Re: [gentoo-user] Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
On Sat, Mar 28, 2020 at 12:13 PM Mark Knecht wrote: > > > > On Sat, Mar 28, 2020 at 8:41 AM Mike Gilbert wrote: > > > > On Sat, Mar 28, 2020 at 5:06 AM wrote: > > > ...because I have experienced neither any restrictions nor any > > > fragilities in the last 12 years of using a MTB setup with that > > > "MS-DOS" setup (technically it is more a MTB-setup. MSDOS refers > > > to V/FAT, which needs to be used with UEFI in the first partition > > > as far as I know...isn't it?) > > > > > > In short: Never chance a runnig system. > > > > > > But let us better discuss on a more technical level, since I asked > > > for technical help and not for more rethorical given question as > > > an answer. > > > > Here's a technical reason to prefer GPT universally: it stores 2 > > copies of the partition table, one at the start of the disk and one at > > the end of the disk. If either copy gets destroyed, you should be able > > to recover the partition table from the other copy. > > Mike, >What's the process for doing that in general? I assume it's fundamentally > to copy the one at the end (what tools) and then to place it back at the > beginning? (what tools? The top hit on Google is this: https://www.rodsbooks.com/gdisk/repairing.html It documents recovery options using the "gdisk" tool (sys-apps/gptfdisk).
[gentoo-user] Re: Courier-Authlib and S.A.S.L.
On Sat, Mar 28, 2020 at 19:23:24 +, Andreas Stiasny wrote: > When I compare the output of "grep -i sasl /etc/postfix/main.cf" to your > configuration the only noticeable difference is that you don't have the line > > smtpd_sasl2_auth_enable = yes Seems the T.L.S.\ encryption was the culprit. Although all the S.A.S.L.\ options were configured correctly, `telnet` wasn't reporting it as T.L.S.\ was optional in general, but required for authentication. smtpd_tls_security_level = may smtpd_tls_auth_only = yes Changing `smtpd_tls_security_level` to `encrypt` seems to fix the issue, such that a STARTTLS is now required for most actions aside from HELO/EHLO. According to [1], the `sql_select: dummy` is only required if cyrus-sasl is built with MySQL/PostgreSQL support, which in this case is not required due to the help of courier-authlib. Thanks for your help. It seems that I also forgot to add `smtpd_sasl_auth_enable = yes` for some reason *; the name isn't exactly cryptic! [1] Gentoo Forums, https://forums.gentoo.org/viewtopic-t-502523.html * Despite me using S.A.S.L. 2 (/etc/sasl2), Postfix doesn't seem to recognise * the `smtpd_sasl2_auth_enable` option, only `smtpd_sasl_auth_enable`. -- Ashley Dixon suugaku.co.uk 2A9A 4117 DA96 D18A 8A7B B0D2 A30E BF25 F290 A8AA signature.asc Description: Digital signature
Re: [gentoo-user] any experience with wayland?
Top posting because the previous parts of the thread are not really relevant, except perhaps for context of how I got here. Short version is I initially had trouble launching plasma under wayland, but eventually figured out (got a clue) how to do so without invoking startx. At the time, I only did so to test a few specific issues with one program. Recently, I decided to give wayland another try. In general, it seems pretty good (i.e., for me few noticeable differences from X) but I do have a few issues I'd love to resolve. First, I have two monitors, and wayland persists in starting with them "backwards." I use systemsettings5 display configuration and drag them to the right position, but I have to do so every time I start a new session. (At some point, I'll just switch the cables to the two monitors, but I would think this setting would hold across sessions. Am I missing something, or is there a bug somewhere that it uses the default every session? Next, I've had a few cases where I come back to my computer in the morning, and find it sitting at the command line. I haven't noticed anything in any log files that looks suspicious to me. I'm open to any suggestions on how to track down the cause if it happens again - but it's been over a week. My current issue is that VirtualBox is segfaulting on startup, and I want to try it in X to see if that's the issue. (In searching for that problem, I found a virtualbox bug that seemed to imply VB will eventually work with wayland, but not yet.) The problem I'm having is that if I start Xorg, it does start, but with no background and no taskbar. Gkrellm is running, presumably started as it was running when I exited the last session, xwin_wayland is running, as is plasma (didn't write down exactlly which binary). I can run konsole with "konsole --display :0.0" from a different virtual terminal, and can launch programs from there (once I "export DISPLAY=:0.0) and at one point even got a taskbar to start (plasmashell, perhaps?). However, it's clear stuff is still missing, as invoking "leave" either from the start menu or right clicking on the desktop has no effect, and the best way I've found to stop the session is to issue "killall xinit" from the other virtual terminal. I start wayland with an alias for ". /home/jack/.winitrc 1>xwlog 2>xwerr" and .winitrc sets some ENV vars, and ends with "dbus-launch --exit-with-session startplasma-wayland". I start X with "startx 1>xxlog 2>xxerr" and the last line of .xinitrc is "exec dbus-launch --sh-syntax --exit-with-session startplasma-x11". Thanks for any pointers of where I need to dig deeper or what config file I've likely got messed up. Jack --- On 2020.01.09 12:57, Jack wrote: Thanks. I think my underlying problem was that all the commands to launch wayland (variants on "dbus-launch startplasma-wayland") were all very close to the last line of .xinitrc when launching X, so I just made changes there, but stupidly didn't recognize that I needed to issue that line directly, not indirectly through startx. Thanks for the necessary clue. I obviously still have tuning to do, but I do have a functioning desktop. (I also don't use any DM, and am happy I was able to get this working without needing to add that layer.) On 2020.01.09 12:40, Franz Fellner wrote: Look here for some hints: https://github.com/elogind/elogind/issues/61 Not everything should apply as you are using Gentoo, but I think there might be a solution for you (pam configuration might be a first step). Unfortunately I can't help with elogind as I'm using systemd and there I have no issues with starting a wayland desktop. Wayland won't work ON TOP OF X11. You can start X applications within Wayland using XWayland (will be done automatically). But trying to start wayland through startx will fail. Am Do., 9. Jan. 2020 um 19:31 Uhr schrieb Mick : > On Thursday, 9 January 2020 17:19:18 GMT Jack wrote: > > On 2020.01.09 11:38, Franz Fellner wrote: > > > Am Do., 9. Jan. 2020 um 18:35 Uhr schrieb Jack < > > > > > > ostrof...@users.sourceforge.net>: > > > > Based on various wiki and forum posts, > > > > I'm using "dbus-run-session startplasma-wayland" as the last line > > > > > > in > > > > > > > .xinitrc, and launching with startx. > > > > > > I stopped reading here. > > > Please think about that again, especially what wayland was meant to > > > REPLACE! > > > > So if you think that's wrong/outdated/inappropriate - please suggest an > > alternative. > > > > I know the ultimate goal is for Wayland to replace xorg-server, but > > everything I have read so far indicates that currently it is really > > just the compositor, and in the case of KDE, that compositing is > > incorporated into kwin (launched as kwin_wayland instead of kwin_x11.) > > I have seen no instructions anywhere
Re: [gentoo-user] laptop closure should not rfkill
On 2020-03-28 11:07, Michael wrote: On Saturday, 28 March 2020 09:43:56 GMT J. Roeleveld wrote: On Friday, March 27, 2020 3:18:47 PM CET n952162 wrote: I want my wireless to continue when I close my laptop lid. Does anyone have a clue what I have to set to make that happen? I've followed it this far: /etc/acpi/events/lm_lid.sh /etc/acpi/actions/lm_lid.sh /lib/udev/lmt-udev (laptop mode tools?) /usr/sbin/laptop_mode /etc/laptop-mode/conf.d /etc/laptop-mode/laptop-mode.conf has this (I don't find any relevant CONTROL_*): /# "Laptop mode" is the mode in which laptop mode tools makes the computer// //# consume less power. This includes the kernel "laptop_mode" feature, which// //# allows your hard drives to spin down, as well as various other settings which// //# can be tweaked by laptop mode tools. You can enable or disable all of these// //# settings using the CONTROL_... options further down in this config file.// // # ##// // //...// // //#// //# Enable laptop mode when the laptop's lid is closed, even when we're on AC// //# power? (ACPI-ONLY)// //#// //ENABLE_LAPTOP_MODE_WHEN_LID_CLOSED=0// / Would it be okay to just *remove* /etc/acpi/*/lm_lid.sh? checking the git-repository for laptop-mode-tools, I see the following file with a change to "disable networking by default": https://github.com/rickysarraf/laptop-mode-tools/blob/lmt-upstream/etc/lapto p-mode/conf.d/ethernet.conf IOW, I would suggest checking the file: /etc/laptop-mode/conf.d/ethernet.conf -- Joost Most desktops have power management facilities from which the behaviour of various inputs (e.g. lid switch) and configured actions can be finely tuned. I mention it here in the off chance it wasn't already explored as an option. Yes, thank you. That's what we're doing, configuring acpi.
Re: [gentoo-user] Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
On 28/03/2020 06:19, tu...@posteo.de wrote: From what I read on the internet: Everything bigger than 2TB needs to be GPT-formatted. Is there anything better than gparted for that job? gdisk? fdisk? Basically, I think pretty much all of the popular linux utilities have been updated. You could format it with an MBR, but that would only allow you to use 2GB. Cheers, Wol
Re: [gentoo-user] Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
> The BIOS of any machine old enough to be using FAT-style MBR cannot cope with > anything newer, so if you have one, you're stuck with it. The BIOS does not care about partitions at all: it just loads and executes whatever is on the first sector of the disk. As long as you are using a bootloader that understands GPT (such as GRUB2), you can BIOS-boot Linux from a GPT disk just fine. andrea
Re: [gentoo-user] Courier-Authlib and S.A.S.L.
On 28.03.20 07:45, Oliver Dixon wrote: I've been attempting to create a Virtual Mail Server using Postfix, MySQL, and Courier, however I've hit a slight brick wall when dealing with the S.M.T.P.\ authentication. As advised at [1], is is generally preferable to use the unified Courier-Authlib interface when interacting with the database, as opposed to having Courier-Authlib *and* cyrus-sasl attaining direct access. Your setup seems to be very similar to mine (Postfix, MySQL, Courier-IMAP, rspamd, clamav). When I compare the output of "grep -i sasl /etc/postfix/main.cf" to your configuration the only noticeable difference is that you don't have the line smtpd_sasl2_auth_enable = yes In /etc/sasl2/smtpd.conf I do not have the line with sql_select, everythig else is the same. Andreas
Re: [gentoo-user] Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
On Sat, 28 Mar 2020 11:40:49 -0400, Mike Gilbert wrote: > > ...because I have experienced neither any restrictions nor any > > fragilities in the last 12 years of using a MTB setup with that > > "MS-DOS" setup (technically it is more a MTB-setup. MSDOS refers > > to V/FAT, which needs to be used with UEFI in the first partition > > as far as I know...isn't it?) The partition table design itself is from MS-DOS, it predates VFAT by many years. > > > > In short: Never chance a runnig system. But it's not a running system because you haven't set it up yet. > > But let us better discuss on a more technical level, since I asked > > for technical help and not for more rethorical given question as > > an answer. > > Here's a technical reason to prefer GPT universally: it stores 2 > copies of the partition table, one at the start of the disk and one at > the end of the disk. If either copy gets destroyed, you should be able > to recover the partition table from the other copy. That's the fragility I was referring to. the restriction is the limit of four partitions unless you add a kludge that makes the partitioning information even more susceptible to damage. -- Neil Bothwick Despite the cost of living, have you noticed how it remains so popular? pgp_8ZQAowKut.pgp Description: OpenPGP digital signature
Re: [gentoo-user] Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
On Sat, Mar 28, 2020 at 8:41 AM Mike Gilbert wrote: > > On Sat, Mar 28, 2020 at 5:06 AM wrote: > > ...because I have experienced neither any restrictions nor any > > fragilities in the last 12 years of using a MTB setup with that > > "MS-DOS" setup (technically it is more a MTB-setup. MSDOS refers > > to V/FAT, which needs to be used with UEFI in the first partition > > as far as I know...isn't it?) > > > > In short: Never chance a runnig system. > > > > But let us better discuss on a more technical level, since I asked > > for technical help and not for more rethorical given question as > > an answer. > > Here's a technical reason to prefer GPT universally: it stores 2 > copies of the partition table, one at the start of the disk and one at > the end of the disk. If either copy gets destroyed, you should be able > to recover the partition table from the other copy. Mike, What's the process for doing that in general? I assume it's fundamentally to copy the one at the end (what tools) and then to place it back at the beginning? (what tools? Thanks, Mark
Re: [gentoo-user] Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
On Sat, Mar 28, 2020 at 5:06 AM wrote: > ...because I have experienced neither any restrictions nor any > fragilities in the last 12 years of using a MTB setup with that > "MS-DOS" setup (technically it is more a MTB-setup. MSDOS refers > to V/FAT, which needs to be used with UEFI in the first partition > as far as I know...isn't it?) > > In short: Never chance a runnig system. > > But let us better discuss on a more technical level, since I asked > for technical help and not for more rethorical given question as > an answer. Here's a technical reason to prefer GPT universally: it stores 2 copies of the partition table, one at the start of the disk and one at the end of the disk. If either copy gets destroyed, you should be able to recover the partition table from the other copy.
Re: [gentoo-user] Dolphin only allowing a single instance after update
J. Roeleveld wrote:
> On Thursday, March 26, 2020 9:47:37 AM CET Dale wrote:
>> J. Roeleveld wrote:
>>> On Thursday, March 26, 2020 7:46:06 AM CET Dale wrote:
J. Roeleveld wrote:
> Dale (and others),
>
> I found the location where this setting is.
> you need to go to "Configure Dolphin" (In my case, I had to add this to
> the
> toolbar as I am missing the drop down menu).
> Then in the "Startup" part, I see "Open new folders in tabs" ticked.
> If you disable that, you get your folders back in new windows.
>
> --
> Joost
I don't see "open new folders in tabs" listed here. Hmmm. What could
cause that. USE flag maybe??
[ebuild R] kde-apps/dolphin-19.12.3:5::gentoo USE="handbook
-activities -debug -semantic-desktop -test"
Well, nothing there that is obvious. I wonder why I don't have that
option here?? It makes sense to have it set that way.
Dale
:-) :-)
>>> This is mine:
>>> [I] kde-apps/dolphin
>>>
>>> Available versions: (5) 19.08.3^t{tbz2} ~19.12.3^t
>>>
>>>{activities debug +handbook semantic-desktop test}
>>>
>>> Installed versions: 19.08.3(5)^t{tbz2}(08:28:36 PM 01/16/2020)
>>>
>>> (activities handbook semantic-desktop -debug -test)
>>>
>>> Homepage:
>>> https://kde.org/applications/system/org.kde.dolphin
>>>
>>> https://userbase.kde.org/Dolphin
>>>
>>> Description: Plasma filemanager focusing on usability
>>>
>>> Due to the current situation, I am slower with updates. Hope it didn't get
>>> removed.
>>> I do have activities and semantic-desktop enabled on mine.
>>>
>>> I attached a screenshot where I found the setting.
>>>
>>> I also found it in the config-file:
>>>
>>> FILE: ~/.config/dolphinrc
>>>
>>> SECTION:
>>> ---
>>>
>>> [General]
>>> OpenExternallyCalledFolderInNewTab=false
>>> Version=200
>>> ViewPropsTimestamp=2016,6,16,18,3,14
>>>
>>> ---
>>>
>>> If you can't find the setting in the menus, maybe this will work.
>>>
>>> --
>>> Joost
>> Interesting. I looked in my dolphinrc config file and the
>> OpenExternally* line is not there at all. I added it to the file, just
>> for giggles. I'll logout Sunday or Monday and then try to figure out
>> what I can use to open a tab and see if that setting works or not.
>>
>> My screen looks just like yours except for the missing bit I need. Just
>> like me to have that huh? lol
>>
>> Dale
>>
>> :-) :-)
> Previously in the thread someone posted a patch to roll-back this change.
> I did not apply that myself, it could be that that roll-back-patch is
> preventing this option from appearing/working.
>
> Do you still have that patch applied on your system?
>
> Also, changing the tick-box worked immediately on my system. Changing the
> config-file directly might require ALL dolphin instances from being killed.
>
> --
> Joost
I do still have the patch there and since there was no error, I assume
it applied. I'll move it to another location and recompile later on. I
do my updates tomorrow anyway.
Dale
:-) :-)
Re: [gentoo-user] Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
On Saturday, 28 March 2020 10:43:50 GMT Michael wrote: > On Saturday, 28 March 2020 06:19:56 GMT tu...@posteo.de wrote: > > Since there is A LOT of stuff to read about UEFI, MTBR and hybrid > > systems I want to sort out, what I need to read in advance. > > What is "MTBR"? > > Mean Time Between Repair? :-/ Motor Torpedo Boat, to those old enough to remember. :) --->8 > What I would consider reasonable in the year of 2020 is to use GPT, which is > a more versatile and useful structure for partitioning disks, without the > restrictions and deficiencies of MBR tables. On the other hand if you have > a really old disk which you are not planning to repartition, I would > consider it reasonable to leave it alone with its existing MBR table and > get on with more important/interesting tasks. The BIOS of any machine old enough to be using FAT-style MBR cannot cope with anything newer, so if you have one, you're stuck with it. -- Regards, Peter.
[gentoo-user] Re: Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
On 2020-03-28, tu...@posteo.de wrote: > The SSD: > > 1.) From the pure technical point of view: Is it possible to > format the SSD with a good ole MTB and boot the system > from it and mount the hardisc (GPT) as usual? Yes. From that point of view, there's no functional difference between SDD and spinning platters. > 2.) Is this reasonable? Using two different label schemes seems overly complex. I'd use GPT for both, just to keep things consistent. -- Grant
Re: [gentoo-user] RYZEN 5: Hyperthreading or no hyperthreading...
On 03/28 05:59, Mark Knecht wrote:
> On Fri, Mar 27, 2020 at 10:58 PM wrote:
> >
> > On 03/27 11:51, Mark Knecht wrote:
> > > On Fri, Mar 27, 2020 at 11:11 AM wrote:
> > > >
> > > > On 03/27 06:04, Andrea Conti wrote:
> > > > > Hello,
> > > > >
> > > > > > Thread(s) per core: 1 <
> > > > > > Does my CPU hyperthread?
> > > > >
> > > > > Definitely not.
> > > > >
> > > > > Your kernel config is fine, chances are hyperthreading (aka "SMT
> mode")
> > > is
> > > > > disabled in your BIOS settings.
> > > > >
> > > > > andrea
> > > > >
> > > >
> > > > Hi Andrea,
> > > >
> > > > I checked that: The BIOS setting was set to use hyperthreading.
> > > >
> > > > But "Number of cores" was set to six. I changed that to 12 and
> > > > Voila! I got two threads per core.
> > > >
> > > > I think "Number of cores" is a little misleading, since there
> > > > are six physical cores (not threads) with a RYZEN 5.
> > > >
> > > > I feeling not that comfortable with this solution.
> > > >
> > > > Is there any way to check for the validity of this setting
> > > > beside a tool, which prints a "2" after the word "threads" ;) ?
> > > >
> > > > Cheers!
> > > > Meino
> > > >
> > > >
> > >
> > > cat /proc/cpu should give info for each thread. I've been running an i7
> 980
> > > Extreme processor @3.33GHz here at home for about 12 years or so. It's 6
> > > cores but shows 12 processors on both Gentoo and now Kubuntu.
> > >
> > > I generally run top and then hit '1' and 'z'. You can watch what
> percentage
> > > each core/thread is using.
> > >
> > > Time a BIG compile job twice, once with each kernel. If it's working
> you'll
> > > measure a significant difference in time. Note that it won't be 2x as
> > > you'll also be limited by disk read/write throughput, but you'll know
> it's
> > > basically working.
> > >
> > > On Gentoo make sure you're compile settings in (I think make.conf - I no
> > > longer run Gentoo much) are set to take advantage of all your cores and
> not
> > > limited to something smaller. Also watch overheating when using more
> > > cores/threads. On older PCs like mine when you possibly have dust in CPU
> > > coolers might not be as efficient as when they are new.
> > >
> > > HTH,
> > > Mark
> >
> > Hi Mark,
> >
> > thank you for your explanations! :)
> >
> > /proc/cpu doesn't exist on my systemmay be you are referring to
> > /proc/cpuinfo?
> >
> > The problem was caused by a kernel misconfiguration by me.
> >
> > In the kernel setup there is a setting "Number of cores" which
> > I had set to six ... since my CPU has 6 physical core.
> >
> > Setting this to twelve (and blurring the syntactically border between
> > threads and cores thereby...) gives me twelves cores in top, htop
> > and such and (as an example) compiling the kernel is faster -
> > so it is not a display gimmick only.
> >
> > I think "Number of cores" is a misnomer...or am I wrong?
> >
> > Cheers!
> > Meino
> >
>
> Meino,
>Yes, /proc/cpuinfo. Sorry.
>
>Well yes, I guess the 'Number of cores' is a misnomer if you're trying
> to equate the language in the kernel against Intel/AMD marketing data for
> physical cores. 6 physical cores with or without hyperthreading is still 6
> physical cores. However 6 physical cores (my processor) _WITH_
> hyperthreading enabled is 12 _LOGICAL_ cores which is more what I think the
> kernel verbiage is about. Semantics I suppose.
>
>I'm glad you found it wasn't a gimmicky number. It really does work,
> within the limits of the hardware being able to figure out what one thread
> should be fetching or writing while the other thread is computing. It's not
> a perfect 2:1 like 12 physical cores might be, but it's a lot less silicon
> and therefore a lot less expensive.
>
> Cheers,
> Mark
Hi Mark,
In the meanwhile I found "glance" and installed it, which is the
bazooka-out-of-the-box-no-configuration terminal-brethren of "conky"
:)
Enough plugins enabled (which come with it preinstalled), you can
watch in realtime, what each core/thread is doing right now...nearly. Big
Brother for the sustem with no bad intention in mind. I am only
curious :)
And you get your sensors diplayed, the workload of your GPU (nvidia in
my case), all processes and lot lot more.
The faster the CPU gets (my previous PC was 12 years old...), the
more the peripheral devices are becoming show stoppers ("stoppers"
in the barest truth of its meaning).
Unfortunatelu the SSD I ordered is in status "ready for delivery"
since 23.03.2020coronayou know...
And with 12 cores enabled on a recent CPU and running for example a
bigger update via emerge (enabled for 12 threads of course)
all the cores are simply waiting a lot
faster..for the harddisc :)
Thanks for your help -- stay healthy!
Cheers!
Meino
Re: [gentoo-user] RYZEN 5: Hyperthreading or no hyperthreading...
On Fri, Mar 27, 2020 at 10:58 PM wrote: > > On 03/27 11:51, Mark Knecht wrote: > > On Fri, Mar 27, 2020 at 11:11 AM wrote: > > > > > > On 03/27 06:04, Andrea Conti wrote: > > > > Hello, > > > > > > > > > Thread(s) per core: 1 < > > > > > Does my CPU hyperthread? > > > > > > > > Definitely not. > > > > > > > > Your kernel config is fine, chances are hyperthreading (aka "SMT mode") > > is > > > > disabled in your BIOS settings. > > > > > > > > andrea > > > > > > > > > > Hi Andrea, > > > > > > I checked that: The BIOS setting was set to use hyperthreading. > > > > > > But "Number of cores" was set to six. I changed that to 12 and > > > Voila! I got two threads per core. > > > > > > I think "Number of cores" is a little misleading, since there > > > are six physical cores (not threads) with a RYZEN 5. > > > > > > I feeling not that comfortable with this solution. > > > > > > Is there any way to check for the validity of this setting > > > beside a tool, which prints a "2" after the word "threads" ;) ? > > > > > > Cheers! > > > Meino > > > > > > > > > > cat /proc/cpu should give info for each thread. I've been running an i7 980 > > Extreme processor @3.33GHz here at home for about 12 years or so. It's 6 > > cores but shows 12 processors on both Gentoo and now Kubuntu. > > > > I generally run top and then hit '1' and 'z'. You can watch what percentage > > each core/thread is using. > > > > Time a BIG compile job twice, once with each kernel. If it's working you'll > > measure a significant difference in time. Note that it won't be 2x as > > you'll also be limited by disk read/write throughput, but you'll know it's > > basically working. > > > > On Gentoo make sure you're compile settings in (I think make.conf - I no > > longer run Gentoo much) are set to take advantage of all your cores and not > > limited to something smaller. Also watch overheating when using more > > cores/threads. On older PCs like mine when you possibly have dust in CPU > > coolers might not be as efficient as when they are new. > > > > HTH, > > Mark > > Hi Mark, > > thank you for your explanations! :) > > /proc/cpu doesn't exist on my systemmay be you are referring to > /proc/cpuinfo? > > The problem was caused by a kernel misconfiguration by me. > > In the kernel setup there is a setting "Number of cores" which > I had set to six ... since my CPU has 6 physical core. > > Setting this to twelve (and blurring the syntactically border between > threads and cores thereby...) gives me twelves cores in top, htop > and such and (as an example) compiling the kernel is faster - > so it is not a display gimmick only. > > I think "Number of cores" is a misnomer...or am I wrong? > > Cheers! > Meino > Meino, Yes, /proc/cpuinfo. Sorry. Well yes, I guess the 'Number of cores' is a misnomer if you're trying to equate the language in the kernel against Intel/AMD marketing data for physical cores. 6 physical cores with or without hyperthreading is still 6 physical cores. However 6 physical cores (my processor) _WITH_ hyperthreading enabled is 12 _LOGICAL_ cores which is more what I think the kernel verbiage is about. Semantics I suppose. I'm glad you found it wasn't a gimmicky number. It really does work, within the limits of the hardware being able to figure out what one thread should be fetching or writing while the other thread is computing. It's not a perfect 2:1 like 12 physical cores might be, but it's a lot less silicon and therefore a lot less expensive. Cheers, Mark
Re: [gentoo-user] Recommended filesystem for an SSD drive and what mount setting to be used?
Hello, Meino. On Fri, Mar 27, 2020 at 06:34:58 +0100, tu...@posteo.de wrote: > Hi, > hopefully in the next daus my first SSD drive will arrive > (corona makes everything more difficult...). Yes. I bought myself a backup Fritzbox (a router) on the last day before the shops were closed here. If the one in service were to die, I'd be left unable to work from home, and with no wired telephone. > To prevent an "installed and works"-experience which ends > a month later in a damaged or over-weared SSD with a drastically > shortened lifetime, I want to ask here for own experiences: > - What is the best filesustem to be used with a SSD, which > will used for /root when it comes to prolong life of that > SSD ? > - What options are recommended for the according mount command? > Thanks a lot for any helpful advice in advance! My experience suggests not to worry too much about these things. On my three year old box I've got no HDDs, just two Samsung 500 GB NVMe M2 SDDs (one mounted on the motherboard, the other on a PCI plug in board). Their partitions are set up in SW RAID-1 (apart from /boot). I just use ext4. I've had no trouble in almost three years with these SSDs. All my emergeing is done on a tmpfs (in RAM), for which I've got 16 GB RAM. I've got swap partitions (non-RAID ;-) on the SSDs, but with 16 GB RAM I suspect they don't get used much. And, yes, I do regular backups. Not that I've ever needed them, but if I stopped, I soon would. ;-) > Cheers! > Meino -- Alan Mackenzie (Nuremberg, Germany).
Re: [gentoo-user] Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
On Saturday, 28 March 2020 06:19:56 GMT tu...@posteo.de wrote: > Hi, > > Since there is A LOT of stuff to read about UEFI, MTBR and hybrid > systems I want to sort out, what I need to read in advance. What is "MTBR"? Mean Time Between Repair? :-/ I'll assume you mean MBR. > The setup will be: > > 512 SSD carrying the root fs with all directories with heavy write-loaded > mounted with/in/at/on/beneath/beyond/under/by/for (or whatever preposition > is correct here ... I am no native speaker, sorry :) tmpfs "on tmpfs" would do, in any case we understand what you mean. > One 3TB harddisc as a "data grave" > > From what I read on the internet: > Everything bigger than 2TB needs to be GPT-formatted. > Is there anything better than gparted for that job? There many partition management tools and it soon becomes subjective which one is /better/ - usually it is the tool you are most familiar with. If you want a GUI tool, GParted tends to be the go to tool although I've come across others. Personally, 'sys-apps/gptfdisk' is my go to tool for partitioning block devices, but parted and GParted will also perform the required task with aplomb. > So I need to format the harddisc with a GPT. > > The SSD: > > 1.) From the pure technical point of view: Is it possible to > format the SSD with a good ole MTB and boot the system > from it and mount the hardisc (GPT) as usual? Yes, both partition table types are readable by modern OS' and suitable partitioning tools. The GUID Partition Table disks are self-identifying, with all partition information stored on the physical disk and in addition with a long list of partition type identifiers more attuned to Linux usage (e.g. for /home, /, swap, etc.). > 2.) Is this reasonable? > > Is this setup reasonable...any ideal/advice is very welcome! > Thanks a lot in advance! > > Cheers! > Meino What I would consider reasonable in the year of 2020 is to use GPT, which is a more versatile and useful structure for partitioning disks, without the restrictions and deficiencies of MBR tables. On the other hand if you have a really old disk which you are not planning to repartition, I would consider it reasonable to leave it alone with its existing MBR table and get on with more important/interesting tasks. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] laptop closure should not rfkill
On Saturday, 28 March 2020 09:43:56 GMT J. Roeleveld wrote: > On Friday, March 27, 2020 3:18:47 PM CET n952162 wrote: > > I want my wireless to continue when I close my laptop lid. Does anyone > > have a clue what I have to set to make that happen? > > > > I've followed it this far: > > /etc/acpi/events/lm_lid.sh > > > >/etc/acpi/actions/lm_lid.sh > > > > /lib/udev/lmt-udev (laptop mode tools?) > > > >/usr/sbin/laptop_mode > > > > /etc/laptop-mode/conf.d > > > > /etc/laptop-mode/laptop-mode.conf has this (I don't find any relevant > > > > CONTROL_*): > > /# "Laptop mode" is the mode in which laptop mode tools makes the > > computer// > > //# consume less power. This includes the kernel "laptop_mode" > > feature, which// > > //# allows your hard drives to spin down, as well as various other > > settings which// > > //# can be tweaked by laptop mode tools. You can enable or disable > > all of these// > > //# settings using the CONTROL_... options further down in this > > config file.// > > > > // > > # > > ##// // > > > > //...// > > // > > //#// > > //# Enable laptop mode when the laptop's lid is closed, even when > > we're on AC// > > //# power? (ACPI-ONLY)// > > //#// > > //ENABLE_LAPTOP_MODE_WHEN_LID_CLOSED=0// > > / > > > > Would it be okay to just *remove* /etc/acpi/*/lm_lid.sh? > > checking the git-repository for laptop-mode-tools, I see the following file > with a change to "disable networking by default": > > https://github.com/rickysarraf/laptop-mode-tools/blob/lmt-upstream/etc/lapto > p-mode/conf.d/ethernet.conf > > IOW, I would suggest checking the file: > > /etc/laptop-mode/conf.d/ethernet.conf > > -- > Joost Most desktops have power management facilities from which the behaviour of various inputs (e.g. lid switch) and configured actions can be finely tuned. I mention it here in the off chance it wasn't already explored as an option. signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] laptop closure should not rfkill
On Friday, March 27, 2020 3:18:47 PM CET n952162 wrote: > I want my wireless to continue when I close my laptop lid. Does anyone > have a clue what I have to set to make that happen? > > I've followed it this far: > > /etc/acpi/events/lm_lid.sh >/etc/acpi/actions/lm_lid.sh > /lib/udev/lmt-udev (laptop mode tools?) >/usr/sbin/laptop_mode > /etc/laptop-mode/conf.d > > > /etc/laptop-mode/laptop-mode.conf has this (I don't find any relevant > CONTROL_*): > > /# "Laptop mode" is the mode in which laptop mode tools makes the > computer// > //# consume less power. This includes the kernel "laptop_mode" > feature, which// > //# allows your hard drives to spin down, as well as various other > settings which// > //# can be tweaked by laptop mode tools. You can enable or disable > all of these// > //# settings using the CONTROL_... options further down in this > config file.// > > //# > ##// // > //...// > // > //#// > //# Enable laptop mode when the laptop's lid is closed, even when > we're on AC// > //# power? (ACPI-ONLY)// > //#// > //ENABLE_LAPTOP_MODE_WHEN_LID_CLOSED=0// > / > > > Would it be okay to just *remove* /etc/acpi/*/lm_lid.sh? checking the git-repository for laptop-mode-tools, I see the following file with a change to "disable networking by default": https://github.com/rickysarraf/laptop-mode-tools/blob/lmt-upstream/etc/laptop-mode/conf.d/ethernet.conf IOW, I would suggest checking the file: /etc/laptop-mode/conf.d/ethernet.conf -- Joost
Re: [gentoo-user] Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
On 03/28 08:49, Neil Bothwick wrote: > On Sat, 28 Mar 2020 07:19:56 +0100, tu...@posteo.de wrote: > > > From what I read on the internet: > > Everything bigger than 2TB needs to be GPT-formatted. > > Is there anything better than gparted for that job? > > > > So I need to format the harddisc with a GPT. > > > > The SSD: > > > > 1.) From the pure technical point of view: Is it possible to > > format the SSD with a good ole MTB and boot the system > > from it and mount the hardisc (GPT) as usual? > > 2.) Is this reasonable? > > Why would you want to use the archaic MS-DOS partition table with all its > restrictions and fragility when you have the more flexible and robust GPT > option. > > > -- > Neil Bothwick > > Y'know how s'm people treat th'r body like a TEMPLE? > Well, I treat mine like 'n AMUSEMENT PARK... S'great... Hi Neil, ...because I have experienced neither any restrictions nor any fragilities in the last 12 years of using a MTB setup with that "MS-DOS" setup (technically it is more a MTB-setup. MSDOS refers to V/FAT, which needs to be used with UEFI in the first partition as far as I know...isn't it?) In short: Never chance a runnig system. But let us better discuss on a more technical level, since I asked for technical help and not for more rethorical given question as an answer. Cheers! Meino
Re: [gentoo-user] Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
On Sat, 28 Mar 2020 07:19:56 +0100, tu...@posteo.de wrote: > From what I read on the internet: > Everything bigger than 2TB needs to be GPT-formatted. > Is there anything better than gparted for that job? > > So I need to format the harddisc with a GPT. > > The SSD: > > 1.) From the pure technical point of view: Is it possible to > format the SSD with a good ole MTB and boot the system > from it and mount the hardisc (GPT) as usual? > 2.) Is this reasonable? Why would you want to use the archaic MS-DOS partition table with all its restrictions and fragility when you have the more flexible and robust GPT option. -- Neil Bothwick Y'know how s'm people treat th'r body like a TEMPLE? Well, I treat mine like 'n AMUSEMENT PARK... S'great... pgpPxzQ7TGUmu.pgp Description: OpenPGP digital signature
[gentoo-user] Courier-Authlib and S.A.S.L.
Hello, I've been attempting to create a Virtual Mail Server using Postfix, MySQL, and Courier, however I've hit a slight brick wall when dealing with the S.M.T.P.\ authentication. As advised at [1], is is generally preferable to use the unified Courier-Authlib interface when interacting with the database, as opposed to having Courier-Authlib *and* cyrus-sasl attaining direct access. I've been at this for a while, and I'm almost annoyed, and somewhat desperate. Although every other element of my mail server works spotlessly, including 'STARTTLS', S.M.T.P.\ authentication through cyrus-sasl and courier-authlib confuses me endlessly, especially due to the lack of information dumped to the logs, despite the maximum levels of logging enabled in the respective configuration files. courier-authlib seems to be working fine, as 'authtest' is capable of retrieving user accounts specified in the MySQL database, which leads me to believe that the problem lies with cyrus-sasl. I also know that Postfix is probably loading cyrus-sasl correctly, as an error in the /etc/sasl2/smtpd.conf file leads to an error when starting Postfix. Fixing the syntax error leads to no errors, but also invokes no mention of a successful load in the syslog. [2] is a list of the Postfix capabilities, as reported via telnet 'EHLO', [3] is the /etc/sasl2/smtpd.conf file, and [4] is the relevant parts of Postfix main.cf. I'm unsure if the erroneous behaviour seen here is a result of a personal fundamental misunderstanding of the virtual mail server stack, or just a silly typo or omission in one of configuration files. Thank you in advance for your assistance. -- [1] S.M.T.P. Authentication, Gentoo Wiki https://wiki.gentoo.org/wiki/Complete_Virtual_Mail_Server/SMTP_Authentication/en [2] Postfix capabilities; notice the upsetting lack of an AUTH response. Issuing an AUTH LOGIN command results in "503 5.5.1 Error: authentication not enabled" being returned. 250-PIPELINING 250-SIZE 20971520 250-ETRN 250-STARTTLS 250-ENHANCEDSTATUSCODES 250-8BITMIME 250-DSN 250-SMTPUTF8 250 CHUNKING [3] smtpd.conf; the courier-authlib socket should have the correct ownership, such that it is owned by root:mail. 'postfix' belongs to the 'mail' group. pwcheck_method: authdaemond mech_list: LOGIN PLAIN sql_select: dummy authdaemond_path: /var/lib/courier/authdaemon/socket log_level: 7 [4] S.A.S.L.-relevant sections of the Postfix main.cf file. (line break on recipient_restrictions added for this e-mail) smtpd_sasl_path = smtpd broken_sasl_auth_clients = yes smtpd_sasl_auth_enable = yes smtpd_sasl_security_options = noanonymous smtpd_sasl_local_domain = smtpd_sasl_authenticated_header = yes smtpd_recipient_restrictions = permit_sasl_authenticated, permit_mynetworks \ reject_unauth_destination -- Ashley Dixon suugaku.co.uk
[gentoo-user] Still questions concerning a reasonable setup of a new system: UEFI &&/|| MTBR
Hi, Since there is A LOT of stuff to read about UEFI, MTBR and hybrid systems I want to sort out, what I need to read in advance. The setup will be: 512 SSD carrying the root fs with all directories with heavy write-loaded mounted with/in/at/on/beneath/beyond/under/by/for (or whatever preposition is correct here ... I am no native speaker, sorry :) tmpfs One 3TB harddisc as a "data grave" >From what I read on the internet: Everything bigger than 2TB needs to be GPT-formatted. Is there anything better than gparted for that job? So I need to format the harddisc with a GPT. The SSD: 1.) From the pure technical point of view: Is it possible to format the SSD with a good ole MTB and boot the system from it and mount the hardisc (GPT) as usual? 2.) Is this reasonable? Is this setup reasonable...any ideal/advice is very welcome! Thanks a lot in advance! Cheers! Meino
