Re: [gentoo-user] systemd unit executing but not persistent later in boot

[Andrew Udvare]

On Sat, 6 Apr 2024 at 21:32, Daniel Frey  wrote:
> What's even stranger is if I manually start it and restart lircd it works!

Have you tried adding Before=lircd.service in the [Unit] section?

Re: [gentoo-user] KWallet refuses to auto open at login

[Andrew Udvare]

Does a new user account work the way it is supposed to?

On Sat, 10 Jun 2023, 07:33 Victor Ivanov,  wrote:

> Hello fellow penguins,
>
> I have to admit I'm at my wits' end with KWallet. This thing has been
> driving me insane for the last couple of weeks, roughly since the
> upgrade to Plasma 5.27 or shortly after.
>
> Every time I log in, it refuses to automatically open and prompts for
> a password whenever an application wants to read secrets. Admittedly,
> this is under Wayland which, following from recent news re this being
> the Gentoo preference, I decided to give a try. But this also happens
> under X11, so I doubt it's got anything to do with Wayland.
>
> I've tried everything I can think of:
> - I've double checked the auto unlock guide in the Gentoo Wiki, made
> sure PAM rules are in place and kwallet-pam is installed. Which should
> be all good regardless, as it used to work just fine;
> - I've also oneshot all of: kde-plasma/ , kde-frameworks/ , kde-apps/
> , and anything returned by "eix -I# pam", "eix -I# xdg", and "eix -I#
> dbus"
> - when the above failed to yield any meaningful resolution, I repeated
> the one-shot step this time with "--noconfmem" and re-reviewed any
> changes from the default configs;
> - finally, I nuked all of my $HOME settings under ~/.config, ~/.cache,
> and ~/.local and started from scratch;
> - and yes, the KWallet was set up and re-setup (plenty of times) with
> Blowfish with the same password as my login;
>
> The irony is, I have an equivalent Gentoo setup on a separate machine,
> also moved to Wayland at the same time as I keep them both up to date
> at the same intervals, and it works flawlessly. I've recursively
> diff'd all config files under /etc between the two hosts and, other
> than some minor, unrelated host-specific differences, everything is
> identical, including configs under /etc/pam.d.
>
> I can get it to kind of work with a blank password, but that's not the
> point and is not a viable "solution". Even then, it still doesn't auto
> open, but at least it doesn't produce annoying prompts to open.
>
> I've always despised KWallet for its flaky behaviour but for the last
> few years it hadn't given me any issues up until now.
>
> I'm truly bewildered. Is there anything I am missing?
>
> Best Regards,
> Victor
>
>

Re: [gentoo-user] InfluxDB python library?

[Andrew Udvare]

On Wed, 10 May 2023, 03:51 J. Roeleveld,  wrote:

> Hi all,
>
> I've been using a python script for the past 4 years to read out my solar
> panels and write the data into InfluxDB to get nice graphs in Grafana.
>
> Unfortunately, the library I use to write to InfluxDB has a dead upstream
> and
> doesn't work with Python 3.11 anymore.
> Does anyone know of a different way to load the data into InfluxDB using
> Python
> that is likely to be supported for the foreseeable future?
>

The line protocol is pretty simple. It is trivial to write a serialiser for
it.

https://docs.influxdata.com/influxdb/v2.7/write-data/developer-tools/api/

If you can't get a good nanosecond timestamp you can reduce it to second
precision.

Re: [gentoo-user] glabel - compatible thermal label printer

[Andrew Udvare]

On 26/07/2022 16:06, the...@sys-concept.com wrote:

Is anybody using "glabel"?
I'm looking for compatible thermal label printer that will works on 
Linux; and easy to install printer driver.






My overlay provides the latest Qt version. 
https://github.com/Tatsh/tatsh-overlay/tree/master/app-office/glabels


This really depends on what you want to do. Most templates in gLabels 
are for printing on 8.5x11 or A4, but there are a number of Dymo 
templates as well. Dymo does officially support Linux and Gentoo has 
net-print/dymo-cups-drivers.


OpenPGP_signature
Description: OpenPGP digital signature

Re: [gentoo-user] mediatek mt7921: no AP?

[Andrew Udvare]

I'm pretty sure none of these built-in wi-fis like that will support AP
mode. They would much rather charge more for that feature.
https://www.524wifi.com/index.php/network-modules-adapters/wifi-6-11-ax-modules/wle3000h2-11ax-4x4-mu-mimo-2-4ghz-qcn-9024-wifi-6-802-11ax-2-4-ghz-single-band-mini-pcie-module-compex.html

https://forum.openwrt.org/t/802-11ax-wifi-ap-mpci-e-cards/63577/2

I own a similar motherboard but I don't use the wi-fi.

On Sun, Jan 2, 2022, 16:25 Jorge Almeida  wrote:

> I just mounted a box with a  ROG STRIX X570-E GAMING WIFI II
> motherboard. I'm not a gamer; one of the reasons I chose this was the
> build-in WiFi. Bad idea! The thing does not support AP mode, which I
> need. Nothing in the mo manual suggests such a limitation.
> Here are the details, in case someone with the same hw has  a better
> experience and would share it. Otherwise, it may be useful as a
> warning. I know I will not purchase ASUS and/or Mediatek again.
>
> Jorge Almeida
>
> $ lshw
> *-network
>   description: Wireless interface
>   product: MEDIATEK Corp.
>   vendor: MEDIATEK Corp.
>   physical id: 0
>   bus info: pci@:03:00.0
>   logical name: wnet2
>   version: 00
>   serial: b4:b5:b6:93:8d:7f
>   width: 64 bits
>   clock: 33MHz
>   capabilities: pciexpress msi pm bus_master
> cap_list ethernet physical wireless
>   configuration: broadcast=yes driver=mt7921e
> driverversion=5.15.12 firmware=01-20211014150922
> ip=192.168.2.131 latency=0 link=no multicast=yes wireless=IEEE 802.11
>   resources: irq:99 memory:e030-e03f
> memory:e040-e0403fff memory:e0404000-e0404fff
>
> $ lsmod
> xt_MASQUERADE  12288  1
> iptable_nat12288  1
> nf_nat 36864  2 iptable_nat,xt_MASQUERADE
> xt_connmark12288  4
> xt_helper  12288  110
> xt_limit   12288  26
> mt7921e77824  0
> mt76_connac_lib28672  1 mt7921e
> mt76   53248  2 mt7921e,mt76_connac_lib
> mac80211  503808  3 mt76,mt7921e,mt76_connac_lib
> cfg80211  372736  4 mt76,mt7921e,mac80211,mt76_connac_lib
> igb   196608  0
> rfkill 24576  1 cfg80211
> i2c_piix4  20480  0
>
> $ iw dev
> phy#0
> Interface wnet2
> ifindex 6
> wdev 0x1
> addr b4:b5:b6:93:8d:7f
> type managed
> txpower 3.00 dBm
> multicast TXQ:
> qsz-byt qsz-pkt flows   drops   marks
> overlmt hashcol tx-bytes tx-packets
> 0   0   0   0   0   0
>  0   0   0
>
> $ iw list | grep "Supported interface modes" -A 8
> Supported interface modes:
>  * managed
>  * monitor
> Band 1:
> Capabilities: 0x9ff
> RX LDPC
> HT20/HT40
> SM Power Save disabled
> RX Greenfield
>
>

Re: [gentoo-user] motherboard compatibility...

[Andrew Udvare]

On 16/11/2021 17:32, Jorge Almeida wrote:

ROG Strix X570-E Gaming


The name is in the model. X570


OpenPGP_signature
Description: OpenPGP digital signature

Re: [gentoo-user] Re: setcap fails: (Operation not supported)

[Andrew Udvare]

On 30/09/2021 13:58, Grant Edwards wrote:

On 2021-09-30, Grant Edwards  wrote:

On 2021-09-30, Grant Edwards  wrote:


I'm trying to add NET_ADMIN capability to an executable that needs to
create a tun inteface. AFACIT, this is the command to do that:

$ sudo setcap cap_net_admin+ep example_app
Failed to set capabilities on file `example_app' (Operation not supported)

The only possible cause for that message Google has been able fo find
is that the FS doesn't have xattr support.


Is Posix ACL support required for setcap?

I can't find any documentation of such a requirement, but it's the
only other thing I can think of...


That's not it. I rebuilt my kernel with POSIX ACL support enabled for
ext4, rebooted, and verified that ACLs now work.

Still can't figure out how to get setcap to work

 # file example_app
 example_app: ELF 64-bit LSB pie executable, x86-64, version 1 (SYSV), 
dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 
3.2.0, with debug_info, not stripped

 #  setcap cap_net_admin,cap_net_raw+eip example_app
 Failed to set capabilities on file `example_app' (Operation not supported)

--
Grant



Not sure if this is it, but do you have CONFIG_EXT4_FS_SECURITY enabled?



OpenPGP_signature
Description: OpenPGP digital signature

Re: [gentoo-user] time to build a new machine ?

[Andrew Udvare]

On 24/09/2021 06:48, Philip Webb wrote:

210924 Andrew Udvare wrote:

On 2021-09-24, at 05:58, Philip Webb  wrote:

While I was asleep yesterday, my machine reported on all  3  Konsoles :
Message from syslogd@  at Thu Sep 23 19:38:11 2021 ...
: mce: [Hardware Error]: CPU 0: Machine Check: 0 Bank 4: 9d0b4c16001d011b
Message from syslogd@  at Thu Sep 23 19:38:11 2021 ...
: mce: [Hardware Error]: TSC 0 ADDR 19e617980 MISC c01a0100
Message from syslogd@  at Thu Sep 23 19:38:11 2021 ...
: mce: [Hardware Error]: PROCESSOR 2:600f20 TIME 1632440315 SOCKET 0 APIC 0 
microcode 6000822
-- end of report --

 From the manpage:


Which man page is that ?


man mcelog



OpenPGP_signature
Description: OpenPGP digital signature

Re: [gentoo-user] time to build a new machine ?

[Andrew Udvare]

> On 2021-09-24, at 05:58, Philip Webb  wrote:
> 
> While I was asleep yesterday, my machine reported on all  3  Konsoles :
> 
> Message from syslogd@  at Thu Sep 23 19:38:11 2021 ...
> : mce: [Hardware Error]: CPU 0: Machine Check: 0 Bank 4: 9d0b4c16001d011b
> 
> Message from syslogd@  at Thu Sep 23 19:38:11 2021 ...
> : mce: [Hardware Error]: TSC 0 ADDR 19e617980 MISC c01a0100 
> 
> Message from syslogd@  at Thu Sep 23 19:38:11 2021 ...
> : mce: [Hardware Error]: PROCESSOR 2:600f20 TIME 1632440315 SOCKET 0 APIC 0 
> microcode 6000822
> 
> -- end of report --
> 
> I don't remember seeing this before : how concerned should I be ?

From the manpage:

   Most  errors  can be corrected by the CPU by internal error correction 
mechanisms. Uncorrected
   errors cause machine check exceptions which may kill processes or panic 
the machine.  A  small
   number  of  corrected errors is usually not a cause for worry, but a 
large number can indicate
   future failure.

   When an uncorrected machine check error happens that the kernel cannot 
recover  from  then  it
   will  usually  panic  the  system.   In  this case when there was a warm 
reset after the panic
   mcelog should pick up the machine check errors after reboot.  This is  
not  possible  after  a
   cold reset.

If you are overclocking, try disabling it.

Re: [gentoo-user] Video plays past end of time it shows

[Andrew Udvare]

On Wed, Aug 25, 2021, 13:44 Dale  wrote:

> Hi,
>
> I have a video that does something weird.  The video plays about 6
> minutes or so past the length it should.  During that extended time, you
> can't fast forward, pause or anything either.  I've ran into this a
> couple times with my deer camera and thought it odd but didn't worry
> about it.  My question is, can this be fixed somehow?  Can I get mplayer
> or something to process the file and adjust the time it thinks the video
> is to what it actually is?
>

You can try to re-encode or maybe remux with ffmpeg.


> Bonus, how in the world did it get that way?
>

More than likely the metadata in the file is off.

This playback issue can happen with VBR MP3s and also Oggs as the length of
the file can be spoofed in the header.


>

Re: [gentoo-user] Windows 10 Pro 64bit in Virtualbox

[Andrew Udvare]

I use with a highly customised image made with NTLite where nearly
everything is removed, especially things like the antimalware, Cortana, etc.

This gives marginal gain in terms of boot up time, and general speed of the
UI (even with all animations disabled). It does however reduce the memory
requirement to about 1-1.5 GiB, which is nicer than average. However, that
barely fits a browser like Chrome or Edge.

I've been trying to perfect a minimal image setup for VirtualBox for a
while. It's taking a long time to test it.

On Sat, May 22, 2021, 17:02  wrote:

> Is anybody running Windows 10 Pro 64bit in Virtualbox?
> Is it stable?
> Is it easy to resize?
>
> Currently I'm still using Windows 7 Pro and it starts much faster then
> Windows 10 pro 32bit
> But Windows 7 is no longer supported and eventually I'll have to select
> Windows 10 Pro 32bit or 64bit to run my programs.
>
>
>

Re: [gentoo-user] Why do we add the local host name to the 127.0.0.1 / ::1 entry in the /etc/hosts file?

[Andrew Udvare]

> On 2021-02-21, at 17:23, Grant Taylor  
> wrote:
> 
> Hi,
> 
> I'm reading Kerberos - The Definitive Guide[1] and it makes the following 
> comment:
> 
>> And to make matters worse, some Unix systems map their own hostname to 
>> 127.0.0.1 (the loopback IP address).
> 
> This makes me think that the local host name /shouldn't/ be included in the 
> 127.0.0.1 (or ::1) entry in the /etc/hosts file.
> 
> However, according to the Gentoo AMD64 Handbook[2], we are supposed to add 
> the local host name to the 127.0.0.1 (and ::1) entry in the /etc/hosts file.
> 
> Will someone please explain why the Gentoo AMD64 Handbook ~> Gentoo (at 
> large) says to add the local host name to the 127.0.0.1 (or ::1) entry in the 
> /etc/hosts file?  What was the thought process behind that?

The thought process is the same as the guide you are reading. 
https://www.oreilly.com/library/view/kerberos-the-definitive/0596004036/ch04s03.html

It says in the last paragraph on this page that the /etc/hosts file should have 
the '127.0.0.1 localhost ' (last part is definitely optional) and 
then that the next entry should have the outer IP address mapped to the FQDN.

-- 
Andrew

Re: [gentoo-user] No news on kernel upgrade?

[Andrew Udvare]

> On 2021-02-09, at 03:43, n952162  wrote:
> 
> If an update requires additional steps, shouldn't that have appeared in the 
> news?

No because how it's done is totally dependent on your needs. There's no one 
correct way to do it. Even in common cases, some may choose to use certain 
kinds of file systems or other things that cannot be covered. Gentoo is all 
about customisability.

Shameless plug: use upgrade-kernel from my upkeep package[1] (which is on 
tatsh-overlay)[2] . It supports updating/rebuilding the kernel plus 
MBR|UEFI/GRUB/Dracut and systemd-boot+UEFI/Dracut/kernel signing (latest HEAD 
only).

1. https://github.com/Tatsh/upkeep#automatic-kernel-update-process
2. https://github.com/Tatsh/tatsh-overlay#installation

Re: [gentoo-user] HOWTO: Freezing/unfreezing (groups of) processes

[Andrew Udvare]

> On 2021-02-05, at 02:45, Walter Dnes  wrote:
> 
> done < /dev/shm/temp.txt

You don't need to write a temporary file. You can pipe this directly into the 
while loop:

while read
do
...
done < <(ps -ef | grep ${1} | grep -v "grep ${1}" | grep -v pstop)

Also to avoid the second grep in Bash at least:

grep "[${1:0:1}]${1:1}"

 $ ps -ef | grep '[l]vmetad'
root 965   1  0 Jan31 ?00:00:00 /sbin/lvmetad -f

^ No grep in output.

--
Andrew

Re: [gentoo-user] Nodejs overlay

[Andrew Udvare]

> On 2021-01-31, at 19:20, Michael Orlitzky  wrote:
> 
> The only thing a package manager does for you is that it allows one
> person (the Gentoo developer, in this case) to invest a little bit of
> time to save everyone else (the Gentoo users) a lot of time. When
> software has been packaged correctly, it integrates well with the rest
> of the system, gets constant maintenance and security updates, and can
> be managed from a central location in a consistent manner.
> 
> "Packaging" software like you describe isn't packaging it in this
> sense.

It still results in the convenience of having Portage as opposed to something 
else manage system files (files not in /home). It gives a consistent way of 
deploying the software on machines with the same distro or package manager.

> When you bundle everything together,
> 
>  1. Nothing is shared between packages so build time and disk 
> usage skyrockets.

This is NodeJS and 99% of stuff is plain JavaScript. Many packages are tiny. 
More time will be spent unpacking tiny distfiles and re-arranging them the way 
Node/Yarn/etc like to see them.

In Rust and Go, with Portage the distfiles are deduplicated as much as 
possible. But it's highly likely that one would have (Rust) random-1.0.crate 
and random-1.1.crate in distfiles because two packages need different versions. 
Wasteful? Yes except when those two packages are really necessary.

> 
>  2. The number of updates and thus the amount of work required
> 
> also skyrockets, for the same reason: when nothing is shared, you 
> 
> have to update each package whenever a dependency of a dependency 
> 
> of a dependency... changes.

I'm not sure what you are referring to here.

I was saying in my previous message that NPM's dependency issues is what makes 
most Node packages (IMHO) infeasible for something like Portage. We have 
smaller scale versions of this, with less conflicts: Python and Perl. They have 
thousands of packages but are shared decently, and generally easily patched to 
reduce the need for slotting.

Node, even without conflicts, would be like this but would have a lot more 
packages. Would Portage actually scale well to this? It's not really known.

> 
>  3. There are no security updates, ever. If you use anything written 
> in Rust or Go on Gentoo, or if you use anything that uses 
> anything written in Rust or Go, or. it will NEVER get a 
> security update. No one even bothers looking for security
> issues in these languages because the "find a bug then fix it"
> algorithm is infeasible.

I'm also not sure what you mean here. Docker is written in Go primarily and 
gets security updates just fine. It is one of the larger players in the Go 
world but still. Then of course on the Rust side, Mozilla is highly concerned 
with security (although their firing of most of the Rust team brings up 
questions).

> 
>  4. You do get to do updates with e.g. "emerge -puDN @world", but not 
> really, because no one is actually updating Gentoo packages every 
> time a dependency of a dependency changes.

That's because for Go, the most important thing is the binary. It doesn't have 
a concept of shared libraries. What goes into a Go binary is the latest version 
of its own code and the acceptable version of the dependencies. All .go source 
files, no libs.

Rust does have libraries, but Gentoo has decided not to deal with them yet. As 
a result, Rust and Go are treated similarly in ebuilds. Crates can be libs, and 
used dynamically, but they do not have to be. 
https://doc.rust-lang.org/reference/linkage.html

In the Node world, the equivalent to Go's bin is the main script executable, 
which usually people install with `npm -g` or `yarn -g`. Nobody would install a 
library like is-boolean globally because it has no real purpose on its own 
without something (the main script) using it. This may seem strange if you like 
to install libraries and then open the interpreter to play with it, but I 
mainly see this in the context of a project that has the node_modules/ 
directory in it, not with user-globally installed libs (the ones in ~/.node/).

> 
> So ultimately, there's little benefit and it introduces security
> vulnerabilities to our users who might be better off just using NPM or
> whatever (or using software written in a sane language).

I don't think it introduces security vulnerabilities more than any other new 
system, but it's very difficult to implement. It needs actual purpose (in the 
sense that Docker is the reason Go keeps being developed), and many users, and 
constant updating (and probably a lot of tooling and automation). I don't see 
much from the Node world to consider installing globally (meaning managed by 
Portage) and using everyday outside of a Node-based (or Yarn-based) project. I 
also don't consider it too much of a burden to add ~/.node/bin to PATH.

I am on a single user PC, but those with a server shared between users may want 
to have a truly 

Re: [gentoo-user] Nodejs overlay

[Andrew Udvare]

> On 2021-01-31, at 18:17, Aisha Tammy  wrote:
> 
> On 1/31/21 8:49 AM, Alessandro Barbieri wrote:
>> I'm trying to package nodejs software and I made an overlay for it here
>> https://github.com/Alessandro-Barbieri/node-overlay 
>> 
>> Could you try it and send feedback?
> 
> 
> I have to confess, I don't like nodejs a lot.
> But don't node packages have a very strict version dependency?
> Does this handle version conflicts?
> 
> X needs >=A-1.2 and <=A-1.3
> Y needs >=A-1.5
> 
> How would this be solved here? Isn't it for this reason that
> package installs are to a local folder for each application?

It can't be done in a sane way. Nodejs, because of their unwillingness to solve 
dependency issues and allowing insane amounts of tiny packages to exist (blame 
the community), allow insane dependency trees that make it impossible to put it 
on top of a normal package manager.

Even if we did find packages that don't have these conflicts, having Portage 
manage thousands of packages just for a single thing like eslint seems like a 
very bad idea.

Our best option is to treat Nodejs stuff the way we treat Rust and Go packages. 
Pretend Nodejs 'binaries' are 'built' statically and therefore, grab all the 
dependencies in the main package ebuild. One example is this overlay's vscode 
ebuild: 
https://data.gpo.zugaina.org/pf4public/app-editors/vscode/vscode-1.52.1.ebuild

The process of generating the SRC_URI can be automated. Eclasses could be 
written for to help with the phases, especially src_unpack and src_prepare.

Re: [gentoo-user] Browser have problems with illegal characters

[Andrew Udvare]

On 25/11/2020 07:24, Helmut Jarausch wrote:

Hi,

if I try to save a web page containing an illegal character (like 
hookrace.net/blog/introduction-to-metaprogramming-in-nim/)
several browser show an error (attached image) and cannot recover from 
it. I have no chance to change the filename.

I have to 'xkill' it

What can I do about this?

(My locale is en_US.iso88591)

Many thanks for a hint,
Helmut


This may be your window manager.



OpenPGP_signature
Description: OpenPGP digital signature

Re: [gentoo-user] Bluetooth and KDE Connect

[Andrew Udvare]

On Sun, Sep 20, 2020, 19:46 Peter Humphrey  wrote:

> Evening all,
>
> I've plugged in a new USB Bluetooth adapter, following the instructions in
> the
> wiki: https://wiki.gentoo.org/wiki/Bluetooth . The idea is to connect to
> my
> Android phone, but I'm only getting part way.
>
> After installing bluez and broadcom-bt-firmware, then booting with the new
> kernel, Bluetooth discovery and pairing work fine, but KDE Connect on each
> device can't see the other: they both report "No Devices".
>
> That nice Mr Google doesn't help, nor do the fine manuals and KDE web
> pages.
> Has anyone succeeded with this?
>

KDE Connect works over wi-fi by having your phone and computer on the same
network. You must run the app on your phone and it should see your computer
if it is running KDE Connect.

Make sure you allow the KDE Connect ports in your firewall. 1714-1764 for
UDP and TCP.

Bluetooth services from your phone are not related to KDE Connect.

Andrew

Re: [gentoo-user] What is the fastest file system for microsd?

[Andrew Udvare]

On Mon, Aug 31, 2020, 15:07 никита степанов  wrote:

> What is the fastest file system for microsd?


Any file system without journaling and other extra parts. I would place my
bets on ext2 in terms of raw speed.

There is also FAT and exFAT which should be fast enough. exFAT is optimised
for flash storage and is now in the kernel.

Re: [gentoo-user] custom mount fstab

[Andrew Udvare]

On Fri, Jul 3, 2020, 09:34 Tamer Higazi  wrote:

> Hi people,
>
> I had a problem with docker on gentoo and found the solution for all my
> problems with a custom mount command:
>
> |sudo mount -t cgroup -o none,name=systemd cgroup /sys/fs/cgroup/systemd
> Can anybody of you tell me how to add that one in /etc/fstab file ?
> best, Tamer |
>

Docker on OpenRC or systemd?

Also with the new cgroups or old?

AFAIK Docker doesn't work or support the new cgroups. :/

Re: [gentoo-user] EFI booting problem - understanding it

[Andrew Udvare]

On 02/07/2020 06:56, Peter Humphrey wrote:
> But then,
> # bootctl set-default 30-gentoo-5.7.7.conf
> Failed to update EFI variable: Invalid argument

Probably the kernel is blocking write access to EFI. This is on purpose
for safety as you can damage your firmware quite easily. systemd-boot
and others do not have this restriction. You also should be careful
writing to the EFI too much as the NVRAM flash may not be of high quality.

https://lwn.net/Articles/674940/

You can try using `chattr -i` against the files like:

chattr -i /sys/firmware/efi/efivars/Boot*

Then you can try with bootctl and others, but this is not guaranteed to
work.

On my ASUS motherboard I haven't been able to write to EFI variables
from within Linux for a long time. I have to add my keys in the BIOS and
set the default in systemd-boot.

The logic to write to a file in efivars is here:

https://github.com/torvalds/linux/blob/master/fs/efivarfs/file.c#L15

If you use strace with bootctl you'll probably see one of these errno
values.

Andrew



signature.asc
Description: OpenPGP digital signature

Re: [SOLVED] [gentoo-user] thunderbird small fonts on hidpi

[Andrew Udvare]

> On 2020-06-25, at 13:40, Valmor de Almeida  wrote:
> 
> On Thu, Jun 25, 2020 at 9:43 AM Andrew Udvare  wrote:
>> 
>> On 25/06/2020 04:59, Valmor de Almeida wrote:
>>> Hello,
>>> Does anyone know how to fix Thunderbird UI fonts on a HiDPI
>>> Gentoo/OpenBox system with no desktop?
>>> A similar problem exists with Firefox but it can be fixed using a CSS
>>> file (userChrome.css). However I can't get Thunderbird to work using a
>>> similar fix. The UI fonts are so small that the app is useless.
>>> Thanks,
>>> --
>>> Valmor
>>> 
>> 
>> Do you have an ~/.Xresources file? It can have something like:
>> 
>> Xft.dpi: 144
>> 
>> Which is also passed to Xorg for me: -dpi 144 (Plasma does this by
>> modifying SDDM service file(s) to systemd).
>> 
> 
> No I don't have it. I thought the line
> 
> xrandr --dpi 192 &
> or
> xrandr --output eDP-1 --mode "3840x2160" --scale "0.75x0.75"
> 
> in ~/.xinitrc was sufficient to fix Firefox; it doesn't fix TB.
> 
> I just added
> Xft.dpi: 192
> to ~/.Xresources and not only TB works but LibreOffice works too.
> Of course all of this defies the purpose of having an HiDPI laptop but
> it is either that or no use at all.

For Qt apps you might want these environment variables in .xinitrc:

QT_SCREEN_SCALE_FACTORS=DVI-I-0=1.5;DVI-I-1=1.5;HDMI-0=1.5;DP-0=1.5;DP-1=1.5;DP-2=1.5;DP-3=1.5;DP-4=1.5;DP-5=1.5;

(These device names are the same as what xrandr shows.)

QT_AUTO_SCREEN_SCALE_FACTOR=0

Currently I have an issue where opening Qt apps from certain contexts results 
in resetting the DPI scale to 2 instead of 1.5. Primarily this happens when 
Chrome launches a Qt-based viewer such as Kate, and sometimes it happens from 
Konsole. Have yet to file bugs.

Re: [gentoo-user] thunderbird small fonts on hidpi

[Andrew Udvare]

On 25/06/2020 04:59, Valmor de Almeida wrote:
> Hello,
> Does anyone know how to fix Thunderbird UI fonts on a HiDPI
> Gentoo/OpenBox system with no desktop?
> A similar problem exists with Firefox but it can be fixed using a CSS
> file (userChrome.css). However I can't get Thunderbird to work using a
> similar fix. The UI fonts are so small that the app is useless.
> Thanks,
> --
> Valmor
> 

Do you have an ~/.Xresources file? It can have something like:

Xft.dpi: 144

Which is also passed to Xorg for me: -dpi 144 (Plasma does this by
modifying SDDM service file(s) to systemd).



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Gentoo-sources 5.7.x [FIXED]

[Andrew Udvare]

On 09/06/2020 12:07, Peter Humphrey wrote:
> On Tuesday, 9 June 2020 15:56:52 BST Peter Humphrey wrote:
> # cat /boot/loader/entries/30-gentoo-5.7.1.conf
> title Gentoo Linux 5.7.1initrd=/intel-uc.img
> linux /vmlinuz-5.7.1-gentoo
> options root=/dev/nvme0n1p4 initrd=/intel-uc.img net.ifnames=0 
> raid=noautodetect

Since you are using systemd-boot or something that fulfils that
specification[1], you can also build your kernel with EFI stub enabled
(CONFIG_EFI_STUB) and then simply put the binary in:

${ESP}/EFI/Linux/vmlinuz-5.7.1-gentoo.efi

You can then run `bootctl set-default vmlinuz-5.7.1-gentoo.efi` or on
the menu, select it and press d to set it within systemd-boot.

(My ASUS motherboard for some reason never lets me write EFI variables
from within Linux so I have to do it from within systemd-boot.)

You can specify the options in the kernel configuration as well:

CONFIG_CMDLINE="root=/dev/nvme0n1p4 initrd=/intel-uc.img net.ifnames=0
raid=noautodetect"

To add the /intel-uc.img to this configuration you either have to
include that in kernel configuration or you can use Dracut to build an
EFI image.

Kernel config:

CONFIG_INITRAMFS_SOURCE="/boot/intel_uc.img"

Or with Dracut:

dracut --force --uefi --uefi-stub
'/usr/lib/systemd/boot/efi/linuxx64.efi.stub' ...

Dracut will automatically pick up your kernel installed to /boot (from
kernel `make install`) and /boot/intel-uc.img (and other similar
things). It will also automatically place the file into

In both cases, you have to remember to update the EFI image/rebuild and
reinstall the kernel whenever you update intel-microcode.

The benefit to this is you don't have to maintain entries files, and you
keep configuration generally in one place: the kernel config. Then you
just drop in EFI binaries into the correct place and they will appear in
the menu. You could have always keep two Linux EFI binaries in
${ESP}/EFI/Linux/ in case the newest one fails.

If you want to do this semi-automatically as part of updates and with
UEFI secure boot signing, use my project: https://github.com/tatsh/upkeep

[1] https://systemd.io/BOOT_LOADER_SPECIFICATION/

--
Andrew




signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Gentoo-sources 5.7.x

[Andrew Udvare]

On 08/06/2020 11:06, Peter Humphrey wrote:
> Afternoon all,
> 
> Is there something special for me to set in the kernel config to enable it to 
> find the root partition? I copied the config from 5.4.38, ran oldconfig and 
> followed most of the suggested answers; but it won't boot.

Sounds like missing drivers. oldconfig didn't do everything it was
supposed to. Moving across multiple major versions, this is to be
expected. A lot of names of things have changed.

Do a comparison of your configuration between old and new.

diff -uN old-config-file /usr/src/linux/.config

Make sure to at least enable NVME with CONFIG_BLK_DEV_NVME=y and try
booting 5.7 again. Other than that, the naming scheme may have changed
but I don't know about this. For better future-proofing, use a UUID of
your root partition rather than a device name.

root=UUID=...

You can get this UUID with the blkid command.

Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] new genkernel problem

[Andrew Udvare]

On Fri, 5 Jun 2020 at 15:02, Jack  wrote:
> Thanks for any thoughts or suggestions.
>
> Jack

I would keep GCC 9 for now. I've run into a few issues with GCC 10
compiling various packages so I have kept GCC 9 on my system. There's
not some serious detriment to not using GCC 10 for 99% of people. Most
distros still use an older version.

Also, have you considered trying Dracut for initramfs generation? I
don't know how well it works with genkernel as I don't use that (I
manually configure my kernels).

Re: [gentoo-user] Creating Blue-ray video discs

[Andrew Udvare]

On Fri, May 29, 2020, 05:02 Dale  wrote:

> Howdy,
>
> A few weeks ago, I ran up on a deal on a Blu-ray burner.  It's a LG and
> smartctrl -i shows this:
>
>
> === START OF INFORMATION SECTION ===
> Vendor:   HL-DT-ST
> Product:  BD-RE  WH16NS40
> Revision: 1.04
> Compliance:   SPC-3
> >> Terminate command early due to bad response to IEC mode page
> A mandatory SMART command failed: exiting. To continue, add one or more
> '-T permissive' options.
>
>
>
> Anyway, I have some HD videos I want to burn to a Blu-ray disc in HD.
> I've used Devede and Devedeng to create DVDs for a while.  I prefer the
> old Devede but the new ng version works well.  It doesn't however seem
> to create Blu-ray discs.  I googled and found how to play some of them
> at least that are commercially made.  I can't find however what software
> is used to create my own.
>
> Does anyone know what software to use to create HD video Blu-ray discs?
>

You're referring to Blu-ray authoring like creating menus and making a
video disc that works in a set-top player. There isn't any software for
Linux I know of that does this, especially the menu part. The menu part can
be in a simple format or it can be more advanced with BD-J.

Studios use Scenarist BD
https://www.scenarist.com/scenarist-bd-professional-blu-ray-disc-authoring/

There's MultiAVCHD for free. Maybe it works with Wine?

Re: [gentoo-user] Kernel config for Docker

[Andrew Udvare]

On 16/05/2020 13:12, Peter Humphrey wrote:
> I can't find any of those. Any clues for the uninitiated?

I am running Docker fine on 5.6.12 and I am missing a lot:

- CONFIG_NF_NAT_IPV4: missing
- CONFIG_IP_NF_FILTER: missing
- CONFIG_IP_NF_TARGET_MASQUERADE: missing
- CONFIG_NETFILTER_XT_MATCH_ADDRTYPE: missing
- CONFIG_NETFILTER_XT_MATCH_CONNTRACK: missing
- CONFIG_NETFILTER_XT_MATCH_IPVS: missing
- CONFIG_IP_NF_NAT: missing
- CONFIG_NF_NAT_NEEDED: missing
- CONFIG_IOSCHED_CFQ: missing
- CONFIG_CFQ_GROUP_IOSCHED: missing
- CONFIG_IP_NF_TARGET_REDIRECT: missing
- CONFIG_EXT3_FS_XATTR: missing
  - CONFIG_INET_XFRM_MODE_TRANSPORT: missing
- CONFIG_NF_NAT_FTP: missing
- CONFIG_NF_CONNTRACK_FTP: missing
- CONFIG_NF_NAT_TFTP: missing
- CONFIG_NF_CONNTRACK_TFTP: missing
- CONFIG_AUFS_FS: missing
- CONFIG_BTRFS_FS: missing
- CONFIG_BTRFS_FS_POSIX_ACL: missing
- /dev/zfs: missing
- zfs command: missing
- zpool command: missing

In regards to NF options, I use nftables and I manage the firewall
manually for Docker (I set {"iptables": false} in
/etc/docker/daemon.json). Docker has been extremely slow at adopting
nftables.

You definitely do not need zfs installed to use Docker. This machine
doesn't have it.

As Victor stated, CFQ is deprecated and gone and BFQ will work fine. And
the script is basically for Red Hat (or corporate) users who still use
iptables, CFQ, ext3 apparently, and a much older kernel.

On my slightly stabler server (running 5.4.38), this is the output and
Docker still works fine. Again on that server I use nftables and manage
the firewall manually. The system has ext4 for its root and the rest of
it is ZFS.

- CONFIG_BRIDGE_NETFILTER: missing
- CONFIG_NF_NAT_IPV4: missing
- CONFIG_IP_NF_FILTER: missing
- CONFIG_IP_NF_TARGET_MASQUERADE: missing
- CONFIG_NETFILTER_XT_MATCH_ADDRTYPE: missing
- CONFIG_NETFILTER_XT_MATCH_CONNTRACK: missing
- CONFIG_NETFILTER_XT_MATCH_IPVS: missing
- CONFIG_IP_NF_NAT: missing
- CONFIG_NF_NAT_NEEDED: missing
- CONFIG_IOSCHED_CFQ: missing
- CONFIG_CFQ_GROUP_IOSCHED: missing
- CONFIG_IP_NF_TARGET_REDIRECT: missing
- CONFIG_EXT3_FS: missing
- CONFIG_EXT3_FS_XATTR: missing
- CONFIG_EXT3_FS_POSIX_ACL: missing
- CONFIG_EXT3_FS_SECURITY: missing
  - CONFIG_INET_XFRM_MODE_TRANSPORT: missing
- CONFIG_NF_NAT_FTP: missing
- CONFIG_NF_CONNTRACK_FTP: missing
- CONFIG_NF_NAT_TFTP: missing
- CONFIG_NF_CONNTRACK_TFTP: missing
- CONFIG_AUFS_FS: missing
- CONFIG_BTRFS_FS: missing
- CONFIG_BTRFS_FS_POSIX_ACL: missing

Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] a day of PAIN.

[Andrew Udvare]

> On 2020-05-12, at 18:45, Alan Grimes  wrote:
> 
> Why is this not a forced-on setting for any machine with UEFI enabled? I
> can't imagine that this would be unacceptable for more than 0.001% of
> the install base.

Because not every user has UEFI and many just use the BIOS compatibility layer 
(CSM) if they don't care about UEFI and Secure Boot. The CSM is going to be 
there for a very long time.

If you don't need UEFI features, then there's nothing really wrong with using 
CSM. I am using UEFI because I like having a signed kernel and signed boot 
loader (systemd-boot).

--
Andrew

Re: [gentoo-user] NVidia-driver, RTX 2060 SUPER. Blender and NO Optix...

[Andrew Udvare]

> On Apr 4, 2020, at 00:59, Dale  wrote:
> 
> tu...@posteo.de wrote:
>> Hi,
>> 
>> I have discussed this on www.blenderartists.org and they asked
>> me to ask here to sort out, whether the problem is a Blender-thing,
>> a Linux-thing or a GENTOO-thing.
>> 
>> My setup is as follows:
>> NVidia RTX 2060 SUPER
>> 
>> NVidia-drivers:
>> [I] x11-drivers/nvidia-drivers
>> Available versions:  (~)304.137-r1(0/304)^md[1] 
>> (~)340.107-r2(0/340)^md[1] 340.108(0/340)^mtd (~)375.82-r2(0/375)^md[1] 
>> (~)378.13-r5(0/378)^md[1] (~)381.22-r3(0/381)^md[1] 
>> (~)384.130-r1(0/384)^md[1] (~)387.34-r1(0/387)^md[1] 
>> (~)390.77-r1(0/390)^md[1] (~)390.87(0/390)^md[1] 390.132-r1(0/390)^mtd 
>> (~)390.132-r2(0/390)^mtd (~)396.24-r2(0/396)^md[1] 
>> (~)396.24.10-r1(0/396.24)^md[1] (~)396.45-r1(0/396)^md[1] 
>> (~)396.51-r1(0/396)^md[1] (~)396.51.02(0/396.51)^md[1] 
>> (~)396.54(0/396)^md[1] 430.64-r1(0/430)^mtd 435.21-r1(0/435)^mtd 
>> 440.64(0/440)^mtd {+X acpi compat +driver gtk3 +kms +libglvnd multilib 
>> pax_kernel static-libs +tools uvm wayland ABI_MIPS="n32 n64 o32" 
>> ABI_RISCV="lp64 lp64d" ABI_S390="32 64" ABI_X86="32 64 x32" KERNEL="FreeBSD 
>> linux"}
>> Installed versions:  440.64(0/440)^mtd(03:03:25 AM 04/03/2020)(X driver 
>> kms libglvnd static-libs tools uvm -acpi -compat -gtk3 -multilib -wayland 
>> ABI_MIPS="-n32 -n64 -o32" ABI_RISCV="-lp64 -lp64d" ABI_S390="-32 -64" 
>> ABI_X86="64 -32 -x32" KERNEL="linux -FreeBSD")
>> Homepage:https://www.nvidia.com/
>> Description: NVIDIA Accelerated Graphics Driver
>> 
>> Blender 2.82a (stable) and
>> Blender 2.83  (deveoper build)
>> 
>> The NVidia RTX-cards offer a new feature called "Optix" which blender
>> can use to speed up rendering and denoising.
>> 
>> When Blender is started one choose "Optix" from the user
>> Preferences->System tab and then the Optix-enabled devices of the
>> system in question are shown.
>> There is a similiar tab, if you want to use CUDA instead.
>> 
>> The CUDA tab shows my graphics card and everything behaves as
>> exsoected. Choosing "Optix" instead says "No Optix enabled
>> device".
>> 
>> Which is not quite right, since the RTX-cards are Optix enabled.

I suggest filing a bug about this regarding the nvidia-drivers package. It's 
possible it's not installing the necessary files.

Re: [gentoo-user] [OFFTOPIC}: Does the MSI Tomahawk MAX is able to boot from a M2 SSD?

[Andrew Udvare]

> On Mar 19, 2020, at 23:29, tu...@posteo.de wrote:
> 
> Beside speed and cost:
> Is there any difference between using a M2 SATA SSD
> or a M2 NVMe SSD?

I can tell you there's a huge difference in speed as NVMe is essentially the 
same as plugging directly to PCI-E. m.2 SATA is basically for backward 
compatibility with older OS.
> 
> Does these drives are working with UEFI boot and
> Legacy boot when it comes to boot Linux?

You shouldn't have a problem with either. I've been booting with my own signed 
Grub for years with UEFI secure boot.

If for some reason the motherboard won't take your self-made certificates/etc, 
return it.

Andrew

Re: [gentoo-user] qt4

[Andrew Udvare]

> On 2020-01-18, at 03:17, k...@aspodata.se wrote:
> 
> Is there a way to install qt4 without resorting to make a new 
> installation ?
> 
> This code requires qt4:
> https://github.com/thliebig/openEMS-Project

Not from what I'm seeing. The README has an outdated line for Ubuntu but the 
QCSXCAD GUI code references Qt5.

https://github.com/thliebig/QCSXCAD/blob/0dabbaf2bc1190adec300871cf309791af842c8e/CMakeLists.txt#L109

Re: [gentoo-user] can you save a filesystem destroyed with mkswap?

[Andrew Udvare]

> On 2020-01-16, at 02:30, n952162  wrote:
> 
> Does anyone have experience with this?  Is it possible to a save an
> ext3/4 system destroyed by mkswap?
> 
> 
> 

Try app-admin/testdisk

Get to a point where you know for sure nothing is writing to the disk (or would 
be) and make an image with dd, then leave the physical disk alone. Then start 
recovery.

Andrew

Re: [gentoo-user] /usr/portage/distfiles vs /var/cache/distfiles

[Andrew Udvare]

On 14/01/2020 17:05, Paul B. Henson wrote:
> I recently installed a new server, and noticed that it is using
> /var/cache/distfiles rather than /usr/portage/distfiles. looking at the
> documentation, it seems that is the new default, and it says "new
> installations" will use it. I have older servers which are still using
> /usr/portage/distfiles. None of them have DISTDIR configured, so I am
> curious how portage is deciding between "old" and new"? If I wanted to
> migrate my older boxes to the new location without explicitly setting
> DISTDIR on them, what do I need to change?
> 
> Thanks…
> 
I attempted to do this migration on two machines but was only successful
once. Pretty sure this is because of not using 17.1 profile on the
failing machine.

The first machine is on a 17.1 profile, and migration was relatively
simple, and worked. I did it based on this post:

https://forums.gentoo.org/viewtopic-p-8373904.html?sid=5c30f451df89c01a6bbe0a8dadd77c36#8373904

On the second machine, I did not change to profile 17.1. After moving
the files around, no matter what I tried, Portage refused to check the
new DISTDIR. I haven't had time to switch this machine to 17.1 yet.

Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] safe use of .gnupg

[Andrew Udvare]

> On Dec 17, 2019, at 20:51, Philip Webb  wrote:
> 
> When encrypting a file, I was told :
> 
>  root:552 root> gpg -c  
>  gpg: WARNING: unsafe ownership on homedir '/home/purslow/.gnupg'
> 
> The file is owned by my user, ie  : .
> This seems to be the default when 'gpg' is installed.

It's probably complaining if you're running as root and you've set the GPG home 
did to be in /home/purslow/.gnupg rather than /root/.gnupg (and owned by 
root:root). Otherwise try setting that directory to 0700 permission (u+rwx 
g-rwx o-rwx).

Andrew

Re: [gentoo-user] Kernel panic on 5.4.2 - not sure of cause yet

[Andrew Udvare]

On 15/12/2019 01:54, Andrew Udvare wrote:
> On 12/12/2019 10:18, Daniel Frey wrote:
>>
>> I have just installed fresh new gentoo installs with 5.4.2 and both
>> machines use nvidia-drivers - I have not seen this at all.
>>
>> I have been doing a fair bit of compiling on one of the machines and
>> haven't had any hiccups whatsoever.
>>
>> Dan
>>
> 
> Thanks for the reply.
> 
> I think I have found the issue in another 3rd party blob-ish driver from
> Magewell, which is for my capture card. I removed the capture card
> (since this would stop the module from loading) and my system has not
> had a panic since. I have not tried to upgrade back to 5.4.2 yet but as
> I said before I was getting the error on 5.4.0 too.
> 
> Andrew
> 

For anyone interested, here are some pictures of the panics:

https://i.imgsafe.org/5e/5e20756985.jpeg (unknown)
https://i.imgsafe.org/5e/5e20754111.jpeg (find_css_set)

The driver in question is the one named ProCapture. I am fairly certain
this is the cause of the recent panics.

I have had some issues with overclocking my system (both CPU and memory)
but I do not think this is related. Just for good measure I reseated all
my RAM modules. I am currently running 3.7 GHz (from 3.5) with memory
overclocked to 2667 MHz. Does not seem to have an issue.

Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Kernel panic on 5.4.2 - not sure of cause yet

[Andrew Udvare]

On 12/12/2019 10:18, Daniel Frey wrote:
> 
> I have just installed fresh new gentoo installs with 5.4.2 and both
> machines use nvidia-drivers - I have not seen this at all.
> 
> I have been doing a fair bit of compiling on one of the machines and
> haven't had any hiccups whatsoever.
> 
> Dan
> 

Thanks for the reply.

I think I have found the issue in another 3rd party blob-ish driver from
Magewell, which is for my capture card. I removed the capture card
(since this would stop the module from loading) and my system has not
had a panic since. I have not tried to upgrade back to 5.4.2 yet but as
I said before I was getting the error on 5.4.0 too.

Andrew



signature.asc
Description: OpenPGP digital signature

[gentoo-user] Kernel panic on 5.4.2 - not sure of cause yet

[Andrew Udvare]

I have been getting relatively consistent kernel panics on some call to
find_css_set and sometimes a stack trace that mentions cgroups.

On 5.4.0 I don't get this same crash and I added blocking of
auto-loading nvidia under the ramdisk just in case that's the issue, as
I was sometimes getting a similar crash on 5.4.0.

/etc/default/grub:

GRUB_PRELOAD_MODULES=lvm
GRUB_CMDLINE_LINUX="init=/usr/lib/systemd/systemd
systemd.legacy_systemd_cgroup_controller=yes rd.driver.blacklist=nvidia
rd.driver.blacklist=nvidia_modeset rd.driver.blacklist=nvidia_drm"
GRUB_GFXPAYLOAD_LINUX="keep"

The reason the legacy argument is there is because Docker won't work
under the new cgroups, for now.

I have a couple of modules, but the one that sticks out most is nvidia.
This is the one I see in the stack trace. I have not seen a bug report
on Gentoo or Nvidia's end.

For now I've masked >5.4.0 gentoo-sources.

Anyone else getting a similar issue?

Thanks
Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] app-text/foliate ebuild?

[Andrew Udvare]

On 27/11/2019 07:50, Helmut Jarausch wrote:
> Hi,
> I came across the "modern ebook reader" foliate
> (github.com/johnfactotum/foliate)
> Since I don't know about meson, I failed to write an ebuild for that.
> 
> Has anybody know about an ebuild for foliate?
> 
> Many thanks for some help,
> Helmut
> 

Use the Meson eclass.

https://devmanual.gentoo.org/eclass-reference/meson.eclass/index.html



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Noah's ArK

[Andrew Udvare]

> On 2019-11-17, at 06:19, Helmut Jarausch  wrote:
> 
> I'd like to "preserve" some packages which do require components I don't like 
> to have "regularly" installed any more like some depending on Python2 or are 
> 32bit.

For Python (any version), use wheels:

https://pypi.org/project/wheel/

You would have to clone/download the packages yourself, and then run `python 
setup.py bdist_wheel` for them. This also ensures that you preserve compiled 
versions of the packages. As time goes on, these older packages will not 
compile against newer GCC/Clang versions without patches. Most popular packages 
already have wheels, although they might not have every combination built.

For anything else, if it doesn't need stuff like hardware 3D acceleration, use 
a VM that retains the packages you want. This also leads into...

If it does need native hardware access, use chroot and/or disk images (that you 
would boot into separately). I prefer to try and create barriers between 
'production' (what I use everyday) and one-off things like some old game that 
only supports 32-bit.

You probably want to retain old versions of toolchains (GCC, binutils, Clang, 
LLVM, etc) in case you need to build anything (built with -mtune not -march). 
Use Gentoo's binpkg format for this with the `qpkg` command. These packages can 
also be deployed on non-Gentoo systems but YMMV as you look for dependencies.

I tend to keep old hardware around for running old software. You might want to 
do the same.

A lot of older (much older) x86 support is being added to MAME everyday. This 
may come in handy in the future to preserve older versions of Linux distros and 
apps/games.

https://www.mamedev.org/releases/whatsnew_0215.txt (search 386)

Andrew

Re: [gentoo-user] problem changing to 17.1 profile

[Andrew Udvare]

On 02/11/2019 01:36, John Covici wrote:
> Hi.  Well, I was finally able to do the change of profile tothe 17.1
> profile.  I have gotten all  the way almost to the end of this process
> to the final step where itwantsto emerge all the 32-bit packages.  I
> am about two from the end of that list and trying to emerge
> x11-libs/gtk+-2.24.32-r1  andI have run into this problem:
> 
> libtool: link: x86_64-pc-linux-gnu-gcc -O2 -mtune=core2 -pipe -ggdb
> -Wall -Wl,-O1 -o decompose-bits decompose-bits.o  -Wl,--as-needed
> /usr/lib/libatk-1.0.so -lpango-1.0 -lcairo -lgdk_pixbuf-2.0
> -lgobject-2.0 -lglib-2.0
> /usr/lib/gcc/x86_64-pc-linux-gnu/8.3.0/../../../../x86_64-pc-linux-gnu/bin/ld:
> /usr/lib/libatk-1.0.so: error adding symbols: file in wrong format
> collect2: error: ld returned 1 exit status
> 
> Now, I re-emerged dev-libs/atk which contains the offending
> library, but no joy.  the gtk package is pulled in by   several other
> packages, so I cannot remove it.

If you re-emerged atk successfully, then the gtk linking stage should
not fail with the same error as /usr/lib/libatk-1.0.so should now be in
the correct format (32-bit instead of 64-bit).

> 
> Thanks in advance for any suggestions asto howto proceed.
> 

You have to pull in some packages in a very specific order.

With KDE we have a sort of circular dependency on freetype[harfbuzz],
because harfbuzz itself depends on freetype. For this reason I had to do
the following:

env USE=-harfbuzz emerge -1 freetype
emerge -1 harfbuzz
emerge -1 freetype

There is currently no way for Portage to do this automatically.

Your issue sounds similar and I would suggest emerging offending
packages separately before continuing the rebuild process for /lib32,
etc. The order is going to depend on other packages you have installed.

As soon as one of these special packages errors, you need to check the
build log. Most likely it was during the configure stage where a
dependency is not being found in /lib because it has not been built yet.
Build that dependency with --oneshot and try again with the failing
package after.

--
Andrew



signature.asc
Description: OpenPGP digital signature

[gentoo-user] Docker + systemd since cgroups v2

[Andrew Udvare]

For reference: uname -r: 5.2.13-gentoo, systemd version 
243_rc2-r1[cgroup-hybrid], ACCEPT_KEYWORDS="~amd64"

My system started failing to start running docker.service automatically and the 
logs weren't too helpful. Finally I ran dockerd on its own and found that it 
gave me this error message:

Devices cgroup isn't mounted

This is not too easy to diagnose as there seem to be a set of solutions but 
none of the main two worked for me. One involved setting 2 options on the 
kernel command line:

cgroup_enable=memory swapaccount=1

And the other was to add USE="cgroup-hybrid" to systemd. I did this, but it too 
it did not work.

The other solution is to simply mount the cgroup manually and this works but I 
did not see why I'd have to do that now when I never had to in the past.

I actually had to add this to my command line:

systemd.legacy_systemd_cgroup_controller=yes

This has been noted in other distros but from what I can tell this is solely 
because runc has not been updated to be able to use cgroups v2.

Anyone else ran into this issue? Is there something I am missing so I wouldn't 
need to pass a kernel command line option?

Reference links:

https://docs.docker.com/install/linux/linux-postinstall/#your-kernel-does-not-support-cgroup-swap-limit-capabilities
https://github.com/opencontainers/runc/issues/654
https://github.com/opencontainers/runc/issues/1175
https://wiki.gentoo.org/wiki/Docker#Docker_service_fails_because_cgroup_device_not_mounted_.28systemd.29


-- 
Andrew Udvare

Re: [EXTERNAL] [gentoo-user] empty cdrom drive is busy or mounted

[Andrew Udvare]

> On 2019-08-22, at 12:31, Laurence Perkins  wrote:
> 
> A common tactic is to use grep twice:
> ps auxf | grep -v grep | grep blah

Or grep with brackets:

ps aux | grep '[f]irefox'

I have a function for this:

psgrep() {
  ps aux | grep "[${1:0:1}]${1:1}";
}

This works because the ps output will have "grep [f]irefox" and the regex can't 
match that line (without escaping the [] again).


-- 
Andrew

Re: [gentoo-user] setup x11/plasma in chroot?

[Andrew Udvare]

> On 2019-08-22, at 20:13,  
>  wrote:
> 
> Can I install x11 and plasma on a new system I've chrooted into or do I have 
> to boot into the new system?  thanks. 
> -- “The whole world is watching! The whole world is watching!”
> 
You can install these things fine. Running them within the chroot can be a bit 
painful, however.

Re: [gentoo-user] version of portaudio conflicts with itsself?

[Andrew Udvare]

On 05/07/2019 22:05, Arve Barsnes wrote:
> Portage would like to install portaudio-19.06.00-r2, but audacity
> requires a version lower than that. I have the same output on my box,
> waiting for an audacity release I guess.

I have similar output especially with packages that are slotted like
wine-mono:

[U] app-emulation/wine-mono
 Available versions:
 (4.6.4) 4.6.4
 (4.7.0) (~)4.7.0
 (4.7.1) 4.7.1
 (4.7.3) (~)4.7.3
 (4.7.5) 4.7.5
 (4.8.0) (~)4.8.0
 (4.8.1) (~)4.8.1
 (4.8.3) (~)4.8.3
 (4.9.0) (~)4.9.0

   Installed versions:  4.7.5(4.7.5)(13:55:46 28/01/19)

 Homepage:
https://www.winehq.org/
 Description: Wine Mono is a replacement for the .NET
runtime and class libraries in Wine


If I run an upgrade this never gets updated. My assumption is that the
Wine ebuild is not requiring >wine-mono:4.7.5.



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Anyone with experience viewing EXIF data in Dolphin?

[Andrew Udvare]

On 05/07/2019 15:37, Andrew Lowe wrote:
> Hi all,
> I'm transferring my install from a spinning disk to an SSD and
> decided to tidy up/customise/stuff up things as I go. I've recently
> taken a lot of photo's whilst travelling and I have the Coords in the
> EXIF data. What I would like is to be able to easily view this data.
> There is a thingy called "ReImage"[1] which allows for a right click
> within Dolphin and you can select an option to display the EXIF data in
> a dialogue box.
> 
> I had to manually install this as I could find nothing on "Dolphin
> Service Menus" within Portage, or the Gentoo wiki or basically on line
> at all, so is there a Gentoo way of dealing with these in the first place?

Not really. You can submit a bug or PR though as some of these do make
into the tree sometimes.

$ eix --homepage store.kde.org

> 
> Secondly, I then came across "kfilemetadata" in Portage. Without
> installing this as well, does this give the same thing? If so, how do
> you use this thing? Acutally, as I'm writing this, I'm still reserching
> and api.kde.org[2] says it's a library, so what apps use it?

Not the same thing.

kfilemetadata is solely a library and should be pulled in by other
packages that need it.

You have to enable USE="semantic-desktop" to make Dolphin and others
pull it in and use it.



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Re: Gvim icon problem

[Andrew Udvare]

On 18/06/2019 12:06, Philip Webb wrote:
> 190615 Nikos Chantziaras wrote:
>> On 15/06/2019 12:00, Philip Webb wrote:
>>> I've just upgraded to Gvim 8.1.1486 & most of the icons have disappeared.
>> Is kde-plasma/kde-gtk-config installed ?
> 
> Yes.
> 
>> Open the KDE "System Settings", and go to "Application Style".
>> Select "GNOME/Gtk Application Style".
>> Make sure "Icon Theme" is set to "Breeze" 
>> & "Fallback theme" is set to "Adwaita".  If there's no "Adwaita" option,
>> then emerge x11-themes/adwaita-icon-theme.
> 
> Done & the icons are restored (screenshot attached),
> but they're not as pretty as those before (see shot in earlier msg).

You should be able to use Oxygen as your icon theme (I do this). The
screenshot looks like you have a mix of Breeze and Adwaita icons.



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Where to put Happauge capture card firmware?

[Andrew Udvare]

On 16/06/2019 04:34, J. Roeleveld wrote:
> There should be a /lib/firmware containing all the firmware files.
> See the linux-firmware ebuild for the exact location.

Thanks. Installing linux-firmware and adding this file to my initrd in
Dracut configuration solved my issue.




signature.asc
Description: OpenPGP digital signature

[gentoo-user] Where to put Happauge capture card firmware?

[Andrew Udvare]

I get this error in my boot log:

Jun 16 03:54:55 limelight kernel: cx25840 2-0044: Direct firmware load
for v4l-cx23885-avcore-01.fw failed with error -2
Jun 16 03:54:55 limelight kernel: cx25840 2-0044: unable to open
firmware v4l-cx23885-avcore-01.fw

I can get this file, but I have not figured out where I am supposed to
put it.

Thanks



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Tracking kernel patches

[Andrew Udvare]

> On 2019-05-25, at 20:39, Adam Carter  wrote:
> 
> I need to use this patch;
> https://marc.info/?l=linux-block=155772952511144=2
> 
> Is there some way to track its progress toward inclusion in the stable 
> release?

https://github.com/torvalds/linux/commit/09324d32d2a0843e66652a087da6f77924358e62#diff-a3fadbb3daba4dd336ddc1846df3e6c8

Can also look at kernel.org.

I just dealt with this with the TRIM error and its patch.

The easiest way is to look at the patch file(s) being changed, and then go to a 
web mirror to check that file's history.

Example:
https://www.redhat.com/archives/dm-devel/2019-May/msg00104.html
https://github.com/torvalds/linux/commits/master/drivers/md/dm.c
https://github.com/torvalds/linux/commit/51b86f9a8d1c4bb4e3862ee4b4c5f46072f7520d#diff-309bcac7077137714d954d0a75c854d1

If you go to the diff for 5.1.4-5.1.5 (GitHub does not have these tags) you can 
see this patch was included in the release:

https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/diff/?id=v5.1.5=v5.1.4=3=0=0

--
Andrew Udvare


signature.asc
Description: Message signed with OpenPGP

Re: [gentoo-user] Re: New Intel CPU flaws discovered

[Andrew Udvare]

> On May 17, 2019, at 01:14, Adam Carter  wrote:
> 
> The classic one is where OPS haven't noticed that disks in a RAID array have 
> died years ago...

This really happened?

Re: [gentoo-user] New network cards default to "Y" with "make oldconfig"

[Andrew Udvare]

> On Mar 23, 2019, at 21:03, Walter Dnes  wrote:
> 
>  When setting up a new kernel with "make oldconfig", almost all new
> device drivers default to "N".  The glaring exception is network cards.
> They all seem to default to "Y".  Is this a bug or a "feature"?

This has been a 'feature' for a while. I find it very annoying.

I suppose the idea is to not have users get annoyed that their network card 
can't be used after building a (generic) kernel. And they consider it justified 
since network connectivity is crucial to get things done. More for the Ubuntu 
users than everyone else.

Re: [gentoo-user] [OT] dmg2img hfs+ disk image

[Andrew Udvare]

On 23/02/2019 04:30, Mick wrote:
> # mount -t hfsplus -o loop 4.hfs /media/dmg
> mount: /media/dmg: failed to setup loop device for 4.hfs.

What does dmesg show? It should give more detail about the problem. Also
try with `mount -v`.

Normally 7z gets into the partition and extracts the files within, but
only if the 'Method' is supported, like 'Zero0 Copy Zero2 ZLIB CRC'.
Yours is 'Copy Zero2 ZLIB CRC'.

Also you might want to try https://github.com/darlinghq/darling-dmg with
the DMG file itself. This is far more active than the kernel module.

Re: [gentoo-user] [OT] dmg2img hfs+ disk image

[Andrew Udvare]

> On Feb 22, 2019, at 19:00, Mick  wrote:
> 
> Hi All,
> 
> A bit off topic, but given I'm trying this on a gentoo system and the high 
> cumulative knowledge of contributors to this M/L, I thought I might as well 
> ask here.  I'm trying to access the various partitions within a .dmg archive, 

This way using the kernel HFS+ driver:

https://github.com/nickl-/xchain-ios#prepare-your-prefix-no-mac

This way without the driver:

https://github.com/Tatsh/xchain#prepare-your-prefix-no-mac

Re: [gentoo-user] Android ADB emerge fails on amd64

[Andrew Udvare]

> On Jan 29, 2019, at 08:12, m4110c  wrote:
> 
> Thus spoke Andrew Udvare (audv...@gmail.com):
>> 
>> Choose another JDK with 'eselect java-vm set system ...'. You might need to 
>> install another like icedtea or Oracle's version.
>> 
> 
> I use Oracle JDK 1.8. I thought maybe I need a 32bit JDK, because it
> installed some dependencies with the abi_x86_32 use flag.
> 
> How would that work then. Do I have to switch (eselect) JDKs whenever I want 
> to
> use Android ADB stuff?

Without other configuration, any Gentoo setup can only have one JVM going (one 
for system and one for local user). So if you have apps that are incompatible 
between JVMs you will have to switch every time.

I'm not sure ADB is a Java app, but I think it's not, so you don't have to 
switch just for ADB.

The Android SDK binary distribution comes with its own copy of the JDK IIRC, if 
you want to go with that.

Re: [gentoo-user] Android ADB emerge fails on amd64

[Andrew Udvare]

> On Jan 29, 2019, at 07:31, m4110c  wrote:
> 
> Hi all,
> 
> I'm trying to install the Android Tools to use "adb" with my phone on
> gentoo.
> 
> When I try to install android-sdk-update-manager it fails on compiling
> its dependency dev-java/swt-3.7.2-r2.
> 
> The output of "emerge --info '=dev-java/swt-3.7.2-r2::gentoo'":

Choose another JDK with 'eselect java-vm set system ...'. You might need to 
install another like icedtea or Oracle's version.

Re: [gentoo-user] dev-php/xdebug-2.6.1 will not build

[Andrew Udvare]

> On Jan 29, 2019, at 06:26, John Covici  wrote:
> 
> Hi.  I have dev-php/xdebug in my world file, but when I tried to do my
> world update I get the following.  Note that I have the following line
> in my make.conf
> PHP_TARGETS="php5-6 php7-0 php-7-1 php-7-2"

The last two are not written correctly. Remove the extra dash: php7-1 php7-2

Re: [gentoo-user] Video database software

[Andrew Udvare]

> On 2019-01-28, at 17:54, Dale  wrote:
> 
> So far, I have installed Griffith and GCStar.  I been googling for
> others but some either are not in the tree or I already know they won't
> do one thing I'd like to see.  I'd also like to be able to point it to a
> directory and let it build the database on its own.  Adding them one at
> a time manually just isn't feasible at all. 

Seems like you could import via command line? 
http://wiki.gcstar.org/en/execution

You can build the database you need locally with something like exiftool or 
MediaInfo, or even ffmpeg https://stackoverflow.com/a/8191228/374110 . I highly 
doubt anyone with serious collections is building their database one item at a 
time.
> 
> Does anyone know of a software package that will sort a lot of videos by
> resolution as well as track other things as well?  It could be that what
> I'd like to have doesn't exist at all.  Then again, maybe I just haven't
> found it yet.  ;-)

The closest thing I can think of is Kodi since it's scanner will retrieve all 
this information and store it in a straightforward database format. You can 
choose SQLite or MySQL (of course MySQL is definitely the better choice for 
larger collections). The downside is the scanner is very slow, especially over 
a network (and not optimised). The only viewer for this data (at the time 
being) is Kodi itself.

-- 
Andrew

Re: [gentoo-user] OT scripting - strip zero if between period and digit

[Andrew Udvare]

On 21/01/2019 18:50, Adam Carter wrote:
> I need to clean up a file which has IP addresses with leading zeros in
> some of the octets so I need to make, say, .09 into .9
> 
> How do i do that in sed/awk/whatever?

A regex would be difficult. Parser is what you want.

You could use Python's ipaddress module (Python 3.3+). It will fix your
IPs (below is all one line):

python -c $'import ipaddress, sys;\nfor x in sys.argv[1:]:
print(ipaddress.ip_address(x))' 1.02.3.4 001.002.003.004

Output:
1.2.3.4
1.2.3.4

Fix that for stdin:

python -c $'import ipaddress, sys;\nfor x in sys.stdin.readlines():
print(ipaddress.ip_address(x.strip()))' <<< $'1.02.3.4\n001.002.003.004'

That way you can do:

python -c $'import ipaddress, sys;\nfor x in sys.stdin.readlines():
print(ipaddress.ip_address(x.strip()))' < list-of-ip-addresses

I'm sure there's a nicer way with modules installed with other languages
but this is built into Python as of version 3.3.

Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] [OT] ...preventing MAC OS from polluting my USBsticks...?

[Andrew Udvare]

> On 2019-01-13, at 13:38, tu...@posteo.de wrote:
> 
> Only to get sure not to break anything (the MAC is not my own and I am
> not at $HOME with Macs)...the
> two commands:
> 
>  defaults write com.apple.desktopservices DSDontWriteNetworkStores true
>  defaults write com.apple.desktopservices DSDontWriteUSBStores true
> 
> need to given as commands at the commandline (the way real UNICES do
> it), aren't they?

Yes. Open Terminal and paste these in.

macOS is POSIX certified: 
https://blog.opengroup.org/2012/07/25/apple-registers-mac-os-x-10-8-mountain-lion-to-the-unix-03-standard/

Re: [gentoo-user] [OT] ...preventing MAC OS from polluting my USBsticks...?

[Andrew Udvare]

> On 2019-01-13, at 08:43, tu...@posteo.de wrote:
> 
> Hi,
> 
> it happens that I use my usbsticks with a Mac. As soon I use the stick
> again with my trusty Linux...I found a lot of added "hidden" files,
> which are shorter and renamed versions of files, which are already
> there...and there are everywere.
> 
> Is there any hack/trick/ to prevent MacOS to write to 
> my usbstick when not instructed to do so?

There are these two defaults settings, and they don't require root to set them:

defaults write com.apple.desktopservices DSDontWriteNetworkStores true
defaults write com.apple.desktopservices DSDontWriteUSBStores true

These are only for stopping Finder from writing .DS_Store files. Anything you 
delete on macOS will still go in .Trash/ at the root of the drive.

-- 
Andrew Udvare

Re: [gentoo-user] Re: mac-fdisk with different block size

[Andrew Udvare]

> On Jan 13, 2019, at 09:24, (Nuno Silva)  wrote:
> 
> On 2019-01-13, Andrew Udvare wrote:
> 
>>> On 2019-01-13, at 07:49, (Nuno Silva)  wrote:
>>> 
>>> I am trying to create an Apple partition map with a block size of 4096
>>> bytes, but I can't find an option to change the block size in mac-fdisk,
>>> which defaults to 512 bytes.
>>> 
>>> Does anybody know of a utility that can create and modify such partition
>>> maps under Gentoo?
>> 
>> 
>> Quick look and it seems that for mac-fdisk the 512 size is hard-coded:
>> 
>> https://github.com/glaubitz/mac-fdisk-debian/blob/bda743065fa2c75a83fec60166bc2e317059ef7a/io.h#L32
> 
> 
> That code appears to be version 0.4 from early 1997. Under Gentoo, the
> README file installed at /usr/share/doc/mac-fdisk-0.1_p18/ mentions 0.4
> as well.
> 
> According to a changelog at apple.com[1], variable block size support
> was added after that, and should be present in version 0.5.

On actual macOS latest, there's a -b option for block size.

Re: [gentoo-user] mac-fdisk with different block size

[Andrew Udvare]

> On 2019-01-13, at 07:49, (Nuno Silva)  wrote:
> 
> I am trying to create an Apple partition map with a block size of 4096
> bytes, but I can't find an option to change the block size in mac-fdisk,
> which defaults to 512 bytes.
> 
> Does anybody know of a utility that can create and modify such partition
> maps under Gentoo?


Quick look and it seems that for mac-fdisk the 512 size is hard-coded:

https://github.com/glaubitz/mac-fdisk-debian/blob/bda743065fa2c75a83fec60166bc2e317059ef7a/io.h#L32

Re: [gentoo-user] Playing video and CPU usage

[Andrew Udvare]

Sent from my iPhone

> On Jan 12, 2019, at 02:12, Dale  wrote:
> Just for giggles, I used VLC for a bit to play a video.  It has a fair
> resolution and is a .mp4.  It uses about the same amount of CPU power as
> Smplayer. I can't tell any difference even with the same video.  It
> appears that it may be something besides the player. 

You have to go into settings and make sure VDPAU is enabled. You need to enable 
the vdpau USE flag. I have it set globally.

Re: [gentoo-user] Playing video and CPU usage

[Andrew Udvare]

> On 2019-01-12, at 00:00, Dale  wrote:
> 
> Howdy,
> 
> As some know, I recently bought a video card.  While not the most modern
> thing, it is a lot faster than my old one.  I have a question tho.  When
> I'm watching TV and playing a video with Smplayer, high resolution or a
> medium resolution, it seems to use a good bit of CPU power.  I notice in
> gkrellm, htop etc that it is using about 20 to sometimes 40 or 50% of
> CPU power.  Yes, I still have the 8 core CPU in here.  In Smplayer, I
> have video driver set to "gl(fast)" but have tried other settings as
> well.  Obviously, some just plain don't work at all.  It causes Smplayer
> to crash.  I did some googling, I think this is the best driver setting
> for my card.  It is nvidia based.  

It's not the best setting. You should avoid SMPlayer (IMHO from my experience 
at this point) and switch to mpv or VLC. You want to enable hardware decoding 
with VDPAU with any player. That card can decode H.264 fully and partially can 
decode MPEG-1/2/VC-1. You will see very little CPU usage with VDPAU enabled. 
The downside is the inability to use software-based filters in real-time, but 
there is a built-in hardware deinterlacer that significantly beats filters like 
Yadif.

-- 
Andrew Udvare

Re: [gentoo-user] @preserved-rebuild how to find out the contents of

[Andrew Udvare]

> On 2018-12-29, at 12:33, Helmut Jarausch  wrote:
> 
> On 12/29/2018 05:03:51 PM, Andrew Udvare wrote:
>> > On Dec 29, 2018, at 10:47, Helmut Jarausch  wrote:
>> >
>> > How can I find out the list of packages in @preserved-rebuild?
>> There's a JSON file in /var/db/portage
> 
> I don't have the folder /var/db/portage here.
> Thanks,
> Helmut
> 
The file you want is /var/lib/portage/preserved_libs_registry

# cat /var/lib/portage/preserved_libs_registry
{
"dev-libs/openssl:0": [
"dev-libs/openssl-1.1.0j",
"62711",
[
"/usr/lib32/libcrypto.so.1.0.0",
"/usr/lib32/libssl.so.1.0.0",
"/usr/lib64/libcrypto.so.1.0.0",
"/usr/lib64/libssl.so.1.0.0"
]
]
}

Re: [gentoo-user] @preserved-rebuild how to find out the contents of

[Andrew Udvare]

> On Dec 29, 2018, at 10:47, Helmut Jarausch  wrote:
> 
> How can I find out the list of packages in @preserved-rebuild?

There's a JSON file in /var/db/portage

Re: [gentoo-user] Is it that hard to add a package, or am I doing wrong?

[Andrew Udvare]

> On 2018-12-19, at 21:24, YUE Daian  wrote:
> 
> Is there anything I can do more?

In your ebuild, remove ./bootstrap and use eautoreconf.

https://devmanual.gentoo.org/eclass-reference/autotools.eclass/

Re: [gentoo-user] Is it that hard to add a package, or am I doing wrong?

[Andrew Udvare]

-- 
Andrew Udvare


> On 2018-12-19, at 21:24, YUE Daian  wrote:
> 
> Hi Gentoo folks,
> 
> Recently I posted a bug report to Gentoo Bugzilla and submitted a
> request to add package Roswell into the package tree.
> 
> https://bugs.gentoo.org/638446


> 
> But...in fact it was not "recent" at all! I submitted the bug one year
> ago and there is literally no news after some point.
> 
> Is there anything I can do more?
> 
> Danny
> 

Re: [gentoo-user] mounting windows 10 partition on linux

[Andrew Udvare]

On 04/12/2018 13:16, the...@sys-concept.com wrote:
> I'm trying to mount windows 10 shared folder on Gentoo home folder but
> I'm getting an error from "smbclinet"

smbclient does not mount. It just connects a CLI client to the server
(see the manpage). The -L argument is for the NetBIOS name alone, not
the UNC. This should work:

smbclient -U OP2 //10.0.0.119/

Since you know the IP, you can skip NetBIOS lookup:

smbclient -I 10.0.0.119 -U OP2 //10.0.0.119/

Assuming you have CIFS built into your kernel, you can use mount as root:

mount -t cifs -o user=OP2 //10.0.0.119 /mount-point

-- 
Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Software for checking CDs and DVDs for errors?

[Andrew Udvare]

> On 2018-12-03, at 23:16, Davyd McColl  wrote:
> 
> Not so much just for verification, but I'd also check out ddrescue. If the 
> tool dumps your media easily, it's probably good. If it struggles, you may at 
> least still have a workable image by the time it is done.

The lasers in older game consoles are very weak and they get even worst with 
age. They don't tend to do well with many scratches. The disc may read in a PC 
with scratches, but then fail to load in the console. I have found PC drives to 
be far more resilient to scratches compared to consoles.

In the case of PlayStation 1, in the interest of space, they went with mode 2 
XA format which makes the discs lack error correction codes. This means if the 
disc fails in a PC it definitely will fail in the console with no way to 
correct.

-- 
Andrew

Re: [gentoo-user] Routing issue with OpenVPN and internal DNS

[Andrew Udvare]

On 03/12/2018 09:49, Michael Orlitzky wrote:
> On 12/3/18 5:55 AM, Andrew Udvare wrote:
>>
>> iptables on server:
>> -A FORWARD -s 10.100.0.0/24 -i tun0 -o enp1s0f0 -m conntrack --ctstate
>> NEW -j ACCEPT
>>
> 
> Is that only forwarding packets for new (i.e. not existing) connections?

Not sure but I do have a rule with using --ctstate ESTABLISHED,RELATED
like yours. I even got rid of the interface argument in case that's a
problem. The box is a router and has 2 NICs going, one for WAN and one
for LAN. enp1s0f0 being the internet, and enp1s0f1 is for 192.168.1.0/24

When I'm connected to the VPN and I'm definitely not on my network, I
can do things like `ssh 192.168.1.xxx` and it works. And HTTP works too.
It's only port 53 that I am having trouble with.

dnsmasq (listening only on enp1s0f1, 192 address) gets the request from
the tun0 interface, which seems to route correctly to the 192 address.
The response that dnsmasq creates (presumably) does not route back to
the originating IP.

Happy to provide any other configuration details and packet dumps if it
helps.

Full iptables (I use a script to reset to sane state, suggestions welcome):

-P INPUT DROP
-P FORWARD DROP
-P OUTPUT ACCEPT
-N DOCKER
-N DOCKER-ISOLATION-STAGE-1
-N DOCKER-ISOLATION-STAGE-2
-N DOCKER-USER
-N SCANS
-A INPUT -p tcp -m tcp ! --tcp-flags FIN,SYN,RST,ACK SYN -m state
--state NEW -j DROP
-A INPUT -f -j DROP
-A INPUT -p tcp -m tcp --tcp-flags FIN,SYN,RST,PSH,ACK,URG
FIN,SYN,RST,PSH,ACK,URG -j DROP
-A INPUT -p tcp -m tcp --tcp-flags FIN,SYN,RST,PSH,ACK,URG NONE -j DROP
-A INPUT -p tcp -m tcp --dport 22 -m state --state NEW -m recent --set
--name BLACKLIST --mask 255.255.255.255 --rsource
-A INPUT -p tcp -m tcp --dport 22 -m state --state NEW -m recent
--update --seconds 10 --hitcount 10 --rttl --name BLACKLIST --mask
255.255.255.255 --rsource -j DROP
-A INPUT -m state --state RELATED,ESTABLISHED -j ACCEPT
-A INPUT -m conntrack --ctstate INVALID -j DROP
-A INPUT -p tcp -m tcp --dport 22 -m conntrack --ctstate NEW,ESTABLISHED
-j ACCEPT
-A INPUT -p tcp -m tcp --dport 80 -m conntrack --ctstate NEW,ESTABLISHED
-j ACCEPT
-A INPUT -p tcp -m tcp --dport 443 -m conntrack --ctstate
NEW,ESTABLISHED -j ACCEPT
-A INPUT -p tcp -m tcp --dport 9222 -m conntrack --ctstate
NEW,ESTABLISHED -j ACCEPT
-A INPUT -p udp -m udp --dport 67 -m conntrack --ctstate NEW,ESTABLISHED
-j ACCEPT
-A INPUT -p udp -m udp --dport 68 -m conntrack --ctstate NEW,ESTABLISHED
-j ACCEPT
-A INPUT -p udp -m udp --dport 53 -m conntrack --ctstate NEW,ESTABLISHED
-j ACCEPT
-A INPUT -p icmp -m icmp --icmp-type 8 -j ACCEPT
-A INPUT -p udp -m udp --dport 33434:33523 -j REJECT --reject-with
icmp-port-unreachable
-A INPUT -s 192.168.1.0/24 -p tcp -m tcp --dport 3306 -m conntrack
--ctstate NEW,ESTABLISHED -j ACCEPT
-A INPUT -s 192.168.1.0/24 -p udp -m conntrack --ctstate NEW,ESTABLISHED
-m udp --dport 137 -j ACCEPT
-A INPUT -s 192.168.1.0/24 -p udp -m conntrack --ctstate NEW,ESTABLISHED
-m udp --dport 138 -j ACCEPT
-A INPUT -s 192.168.1.0/24 -p tcp -m conntrack --ctstate NEW,ESTABLISHED
-m tcp --dport 139 -j ACCEPT
-A INPUT -s 192.168.1.0/24 -p tcp -m conntrack --ctstate NEW,ESTABLISHED
-m tcp --dport 445 -j ACCEPT
-A INPUT -p tcp -m tcp --dport 4242 -m conntrack --ctstate
NEW,ESTABLISHED -j ACCEPT
-A INPUT -s 192.168.1.0/24 -i enp1s0f1 -p udp -m udp --dport 5353 -m
conntrack --ctstate NEW,ESTABLISHED -j ACCEPT
-A INPUT -p udp -m udp --dport 12112 -m conntrack --ctstate
NEW,ESTABLISHED -j ACCEPT
-A INPUT -s 205.171.2.64/32 -p ipv6 -j ACCEPT
-A INPUT -i lo -j ACCEPT
-A FORWARD -j DOCKER-USER
-A FORWARD -j DOCKER-ISOLATION-STAGE-1
-A FORWARD -o docker0 -m conntrack --ctstate RELATED,ESTABLISHED -j ACCEPT
-A FORWARD -o docker0 -j DOCKER
-A FORWARD -i docker0 ! -o docker0 -j ACCEPT
-A FORWARD -i docker0 -o docker0 -j ACCEPT
-A FORWARD -i enp1s0f1 -j ACCEPT
-A FORWARD -i enp2s0 -j ACCEPT
-A FORWARD -i enp1s0f0 -j ACCEPT
-A FORWARD -i br0 -j ACCEPT
-A FORWARD -m conntrack --ctstate RELATED,ESTABLISHED -j ACCEPT
-A FORWARD -s 10.100.0.0/24 -j ACCEPT
-A FORWARD -i tun0 -j ACCEPT
-A OUTPUT -o lo -j ACCEPT
-A DOCKER-ISOLATION-STAGE-1 -i docker0 ! -o docker0 -j
DOCKER-ISOLATION-STAGE-2
-A DOCKER-ISOLATION-STAGE-1 -j RETURN
-A DOCKER-ISOLATION-STAGE-2 -o docker0 -j DROP
-A DOCKER-ISOLATION-STAGE-2 -j RETURN
-A DOCKER-USER -j RETURN
-A SCANS -p tcp -m tcp --tcp-flags FIN,PSH,URG FIN,PSH,URG -j DROP
-A SCANS -p tcp -m tcp --tcp-flags FIN,SYN,RST,PSH,ACK,URG
FIN,SYN,RST,PSH,ACK,URG -j DROP
-A SCANS -p tcp -m tcp --tcp-flags FIN,SYN,RST,PSH,ACK,URG NONE -j DROP
-A SCANS -p tcp -m tcp --tcp-flags SYN,RST SYN,RST -j DROP

iptables -nat --list-rules:

-P PREROUTING ACCEPT
-P INPUT ACCEPT
-P OUTPUT ACCEPT
-P POSTROUTING ACCEPT
-N DOCKER
-A PREROUTING -m addrtype --dst-type LOCAL -j DOCKER
-A OUTPUT ! -d 127.0.0.0/8 -m addrtype --dst-type LOCAL -j DOCKER
-A POSTROUTING -s 172.17.0.0/16 ! -o docker0 -j MASQUERADE
-A POSTROUTING -s 10.100.0.0/24 -j MASQUERADE
-A POSTROUTIN

Re: [gentoo-user] Software for checking CDs and DVDs for errors?

[Andrew Udvare]

On 03/12/2018 17:50, Dale wrote:
> Jack wrote:
> 
> I was wondering if the checksums could be checked?  If the OP can find
> the checksum for the DVD as it comes from the factory, then he could
> check what he has against that.  Question is, is that info even
> available or does it vary over batches of DVDs?  If it varies, it may
> not be possible to test that way even if checksums are available. 
> 
> Just throwing that out there in case checksums are available and it
> could be done that way.  Sort of doubt it tho. 

Checksums do exist. The Redump project is trying to collect all relevant
disc information for every video game ever made, basically.

http://redump.org/disc/57802/

The process to generate this information is very specific and requires
software that isn't made for Linux unfortunately. But the MD5, CRC32,
SHA1 is there which can be usually be checked with `dd /dev/sr0 |
sha1sum` for a basic disc.

PS1 and PS2 games can be checked without special hardware in this case,
but for others, specific hardware is required.

In my case I have a JTAG Xbox 360, so the best way for me to check a
game is to have my Xbox 360 dump the disc contents entirely with an app
that would not be available on a normal Xbox 360. If this fails at any
point it's a bad disc (or it needs cleaning/repair).

-- 
Andrew



signature.asc
Description: OpenPGP digital signature

[gentoo-user] Routing issue with OpenVPN and internal DNS

[Andrew Udvare]

Very confused here, but I feel like I'm missing a route on either the client 
side or the server side. Or it is a firewall rule but that doesn't seem likely.

My OpenVPN server/client config is almost identical to that on the wiki page: 
https://wiki.gentoo.org/wiki/OpenVPN#Configuration

After connecting from another ISP and then connecting to the VPN, I use dig to 
query the internal server:

dig @192.168.1.254 pi.hole

The server can see the DNS request from the client:

 # tcpdump -i tun0 'port 53'
dropped privs to tcpdump
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on tun0, link-type RAW (Raw IP), capture size 262144 bytes
02:21:07.996606 IP 10.100.0.6.50312 > 192.168.1.254.domain: 13933+ [1au] A? 
pi.hole. (36)

Even the dnsmasq log is showing the request:

Dec  3 10:21:45 dnsmasq[1026]: 4744 71.214.144.51/49349 query[A] pi.hole from 
71.214.144.51
^ shows the external IP. Is this normal?

On the client side, dig times out.

So if it can come through client-to-server but won't go back server-to-client, 
where could the issue be?

I even changed the iptables policies for INPUT and FORWARD to allow to see if 
anything would change but the result was the same.

I have HTTP running on the server and accessing via the IP works fine:

 $ curl https://192.168.1.254 -k
...
< Server: nginx
< Date: Mon, 03 Dec 2018 10:22:45 GMT
< Content-Type: text/html
< Content-Length: 574
< Connection: keep-alive
< Keep-Alive: timeout=20

OpenVPN server configuration relevant lines:

# OpenVPN 'virtual' network infomation, network and mask
server 10.100.0.0 255.255.255.0
push "redirect-gateway def1"
push "route 192.168.1.0 255.255.255.0"
push "dhcp-option DNS 192.168.1.254"

Routing table on server:
 $ route -4
Kernel IP routing table
Destination Gateway Genmask Flags Metric RefUse Iface
default xx-xx-xx-xx.0.0.0.0 UG1024   00 enp1s0f0
10.100.0.0  10.100.0.2  255.255.255.0   UG0  00 tun0
10.100.0.2  0.0.0.0 255.255.255.255 UH0  00 tun0
xx.xxx.144.00.0.0.0 255.255.255.0   U 0  00 enp1s0f0
xx-xxx-144-254. 0.0.0.0 255.255.255.255 UH1024   00 enp1s0f0
172.17.0.0  0.0.0.0 255.255.0.0 U 0  00 docker0
192.168.1.0 0.0.0.0 255.255.255.0   U 0  00 enp1s0f1

iptables on server:
-A FORWARD -s 10.100.0.0/24 -i tun0 -o enp1s0f0 -m conntrack --ctstate NEW -j 
ACCEPT

-t nat:
-A POSTROUTING -s 10.100.0.0/24 -j MASQUERADE

On the client (en0 is the wifi, utun4 is created by OpenVPN):
 $ netstat -nrtf inet
Routing tables

Internet:
DestinationGatewayFlagsRefs  Use   Netif Expire
0/110.100.0.5 UGSc   840   utun4
default172.20.10.1UGSc30 en0
10.100.0.1/32  10.100.0.5 UGSc00   utun4
10.100.0.5 10.100.0.6 UHr140   utun4
71.214.144.51/32   172.20.10.1UGSc10 en0
127127.0.0.1  UCS 00 lo0
127.0.0.1  127.0.0.1  UH  5   140998 lo0
128.0/110.100.0.5 UGSc40   utun4
169.254link#9 UCS 00 en0  !
172.20.10/28   link#9 UCS 10 en0  !
172.20.10.1/32 link#9 UCS 20 en0  !
172.20.10.1b2:70:2d:2:18:64   UHLWIir 8   23 en0   1184
172.20.10.3/32 link#9 UCS 00 en0  !
172.20.10.15   ff:ff:ff:ff:ff:ff  UHLWbI  0   18 en0  !
192.168.1  10.100.0.5 UGSc00   utun4
224.0.0/4  link#9 UmCS20 en0  !
224.0.0.2511:0:5e:0:0:fb  UHmLWI  00 en0
239.255.255.2501:0:5e:7f:ff:faUHmLWI  04 en0
255.255.255.255/32 link#9 UCS     00 en0  !

-- 
Andrew Udvare

Re: [gentoo-user] Odd problem with Xorg

[Andrew Udvare]

> On Nov 28, 2018, at 15:09, Daniel Frey  wrote:
> 
> I've attached the two log files.

I would like to see your syslog in both cases; dmesg output.

Re: [gentoo-user] Help with dev-util/cargo blocking dev-lang/rust.

[Andrew Udvare]

On 18/11/2018 00:23, Grant Taylor wrote:
> 
> I'm not completely sure what that means.  I'm guessing you're asking
> about accept_keywords.  I'm mostly (?) running stock amd64.  I have
> added ~amd64 to some packages for various reasons over the 2+ years that
> this install has been around.

I switched fully to ACCEPT_KEYWORDS="~amd64" (make.conf) after running
mixed for a while. These kinds of issues come up too often and I don't
have a lot of time to solve them, plus for my dev machine I just don't
notice stable vs unstable most of the time. If something is truly
unstable from my own experience I will mask that version and downgrade.

My recommendation would be to switch over to ~amd64 for a workstation
generally. On my home server I run stable with a few unmasks, but most
of the time these are packages that have no dependencies.

-- 
Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] systemd/dracut: Correctly boot using flash drive with LUKS key?

[Andrew Udvare]

On 16/11/2018 14:43, Rich Freeman wrote:
> On Fri, Nov 16, 2018 at 12:15 PM Andrew Udvare  wrote:
>>
>> I am not sure if there is a way to move the systemd-cryptsetup@home.service 
>> up the dependency tree once it's working, which would then remove the 
>> mnt-chuan.mount dependency.
>>
> 
> Ok, I did a bit more reading.  You're using the cryptsetup generator
> most likely.  It sets up units to be oneshot+remainafterexit, which
> means they're "active" whenever the LUKS device is mounted (without
> any processes - but they show as active so that you can stop them and
> unmount the device).  It sets the RequiresMountsFor parameter for the
> device the key file is contained on, which makes that mount service a
> Required dependency.  That means that it can't be unmounted while the
> cryptsetup device is in use, and in theory attempting to unmount the
> key file should make systemd attempt to unmount the cryptsetup device
> (though busy filesystems could interfere with that).

So it is a bit strange that /mnt/chuan was considered a dependency just
because of mention in /etc/crypttab. However I found out that the reason
has something to do with the /mnt/chuan entry in /etc/fstab in my real
root, and this is not a necessary line (it is the only entry in the
initrd fstab). I removed the line and now the dependency is still show
with list-dependencies, but it is white instead of red. My system is
still shown as running rather than degraded.

Removing the line from /etc/fstab only partially solves the problem, as
it's not explained what happens with the USB drive once the root is
switched because after that it's not shown to be mounted. I am pretty
sure it's not safely unmounted before the switch, which leaves it in a
strange state requiring fsck. Don't know the best way around this other
than wait till systemd supports the keyscript option in /etc/crypttab.

-- 
Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Help with dev-util/cargo blocking dev-lang/rust.

[Andrew Udvare]

On 17/11/2018 23:53, Grant Taylor wrote:
> On 11/17/2018 07:58 PM, Adam Carter wrote:
> 
> These are the packages that would be merged, in order:
> 
> Calculating dependencies... done!
> [ebuild  N ] dev-util/cargo-0.30.0  USE="-debug -doc -libressl"
> ABI_X86="(64) -32 (-x32)"
> [ebuild  N ] virtual/cargo-1.29.1
> [blocks B  ] dev-util/cargo ("dev-util/cargo" is blocking
> dev-lang/rust-1.30.1-r1)
> 
>  * Error: The above package list contains packages which cannot be
>  * installed at the same time on the same system.
> 
>   (dev-util/cargo-0.30.0:0/0::gentoo, ebuild scheduled for merge) pulled
> in by
>     =dev-util/cargo-0.30.0* required by
> (virtual/cargo-1.29.1:0/0::gentoo, ebuild scheduled for merge)
> 
>   (dev-lang/rust-1.30.1-r1:stable/1.30::gentoo, installed) pulled in by
>     =dev-lang/rust-1.30.1* required by (virtual/rust-1.30.1:0/0::gentoo,
> installed)

It looks like you need to unmask virtual/cargo because you need to have
virtual/cargo 1.30.1. These version numbers have to match. Are you
running a mixed keyworded system?

Once you unmask, you can install what you need for Firefox/Thunderbird
beforehand:

emerge -1 virtual/cargo dev-lang/rust

--
Andrew Udvare



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Help with dev-util/cargo blocking dev-lang/rust.

[Andrew Udvare]

> On 2018-11-17, at 20:16, Adam Carter  wrote:
> 
> > On 2018-11-17, at 19:24, Grant Taylor  
> > wrote:
> > 
> > So, will someone help me fix this cargo / rust blockage so that I can 
> > finish my @world emerge?
> 
> Uninstall dev-util/cargo and emerge -1 dev-lang/rust. dev-lang/rust comes 
> with Cargo. 
> 
> 
> I also had to
> 
>  ln -s /usr/bin/cargo-1.30.1 /usr/bin/cargo
> 
> To get things working, YMMV.

The correct way is to run `eselect rust set 1`. There are other symlinks 
besides cargo.

-- 
Andrew Udvare

Re: [gentoo-user] Help with dev-util/cargo blocking dev-lang/rust.

[Andrew Udvare]

> On 2018-11-17, at 19:24, Grant Taylor  
> wrote:
> 
> So, will someone help me fix this cargo / rust blockage so that I can finish 
> my @world emerge?

Uninstall dev-util/cargo and emerge -1 dev-lang/rust. dev-lang/rust comes with 
Cargo. 

-- 
Andrew Udvare

Re: [gentoo-user] systemd/dracut: Correctly boot using flash drive with LUKS key?

[Andrew Udvare]

> On 2018-11-16, at 08:16, Rich Freeman  wrote:
> 
> On Fri, Nov 16, 2018 at 12:11 AM Andrew Udvare  wrote:
>> 
>> - use key to decrypt root partition (systemd job in the first systemd
>> launched within initramfs)
>> ...
>> Nov 16 00:01:52 limelight systemd[1]: Dependency failed for /mnt/chuan.
>> Nov 16 00:01:52 limelight systemd[1]: mnt-chuan.mount: Job
>> mnt-chuan.mount/start failed with result 'dependency'.
> 
> So, I haven't messed around that much with systemd and removable
> media, but a few things I'd start by looking at:
> 
> systemctl show mnt-chuan.mount
> systemctl list-dependencies --reverse --all mnt-chuan.mount

So this output seems to make sense, and running systemctl stop on the mount 
gave me similar messages about dependencies. I"m not sure there is a way around 
this, as the dependency tree always relies on that mount before 
systemd-cryptsetup@root can work. If unmounted it will be flagged as in invalid 
state.

I am not sure if there is a way to move the systemd-cryptsetup@home.service up 
the dependency tree once it's working, which would then remove the 
mnt-chuan.mount dependency.

Tree from the device:

dev-disk-by\x2...device
● └─mnt-chuan.mount
│ └─...
●   ├─systemd-cryptsetup@home.service
●   │ ├─dev-disk-by\x2duuid-device
│ │   └─...
●   │ ├─dev-mapper-home.device
●   │ └─cryptsetup.target
●   │   └─sysinit.target

systemd-cryptsetup@home.service is red.

-- 
Andrew Udvare

[gentoo-user] systemd/dracut: Correctly boot using flash drive with LUKS key?

[Andrew Udvare]

I have a set up like this to boot:

GRUB - access non-encrypted EFI partition to find kernel, initramfs, etc
initramfs has enough on it to boot minus the encryption key
(Removable) storage must be inserted (UUID matching) to:
- mount to expected point
- use key to decrypt root partition (systemd job in the first systemd
launched within initramfs)

This set up works, but it has one annoying thing coming from systemd
after initramfs is done.

When I take out the flash drive after booting up, I continuously get a
message that the flash drive wants to be mounted again.

Nov 16 00:01:52 limelight systemd[1]:
dev-disk-by\x2duuid-BD3B\x2d03BD.device: Job
dev-disk-by\x2duuid-BD3B\x2d03BD.device/start timed out.
Nov 16 00:01:52 limelight systemd[1]: Timed out waiting for device
Silicon-Power16G CHUAN.
Nov 16 00:01:52 limelight systemd[1]: Dependency failed for /mnt/chuan.
Nov 16 00:01:52 limelight systemd[1]: mnt-chuan.mount: Job
mnt-chuan.mount/start failed with result 'dependency'.
Nov 16 00:01:52 limelight systemd[1]:
dev-disk-by\x2duuid-BD3B\x2d03BD.device: Job
dev-disk-by\x2duuid-BD3B\x2d03BD.device/start failed with result 'timeout'.

I think this is because of the entry in /etc/fstab.

Anyone have a similar issue? And how can I fix this in Dracut?

Thanks
Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] What's with KDE?

[Andrew Udvare]

> On 2018-11-03, at 00:43, Alan Grimes  wrote:
> 
> I have not seen KDE add a single feature or even fix a single bug (while
> letting new bugs pile up) in FIVE YEARS. They version bump their entire
> suite for no reason every five minutes but nothing gets even slightly
> better. KDE has'nt been worth anything since the QT4 cataclysm, 3.59 was
> pretty darn good. =\ Kdevelop3 only needed to be restarted once a
> week... You could binge program with it for an ENTIRE WEEK without
> restarting it... It had some wonderful class inheratance heirarchy
> graphing tools so that you could learn codebases nice and quick and go
> in and refactor them.

There are a lot of new features in KDE but you may not be aware of them or you 
do not care for them. But a lot of KDE 4's changes was to upgrade to KDE 4 
without inheriting the code issues of Qt 3/KDE 3.5 code (KF5 continues this 
with Qt 5). Also with KF5 they are splitting up the code base so that it can be 
used in a more modular form. Prior to KF5, using KDE code in other projects was 
almost impossible without being dependent on the very large kdelibs library 
(which is an issue on other platforms as well where loading that many libraries 
at launch time takes a while).

My main criticism of KDE 4 and onward is the introduction of Akonadi. I've 
never had much success with it especially with KMail.

> How did they make it so that 40% of ordinary zip files I try to open
> with konqueror fail CRC (but work perfectly from the command line)
> 
> It used to have nice large icon mode with previews, and nice small-icons
> in normal mode... Broken too for many months now. =\
 
You have to give up on Konqueror as a file manager. It just doesn't get much 
support for that anymore. I've abandoned Konqueror completely as it's far 
behind as a web browser compared to Firefox/Chrome.

Dolphinpart which runs within Konqueror to make it a file manager is almost the 
same as Dolphin. Although there are things that annoy me in Dolphin, it gets 
more maintenance and a bug report is much more likely to be addressed.

> 
> Akregator crashes all the time if I simply try to close a tab... It went
> about three years without saving anything to disk when it's supposed to
> buffer my RSS feeds. Now it's just crashtastic in the extreme...

Even Google completely dropped their RSS reader. I have my doubts that this app 
gets a lot of maintenance.
> 
> Microsoft must be paying Good Money (tm) to the FS developers to keep
> breaking stuff at this rate. =|

Trouble is afoot! Given the track record of late (like owning GitHub) I don't 
think so.

-- 
Andrew Udvare

Re: [gentoo-user] Nvidia driver surprise

[Andrew Udvare]

> On 2018-10-28, at 10:42, Davyd McColl  wrote:
> 
> Or perhaps go team red. Personally, I've had enough of team green's drivers. 
> My 660ti is still working fine, but if it dies, I think I'm going red. Not 
> saying the grass is 100% greener there (heh) but at least you can't be held 
> hostage with binary blobs. And they're cheaper. /2c
> 
> -d

We're not in good shape on either side here. I would use AMD if they had better 
performance on Linux with the open source driver.

In terms of the long term, I don't see either open source driver supporting 
every device for as long as the drivers are supported. At some point there will 
probably be a split and then eventually those legacy drivers will be removed 
from the kernel due to lack of maintainers.

Andrew

Re: [gentoo-user] Nvidia driver surprise

[Andrew Udvare]

> On Oct 28, 2018, at 09:53, Philip Webb  wrote:
> 
> Perhaps I should buy a more upto-date graphics card.
> 
> Does anyone have comments or suggestions ?

Upgrade to the 1050 or so and you won’t have to upgrade for years. I am still 
on my GTX 980 since 2015.

Re: [gentoo-user] libfilezilla not available ?

[Andrew Udvare]

> On 2018-10-26, at 22:38, tu...@posteo.de wrote:
> 
> Since it is missing on different servers...is the nameing wrong?

Seems to be working now. Probably just mirrors not in sync.

-- 
Andrew

Re: [gentoo-user] Gentoo hardware database

[Andrew Udvare]

> On 2018-10-26, at 05:27, Andrey Ponomarenko  
> wrote:
> 
> Hi,
> 
> The Linux-Hardware.org database has been divided recently into a set of 
> databases, one per each Linux distro. The one for Gentoo is available at:
> 
> https://linux-hardware.org/?d=Gentoo
> 
> Everyone can contribute to the database with the help of 
> https://github.com/linuxhw/hw-probe (various packages for the tool are 
> available: AppImage, Snap, Flatpak, Docker, ebuild 
> http://gpo.zugaina.org/sys-apps/hw-probe, etc.). The tool is intended to 
> simplify collecting of logs necessary for investigating hardware related 
> problems. You need to execute only one simple command to collect all system 
> logs at once:
> 
>sudo hw-probe -all -upload
> 
> Hardware failures are highlighted in the collected logs (smartctl, dmesg, 
> xorg.log). Also it's handy to search for particular hardware configurations 
> in the community and review logs for errors to check operability of devices 
> on board (for some devices this is done automatically by hw-probe — see 
> statuses of devices in a probe).

I submitted my machines. What I find unusual is that one of my machines says 
'failed' for the video card. On that machine, I have not installed 
Nvidia-drivers as it is not necessary (not using X or the card for anything 
except output since the motherboard does not have one built-in). Otherwise not 
sure how it detects failed vs works vs detected.

https://linux-hardware.org/?probe=9cb0ac857a

-- 
Andrew Udvare

Re: [gentoo-user] qt-mismtach?

[Andrew Udvare]

> On 2018-10-24, at 23:03, tu...@posteo.de wrote:
> 
> How can I get around this? Removing linguist is not possible since
> being used by a lot of packages.

You should be able to run:

emerge -1 linguist-tools

This should update linguist-tools to 5.11.2.

-- 
Andrew

Re: [gentoo-user] Per-process-tree memory quotas?

[Andrew Udvare]

> On 2018-10-19, at 23:24, Alan Grimes  wrote:
> 
> How do I do this?

Cgroups were sort of invented for this reason. Yes it requires Systemd.

https://wiki.archlinux.org/index.php/cgroups

I too have 32 GiB of RAM and I'm curious how Chromium acts under a constrained 
environment, or what limits can be placed especially on the CPU. In my 
experience using Chrome inside VMs, it acts very poorly.

-- 
Andrew

Re: [gentoo-user] Is this a genlop bug?

[Andrew Udvare]

> On 2018-10-01, at 05:25, Peter Humphrey  wrote:
> 
> Hello list,
> 
> When I run "genlop -c" is it supposed to include the whole cat/pkg name in 
> its 
> calculations, or just the pkgname? I saw this today:
> 
> $ genlop -c   
> Currently merging 8 out of 16
> * dev-lang/rust-1.29.1 
>   current merge time: 13 minutes and 38 seconds.
>   ETA: 1 minute and 58 seconds.
> $ genlop -t rust | grep second
>   merge time: 29 minutes and 26 seconds.
>   merge time: 4 seconds.
>   merge time: 10 seconds.
>   merge time: 32 minutes and 44 seconds.
> $ genlop -t dev-lang/rust | grep second
>   merge time: 29 minutes and 26 seconds.
>   merge time: 32 minutes and 44 seconds.
> 
> This looks like a bug to me.

It is not a bug. The manpage says it takes name or category/name. In the former 
case it has to match anything named rust. There are two packages named exactly 
'rust': dev-lang/rust and virtual/rust.

Andrew

Re: [gentoo-user] Re: OT: Noteworthy change in the way Chrome logs in

[Andrew Udvare]

> On Sep 24, 2018, at 10:20, Nikos Chantziaras  wrote:
> 
> There is an option for this, actually. Go to:
> 
>  chrome://flags/#account-consistency
> 
> and set it to "Disabled."
> 

Flags come and go and they are not intended for use by regular users. Based on 
the fact that the flag can be enabled automatically via random field testing, I 
assume the flag will disappear soon in the future, and the feature will be on. 
This is the price of using Chrome unfortunately.

Andrew

Re: [gentoo-user] OT: Noteworthy change in the way Chrome logs in

[Andrew Udvare]

> On 2018-09-24, at 06:29, Mick  wrote:
> 
> 
> Unless I understood this wrong, 'Google Chrome' will not be able to perform 
> this (dis)service, unless you have enabled the "Offer to save passwords" 
> feature AND|OR you use Chrome to sign in to an Alphabet related website.

This is not necessarily true because AFAIK it works by taking the session 
generated from within Google's main sign on page and applying that session to 
the browser.

> 
> Is Chromium also affected by this?

Yes.

-- 
Andrew Udvare

Re: [gentoo-user] OT: Noteworthy change in the way Chrome logs in

[Andrew Udvare]

> On 2018-09-24, at 02:47, Adam Carter  wrote:
> 
> From: 
> https://blog.cryptographyengineering.com/2018/09/23/why-im-leaving-chrome/
> 
> "A few weeks ago Google shipped an update to Chrome that fundamentally 
> changes the sign-in experience. From now on, every time you log into a Google 
> property (for example, Gmail), Chrome will automatically sign the browser 
> into your Google account for you. It’ll do this without asking, or even 
> explicitly notifying you. (However, and this is important: Google developers 
> claim this will not actually start synchronizing your data to Google — yet. 
> See further below.)"

I use the Google features, with many Google accounts including ones I have no 
control of like work accounts. I have many Chrome profiles to avoid the 
confusion that Google is trying to solve with this feature. I wish they would 
solve it in a better way like allowing multiple user accounts open in one 
session complete with multi-account Gmail and others. I can understand that 
other users may not ever figure out to create separate profiles.

I can understand not wanting this feature but unfortunately Google does not 
believe in options when it comes to Chrome. They are always removing or 
obscuring them. They seem to have a feeling that if there are options (more 
than one good default setting), something is wrong.

-- 
Andrew Udvare

Re: [gentoo-user] march cflag for Intel Pentium Dual-Core E2160 SLA8Z Malay processor

[Andrew Udvare]

> On 2018-09-19, at 17:54, Jack  wrote:
> 
> Unless I'm missing something, I'm surprised nobody has mentioned one 
> potential problem.  Unless you are careful, code you compile for an AMD may 
> not run on an Intel.  If you are just going to compile binary packages to 
> move and install, this might be OK (assuming you don't trip over trying to 
> run test during a build.) You should be able to come up with settings so 
> compiled code will run on either architecture, but I have no idea what the 
> side effects might be, such as larger and/or less efficient code.  I almost 
> wonder if this doesn't count as cross-compiling, even though it is not a 
> totally incompatible architecture.  Hopefully someone can either expand on 
> this, or tell me why I'm barking up the wrong tree.

I've dealt with this problem before. The key is flags like -mtune=generic and 
not using -march=native. Optimisations should still be okay.

Andrew

Re: [gentoo-user] wxPython/wxWidgets release number mismatch

[Andrew Udvare]

> On 2018-09-18, at 18:50, Mart Raudsepp  wrote:
> 
> 
> Yes, it'd be nice if we had wxpython-4. Help welcome. Also help welcome
> in having wxpython-3 use wxGTK:3.0-gtk3 or have a separate SLOT for
> that itself as well.

Curiosity: what is the reason for wxGTK:3.0 and wxGTK:3.0-gtk3 ?

Andrew

Re: [gentoo-user] wxPython/wxWidgets release number mismatch

[Andrew Udvare]

> On 2018-09-17, at 19:51, Grant Edwards  wrote:
> 
> Sounds like a mess.  The Gentoo stable versions _seem_ to work OK
> together at the moment (for the apps I've tried), but the warning
> every time you run one doesn't inspire confidence.

That would be enough for me to stop using packages that rely on it. 2014 is a 
long time ago for any major package like this. I would expect crashes and 
instability with such apps. If I really need it, I would find another way to 
get the latest versions of wxPython and wxGTK installed. Maybe an overlay has 
that set up. Or you can build a local copy of the packages and use them that 
way.

Andrew

Re: [gentoo-user] Re: wxPython/wxWidgets release number mismatch

[Andrew Udvare]

On 9/17/18 5:02 PM, Grant Edwards wrote:
> 
> It wants to re-install wxpython-3.0.2.0, wxGTK-3.04 and
> wxGTK-304-r300.  I've already done that a few times, but I answered
> 'y' anyway and let it reinstall them again.  It didn't help:
> 
> $ python -c "import wx"
> 
> /usr/lib64/python2.7/site-packages/wx-3.0-gtk2/wx/_core.py:16629: 
> UserWarning: wxPython/wxWidgets release number mismatch
>   warnings.warn("wxPython/wxWidgets release number mismatch")
> 

It needs to be version bumped. https://bugs.gentoo.org/632602

The current version of wxPython is actually 4.0.x and is not compatible
with wxGTK 3.0.4. wxGTK needs to be bumped as well
https://bugs.gentoo.org/577030 but there are some breaking changes in
3.1 vs 3.0.

wxPython 3.0.2.0 is considered 'classic' and was released in 2014.

I tried to use a virtualenv with system package access and tried `pip
install wxpython` but the oldest sane version you can go back to 4.0.3
which won't build with 3.0.4 version of wxGTK.

This discrepancy between the two does not look like it will be fixed any
time soon.

Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] wxPython/wxWidgets release number mismatch

[Andrew Udvare]

On 9/17/18 3:48 PM, Grant Edwards wrote:
> Does anybody have any idea what the below is trying to tell me?
> 
> _WHAT_ two compenents are mismatched?
> 
> WTF is the point of printing a "release number mismatch" warning
> without tell the user which two components are mismatched?
> 
> You might as well print out
> 
> WARNING: something might or might be wrong!
> 
> I've reinstalled wxpython, pygtk, and wxGTK, and it didn't help.

I think you need to try this:

revdep-rebuild -L wx_baseu-3.0 -- --ask

I had an issue where Audacity and a few others could not start due to
version mismatches. Got an issue where the ABIs were different. After
running the above command and reinstalling, these apps started working
again.

Calculating dependencies... done!
[ebuild   R] x11-libs/wxGTK-3.0.4
[ebuild   R] x11-libs/wxGTK-3.0.4-r300
[ebuild   R] app-editors/wxhexeditor-0.24
[ebuild   R] games-emulation/pcsx2-1.4.0
[ebuild   R] app-arch/p7zip-16.02-r4
[ebuild   R] net-ftp/filezilla-3.36.0
[ebuild   R] media-video/mediainfo-18.08.1
[ebuild   R] sci-visualization/gnuplot-5.2.4
[ebuild   R] media-sound/audacity-2.2.2
[ebuild   R] games-emulation/dolphin-5.0

I would presume wxPython will be in your list.

Andrew



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] pycharm-community vs pycharm-professional

[Andrew Udvare]

> On Sep 15, 2018, at 18:31, gevisz  wrote:
> 
> In short: What is the difference between pycharm-community and
> pycharm-professional Gentoo packages.

Pretty sure one is completely free and limited in functionality and 
professional requires a license from JetBrains.

If you have a license, install pycharm-professional.

Andrew

Re: [gentoo-user] Gentoo on Apple TV 1

[Andrew Udvare]

On 8/31/18 2:41 PM, Mick wrote:
> 
> What I have not fathomed yet is how to compile into the mach_kernel the 
> vmlinuz and initrd the boot.efi uses to boot linux.  :-/
> 

(Note that I am making assumptions that the Apple TV 1st gen can be
treated kind of like a Mac.)

You probably should try rEFInd to help. You can get rid of it once you
are comfortable. rEFInd can be avoided:

https://glandium.org/blog/?p=2830

This is what I used years back on a MacBook Pro but I was not successful
in getting an EFI stub to boot correctly. The issue was a bug with USB
2/3 initialisation or something at the time in the kernel, which you
probably won't run into. I had to use the BIOS emulation which you might
have the ability to do. So it was rEFInd -> BIOS emulation (calls it
Windows) -> LILO (GRUB didn't work) and then Linux.

Here is what it looked like (holding C at boot time):
https://i.imgtc.com/jjBY8AF.jpg (OS is macOS, "Windows" CD in the
picture was just Gentoo live CD).

Your problem can be made simpler if you have a) no desire to dual-boot
and b) no disk encryption. This would mean you only have your VFAT
partition for EFI and your main partition.

-- 
Andrew



signature.asc
Description: OpenPGP digital signature