home router went out of support
recently and needed to be upgraded, so I went with another Gl.iNet box
because I'm willing to pay for something less than 5 years old and
they've done the heavy lifting getting OpenWRT on it.
Since then I've been seeing entries like
On Tue, 19 Nov 2024 18:18:53 -0500
Kent Borg wrote:
> P.S. Is this something I can blame systemd for? No? Well, it was
> worth the try.
No, for a change. ACPI is a bloated mess of specifications that that
nobody, not one hardware vendor in the world, has ever correctly
implemented.
--
\m/ (--)
On Thu, 14 Nov 2024 21:45:42 -0500
Kent Borg wrote:
> On 11/14/24 9:35 PM, Rich Pieri wrote:
> > The only kinda-problem with Brother is […]
>
> What about software compatibility with Linux?
Seem to be well-supported by CUPS and OpenPrinting.
On Thu, 14 Nov 2024 20:39:00 -0500
Kent Borg wrote:
> I see Microcenter in Cambridge has various inexpensive Brother models…
The only kinda-problem with Brother is their toner cartridge sensors
are calibrated to "require" you to buy new cartridges while there is
plenty left. I put require in quo
On Mon, 11 Nov 2024 13:05:02 -0500
dan moylan wrote:
> i just upgraded my three laptops from fc40 to fc41. none
> of them now mounts my SSD harddrive. df now shows all these
> /run/credentials files which i never saw before. what's
https://systemd.io/CREDENTIALS/
> going on? how can i moun
On Sat, 2 Nov 2024 15:44:07 -0700
Kent Borg wrote:
> Except last I looked a lot of hardware doesn't run a current version
> of OpenWRT.
Did you look at the OpenWRT web site/table of hardware?
--
\m/ (--) \m/
___
Discuss mailing list
Discuss@lists.blu
On Sat, 2 Nov 2024 10:53:25 -0700
Kent Borg wrote:
> But the GL people seem to ship OpenWRT on all their hardware. That
> suggests that there will be longer support for this hardware.
GL ship a fork of OpenWRT, but the fork is their own UI on top of
vanilla. Full LuCI is available underneath, a
Why hack a solution when I can throw some money at it?
I've long held the belief that a portable/personal firewall device
would solve many problems. I just never dug into it. Until now: I
bought a GL.iNet travel router. Pocket-sized device, USB-C power (15W),
runs OpenWRT, bridges to existing WiFi
On Tue, 29 Oct 2024 19:31:13 -0400
dan moylan wrote:
> sorry, i forgot to mention at the outset i disabled selinux.
Double-check that to be sure nothing has quietly re-enabled SELinux.
Otherwise, the directory being written needs to be writable by the UID
or GID doing the writing, and all paren
On Wed, 23 Oct 2024 12:07:58 -0400
dan moylan wrote:
> root ~[232] mysql -u root
> ERROR 2002 (HY000): Can't connect to local server through
>socket '/var/lib/mysql/mysql.sock' (2)
>
> suggestions?
You probably need to enable and start the service:
systemctl enable --now ma
On Mon, 21 Oct 2024 11:46:09 -0700
Kent Borg wrote:
> Just so long as you don't need access to the local version of those
> addresses. (The hotel's gateway to get to the rest of the internet?
> DNS? Room service?)
Exactly. In my case my home network address space is local to the hotel
network a
On Sun, 20 Oct 2024 18:42:53 -0400
Dan Ritter wrote:
> There's always something, and in this case, you can do something
> really weird.
Yup. That is really weird. I'll have to look up network namespaces when
I get home.
Thank you.
--
\m/ (--) \m/
__
I'm traveling a bit this weekend and I ran into some network wonk with
my Wireguard VPN: My home network is 192.168.1.0/24. The place I'm
staying uses 192.168.0.0/20 for their WiFi network. Because my home
network overlaps their network, traffic to my home network doesn't go
out the Wireguard inter
On Sat, 12 Oct 2024 20:59:16 +0200 (CEST)
Zachary Liebl wrote:
> Do you think is suspicious, and even if not, do you think this is a
> GPL violation?
Sus? Yes, though the notes by Toolybird do help ease some of the
suspicion. Still... I'm not going to be using Ventoy for the
foreseeable future w
https://github.com/openzfs/zfs/issues/11140
Briefly: Flatpak uses fsync extensively. For reasons nobody is entirely
sure about, this causes ZFS to thrash *hard* which I discovered earlier
this week as I finally got around to switching my Calibre install from
Debian's packages (which are woefully o
On Wed, 18 Sep 2024 12:38:22 -0400
Rich Pieri wrote:
> ReaR is very configurable. To exclude /data/excluded you add this line
> to your /etc/rear/local.conf:
>
> EXCLUDE_BACKUP=( ${EXCLUDE_BACKUP[@]} /data/excluded )
Possible correction. The setting might be BACKUP_
On Wed, 18 Sep 2024 12:31:00 -0400
Daniel Barrett wrote:
> Thanks for sharing your setup, Rich. I'm curious: how protected are
> you if machine A's disk dies in the middle of a ReaR backup,
> potentially corrupting the backup? How would you recover?
I keep two versions of the ReaR backups on eac
On Wed, 18 Sep 2024 11:47:33 -0400
Steve Litt wrote:
> I'll be adding this to my backup routine, if I can set it to only back
> up the system stuff, leaving the data to be backed up by rsync.
ReaR is very configurable. To exclude /data/excluded you add this line
to your /etc/rear/local.conf:
EX
The deduplication discussion got me thinking about how I do backups of
my two "work" stations (read: where I play games, watch movies, and do
all my other not actually work things like writing this). I had been
using rsync to replicate data to my file server (ZFS, redundant
storage, etc.) but this
On Thu, 5 Sep 2024 09:32:09 -0400
Rich Pieri wrote:
> Aside: The ext# family don't have CoW capability so dupremove can't
> work on them.
Aside 2: Compression is typically a much better value than dedup at the
small end.
--
\m/ (--) \m/
___
I think deduplication is kind of overrated and impractical. As was
pointed out several times in the EFI thread: big, fast drives are
cheap. So what if there are two or three copies of a file on a backup
set? The dedup overhead is more costly than the storage.
Where deduplication starts becoming pr
On Tue, 3 Sep 2024 22:05:59 -0400
Steve Litt wrote:
> Tell me more about rEFInd. This sounds interesting.
rEFInd is a fork of rEFIt. They're UEFI programs -- why is everything
an OS? So that we can have nice things like rEFIt and rEFInd. rEFIt
originally was intended for UEFI implementations lik
On Tue, 3 Sep 2024 19:51:43 -0400
Steve Litt wrote:
> >It also brings cross-architecture portability;
>
> I'm not sure in what way it does this, but I'm sure it could have been
> done in a much simpler way.
The principle being that your UEFI code will run on ARM, AMD64, IA64,
or whatever oth
On Tue, 3 Sep 2024 13:26:45 -0700
Kent Borg wrote:
> When I first poked my head down into EFI I was horrified: A whole
> damn OS down there. (Why does everything need to be an OS‽‽)
A number, in fact a litany of reasons.
Starting with securing the boot process against attack.
It also brings c
"Future Possibilities: Data, Hardware, Software, and People" presented
by Capt. Grace Hopper at NSA in 1982. Two parts:
https://www.youtube.com/playlist?list=PLWvkfYC3snpW3GYOXoBYcmReXN2Blc9T8
--
\m/ (--) \m/
___
Discuss mailing list
Discuss@lists.blu.
On Wed, 28 Aug 2024 19:26:03 -0400
jbk wrote:
> Could be, but you still have to edit the installed copy for
> UUID's, IP addresses and hostname. And, doing the snapshot
> probably takes the same amount of time as what I did to copy
> the files directly to disk. The advantage to my method is
>
On Wed, 28 Aug 2024 13:52:34 -0400
"Dale R. Worley" wrote:
> I'm told ZFS is popular and supports copy-on-write, but it adds
> another layer of volume management, so I chose XFS as the path with
> lowest learning curve.
I say exactly the opposite. Best practice is you give ZFS the entire
device
On Wed, 28 Aug 2024 07:23:52 -0400
jbk wrote:
> I guess I'm lazy in that regard. There is Ansible that RH
> advocates in doing just this as you describe. On all my
I'm very much an Ansible advocate but it's too much for this specific
purpose: more time spent crafting and testing Ansible plays
My take on the process is rather mundane: I write it down. Every
package I install and every custom change I make, I write it in text
file which functions as a play book for deploying a functionally
identical system regardless of the underlying hardware. And then I
restore my home directory and non
On Wed, 21 Aug 2024 08:58:21 -0700
Kent Borg wrote:
> The things I am worried about are:
>
> - Hardware compatibility. If Debian works (I'm thinking it does), how
> likely is Devuan?
Should be the same.
> - Navigating the installer. Putting btrfs on top of encrypted LVM
> doesn't seem to inv
On Tue, 20 Aug 2024 18:56:48 -0400
Dan Ritter wrote:
> Sounds like you might want Debian with a different init. You
> could do that on Debian with a little tinkering, or you could use
> Devuan.
As a full time user of the former method, I can unequivocally state
that sysvinit is not a first class
On Tue, 20 Aug 2024 11:50:52 -0700
Kent Borg wrote:
> Debian user might like better in this year 2024?
Debian is a perfectly good server operating system but it's not a great
desktop foundation. OpenSUSE Leap is good. It is to SuSE Linux
Enterprise what CentOS used to be to RHEL so you have that
On Mon, 19 Aug 2024 08:22:29 -0400
Daniel Barrett wrote:
> I jut came across a fantastic site for testing SPF, DKIM, and DMARC
> records for email delivery:
>
> https://www.learndmarc.com/
I'll also note that Microsoft have a very useful suite of mail analysis
tools:
https://testconnectivity
On Mon, 26 Feb 2024 14:24:28 -0500
Bill Bogstad wrote:
> If I look into the /tmp/.X11 directory I can see multiple socket files
> which appear to correspond to the $DISPLAY variable that X clients
> use to find their server. So if you subscribe to the idea that
> /dev/tty is an example of 'ever
On Mon, 26 Feb 2024 01:30:55 -0500
John Hall wrote:
> Ok So this is odd. I figured I'd look. dd is cpu bound ! WEIRD!
> 104909 root 20 06628 1248756 R 73.2 0.0 228:02.10
> mount.ntfs
> 105003 root 20 05540 96 0 S 32.5 0.0
> 101:36.29 dd And mount.ntfs
On Wed, 21 Feb 2024 14:00:22 +
Edward wrote:
> A part of Liam Proven's FOSDEM 2024 talk.
>
> https://www.theregister.com/2024/02/21/successor_to_unix_plan_9/
Plan 9 as an operating system died, but some very useful bits from Plan
9's development are in active use today, most notably the 9P
On Tue, 20 Feb 2024 14:00:10 -0500
Kent Borg wrote:
> P.S. Does X do things as files? I've only ever been a user,
No. X11 is a display server and a network protocol stack. It is OS and
architecture agnostic. X11 clients might do things as files, or they
might not, but this is entirely separate f
On Mon, 29 Jan 2024 21:26:05 -0500
Dan Ritter wrote:
> > Yes, but this isn't the question. Let's say the exports are
>
> Yes, but it was the answer.
It's an answer to a question I did not ask. I already know how NFS v4
works with nested volume exports. What I don't know, and what I asked
abou
On Mon, 29 Jan 2024 20:01:25 -0500
Brandon Vogel wrote:
> Once you see them all mounted in Dolphin, do you see them as separate
> mounts to the system when you run the 'mount' command?
Yes, they are separate mounts according to the mount and df commands,
and they are in /etc/mtab. autofs is not
On Mon, 29 Jan 2024 18:22:41 -0500
Dan Ritter wrote:
> If all clients should see foo/quz and foo/rat, client B should see
> foo/bar and client C should see foo/baz, I think you will have to
> rearrange this so that you export foo/quz and foo/rat to all and
> foo/bar and foo/baz separately.
Yes, b
I have a server, ZFS on Linux, using ZFS' NFS server to export some
nested datasets:
tank/foo
tank/foo/bar
tank/foo/baz
zfs set sharenfs="rw,client.home.net" tank/foo
zfs share tank/foo
This exports three NFS volumes because that's how ZFS works. This is
fine.
I mount the top level tank/foo on
On Fri, 12 Jan 2024 12:33:06 -0500
Daniel Barrett wrote:
> Also, what's the business need behind your question? In case there's a
> better approach than "finding an image viewer that can open ZIP
> files."
I want an image viewer that can open zip files and similar archives.
Nothing more than thi
I'm looking for an image viewer which can open ZIP (ie .cbz) files but
is *not* a comic reader or library manager. Just an image viewer which
can open archives.
Any suggestions?
--
\m/ (--) \m/
___
Discuss mailing list
Discuss@lists.blu.org
http://list
On Mon, 8 Jan 2024 22:45:54 -0500
Bill Ricker wrote:
> This depends upon what Window Manager and perhaps Desktop Manager
> you're running ?
This, and as noted the X11 vs. Wayland consideration. And the fact that
this all assumes the pty is associated with an xterm (or whatever) and
not something
Quick updates on this front
On Thu, 14 Dec 2023 17:06:03 -0500
Rich Pieri wrote:
> My Windows 10 machines are going to be rebuilt with openSUSE
> Tumbleweed early next year.
Converted the secondary (and more work) PC to Tumbleweed this week.
It's a notebook with nVidia/Optimus so
On Thu, 4 Jan 2024 21:04:19 -0500
Daniel M Gessel wrote:
> Doesn't MS support Linux for Windows now? Do they have an official
> distro, or do you have to do a third party install (e.g. Ubuntu or
> "vanilla" Debian)?
Windows Subsystem for Linux is a (well, two, technically) ways of
installing a
On Thu, 4 Jan 2024 11:19:30 -0500
Sid Koul wrote:
> For those of us not in the know, how did you do a search and replace
> on a Windows box?
Powershell:
(Get-Content "path\to\file").Replace('old', 'new') | Set-Content "path\to\file"
--
\m/ (--) \m/
_
On Tue, 2 Jan 2024 19:43:04 -0500
Dan Ritter wrote:
> - a reference to a CSS file on another machine was changed from
> http to https without checking that it's actually available that
> way
Or that the CSS files are on the same machine but in a different
location that is not being SSL-wrapped.
On Fri, 29 Dec 2023 13:00:03 -0500
"Derek Atkins" wrote:
> I would remove gstreamer1-plugins-ugly and
> gstreamer1-plugins-ugly-free, re-run the upgrade, and then attempt to
> re-install (one of) these packages.
> -derek
Agreed.
My experience with this problem is RHEL and non-RHEL packages, but
On Fri, 22 Dec 2023 16:48:09 -0500
Derek Martin wrote:
> I came across this, and I thought it was interesting:
>
>
> https://www.tomshardware.com/pc-components/cpus/risc-v-handheld-gaming-system-announced-linux-as-the-basis-for-a-retro-gaming-platform
Amusing that the image credit isn't retr
Assuming that the drives are good, you might be hitting the standby
problem. So-called "green drives" park their heads and go into standby
mode very quickly (a set of Western Digital Green series drives I have
park after 8 seconds idle). It can take a long time (tens of seconds)
for them to spin up
On Fri, 15 Dec 2023 08:36:10 -0800
Kent Borg wrote:
> Network Solutions wants me to confirm data on a domain I control.
> They sent me a link, but to some site I don't have any record of a
> password to. I'm not going to just start trying stuff, that's what
> phishers want us to do. Can I just ig
On Thu, 14 Dec 2023 14:35:55 -0500
Daniel M Gessel wrote:
> I chose to switch from Ubuntu because of Snap and now use Debian 12
> (Bookworm) with Sway WM for Wayland. It's pretty much my dream
> config! If I had stuck it out longer, I would definitely be switching
> from 23.10...
I use and admi
I learned this week that Canonical are forcing Snap packages on
everyone with Ubuntu 23.10. This apparently is how Canonical are
letting users know about their new Snap store.
https://www.webpronews.com/ubuntu-23-10s-app-store-will-block-deb-files-when-a-snap-is-available/
Choice quote from Canon
On Mon, 27 Nov 2023 09:55:04 -0800
Kent Borg wrote:
> > and that any attempt to read the raw text of the email had been
> > blocked in Thunderbird.
> They manage to disable "View"->"Message Source Ctrl+U"? That is
> impressive.
If they buried the whole thing in the PDF file then there is no
On Thu, 2 Nov 2023 14:34:06 -0400
Kent Borg wrote:
> What is the best way to manage this? I don't want to accidentally
> install the standard kernel on top of my custom kernel, but I would
> like to be prodded to compile a new kernel by the availability of a
> new kernel .deb.
Preventing new ke
On Sun, 8 Oct 2023 03:01:48 -0400
John Abreau wrote:
> I was checking server prices on ebay this evening, and I found
> something nice that has an Intel Xeon E5-2690 V2 cpu, but I'm having
> no luck finding a way to determine if this cpu is "x86-64-v2" as
> opposed to just "x86-64".
https://deve
On Tue, 22 Aug 2023 13:35:15 -0400
Ben Kallus wrote:
> Has anyone here dealt with this problem in the past? Is anyone aware
> of an adapter with VGA input and digital output that supports
> 1920x1200?
Yes; and they're all garbage.
--
\m/ (--) \m/
___
On Sat, 24 Jun 2023 19:37:26 -0400
grg wrote:
> even if only one is used at a time, the system still has two paths to
> the resource (which I'd argue is twice as many as a clean design
> ought to have...)
An argument that I would agree with in principle, but you have to walk
and merge /bin and /
On Sat, 24 Jun 2023 10:14:01 -0400
grg wrote:
> I agree usrmerge is a good thing, but I'd say it's actually adding
> bloat rather than trimming it: post-usrmerge there are (at least) two
> paths for every binary, two linkings for every library. pre-usrmerge
> there was usually just one on a given
On Fri, 23 Jun 2023 18:49:40 -0400
Daniel M Gessel wrote:
> As a hobby software developer, I see the benefit: *nix isn't static,
> so simplification is generally "a good thing". More power to those
> who use Occam's razor to trim some bloat.
One of the more amusing things about this for me is th
On Thu, 22 Jun 2023 20:07:38 -0500
Derek Martin wrote:
> I'm not sure what nonsense you mean--I'm still waiting for you to
> refute a single point I made with actual facts, but I'm glad you're
> enjoying yourself. :)
--%< cut here %<--
#!/bin/sh
# Little bit of code to identify
On Thu, 22 Jun 2023 17:23:38 -0500
Derek Martin wrote:
> But if you're not the admin in your environment and yours stubbornly
> refuses to do that for whatever reason, you are SoL. You'll probably
> have no choice but to use env, or I s'pose maybe create multiple
> copies... in which case they'l
On Thu, 22 Jun 2023 15:07:19 -0500
Derek Martin wrote:
> 1. As I indicated in the other message, if the program is intended to
>run exclusively in the security context of the user running it, and
>does not at any point require elevated privileges (which needs to
>be evaluated carefull
On Thu, 22 Jun 2023 17:05:47 -0400
Daniel M Gessel wrote:
> With that in mind, is the burden mostly on WSL users having issues
> upgrading to Debian 12? I can see that cost being discounted by
> hardcore *nix devotees...
It's the only one widespread enough for the Debian maintainers to
specific
On Wed, 21 Jun 2023 19:26:55 -0400
Dan Ritter wrote:
> If you're packaging for a particular distro, you know where they
> put it.
For systems that implement merged /usr you just use /usr/bin/perl. For
systems with optional merged /usr then your post-install scripts will
need to adjust the interp
On Wed, 21 Jun 2023 11:35:08 -0500
Derek Martin wrote:
> I think it's worth expanding on this just a bit. This IS good
> practice, and you should do it in your shell scripts--particularly
> when you need to execute system utilities but can't be sure in which
> system path they will live--and you
On Tue, 20 Jun 2023 14:08:11 -0500
Derek Martin wrote:
> Obviously I disagree, particularly when the customizations are
> basically trivial, one-time work.
That's the rub. You see it as a one-time thing. Whereas I see it as a
one-time thing times the approximately 4000 machines and counting that
On Sun, 18 Jun 2023 16:57:30 -0400
dan moylan wrote:
> moylan ~[1140] ll -d .ssh
> drwx--. 1 moylan moylan 104 230618:1645 .ssh/
> moylan ~[1141] ll .ssh
> total 16K
> -rw---. 1 moylan moylan 553 230612:1636 authorized_keys
> -rw---. 1 moylan moylan 1.3K 230615:1316 known_hosts
> -rw
How do you all go about accessing host filesystems from KVM guests?
KVM is running as user libvirt-qemu, not as root.
--
\m/ (--) \m/
___
Discuss mailing list
Discuss@lists.blu.org
http://lists.blu.org/mailman/listinfo/discuss
On Sun, 18 Jun 2023 16:06:19 -0400
dan moylan wrote:
> moylan cmd[449] rsync -r -e ssh -i /home/moylan/.ssh/rsync.key
> /home/moylan/foo moylan@aldeberan:
>
> asks for a password. i'm doing something wrong, but haven't
> a clue as to how to find out what. any suggestions?
Check permissi
On Fri, 16 Jun 2023 19:07:19 -0500
Derek Martin wrote:
> No. A symlink solves that problem if it's a concern in your
> environment--it never has been in any of mine, even with a mix of
> SunOS, Solaris, HP-UX, and Linux machines. And this is not actually
Adding custom symlinks to areas owned by
On Fri, 16 Jun 2023 17:54:25 -0500
Derek Martin wrote:
> I do not see one. "It is no longer necessary to do it this way," is
> not one. Change for no practical reason is bad change.
I would say, rather, "we've always done it this way" is the worst
reason possible for doing something a particu
On Fri, 16 Jun 2023 16:41:21 -0500
Derek Martin wrote:
> I'm curious if this change is thought to have any genuine practical
> benefit, or if it's just the usual, "I'm a bored developer, time to
> break something completely arbitrarily, that's working perfectly fine,
> that people have been used
On Fri, 16 Jun 2023 19:12:22 +
Edward wrote:
> Should the average end-user be concerned about this change to /usr?
> The internal HDD/SSD installs are Debian Sid (unstable). originally
> installed with a Debian 11 image, then 'upgraded'.
No. The process creates symlinks for formerly split bi
Debian 12 was released last week(end). With it comes the mandatory
change to merged /usr.
https://wiki.debian.org/UsrMerge
This can break on WSL1 installs due to how WSL locks parts of the
filesystem. Debian's official recommendation is to reinstall the base
OS if you don't want to convert to WSL
I'm trying to upgrade my Pi VPN box from an old Pi 2/32-bit to a new Pi
4/64-bit using PIVPN and Wireguard. I followed the backup, new setup,
and restore instructions. My endpoints connect to the new server and I
see traffic (pivpn -c) but all DNS lookups fail. This is setting either
my internal DN
On Fri, 9 Jun 2023 10:15:41 -0400
dan moylan wrote:
> just where are the authentication logs? i tried journalctl and
> didn't see anything appropriate. looking at .fetchmailrc
Depends on your mail server. Dovecot on Debian logs session
authentication in /var/log/mail.log by default but this is
On Thu, 8 Jun 2023 11:17:49 -0400
dan moylan wrote:
> trying fetchmail again after some years, but getting authentication
> failure.
Possibly not helpful here but I switched from fetchmail to getmail
around 5 years ago. Easier setup than fetchmail, doesn't require a local
MTA (but can do things
Just some notable notes from my experience the past week getting
Nextcloud working the way I want.
I still hate Docker. The more I try to do simple things with it, the
more ways I see how terrible the core design is. Yes, I get it that it
scales out very well, but due to early design decisions mad
1 - 100 of 261 matches
Mail list logo
