Re: UEFI Migrate to new hard drive

2019-02-11 Thread Richard Shaw
On Mon, Feb 11, 2019 at 6:19 PM Chris Murphy 
wrote:

> On Sun, Feb 10, 2019 at 2:21 PM Richard Shaw  wrote:
>
> > efibootmgr -B 0001 (which was the fedora entry)
> > and then
> > efibootmgr -c -d /dev/sda -p 1 -l "EFI/fedora/shim.efi" -L Fedora
> >
> > then it worked...
> >
> > NOTHING ABOUT THIS IS INTUITIVE!
>
> It's not intuitive, it's also not self-describing. But there is a
> (massive) spec that describes the intended behavior. The end result
> is, with UEFI comes with a stick, and with BIOS you're left on your
> own to learn you need a stick and then find it.
>

Well, I was pretty frustrated at that point :) But there are some things
that would make it easier...

For one, why did my PARTUUID change? I used gparted and moved both /boot
and /boot/efi over to the new disk. The UUID was unchanged, and I had no
idea that EFI used PARTUUID.

Also, why does efibootmgr silently act like it's working? I tried
"changing" an entry and it exited like it worked but couldn't because
efivars was still mounted RO... I can't remember what I did but a different
command (maybe remove?) was the one that got me the hint that I needed to
remount efivars as RW.

Thirdly... I wasn't sure if partitions in EFI start with 0 or 1 and it I
couldn't find ANYWHERE that said one way or the other...

Thanks,
Richard
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org


Re: UEFI Migrate to new hard drive

2019-02-11 Thread Chris Murphy
On Sun, Feb 10, 2019 at 2:21 PM Richard Shaw  wrote:

> efibootmgr -B 0001 (which was the fedora entry)
> and then
> efibootmgr -c -d /dev/sda -p 1 -l "EFI/fedora/shim.efi" -L Fedora
>
> then it worked...
>
> NOTHING ABOUT THIS IS INTUITIVE!

It's not intuitive, it's also not self-describing. But there is a
(massive) spec that describes the intended behavior. The end result
is, with UEFI comes with a stick, and with BIOS you're left on your
own to learn you need a stick and then find it.

Anyway, another option is you can remove all the boot entries (for
operating systems, you can leave device entries). The spec defines
that if none of the NVRAM boot entries point to a bootable system,
then the firmware starts looking on the ESP (there are rules for the
order of ESP checking), for a BOOT directory, and blindly executes the
BOOT.EFI found there. On x86_64, the BOOTX64.EFI is actually
shim.efi which has the ability along with fallback.efi to add the
proper Fedora entry to NVRAM if one doesn't exist already.



-- 
Chris Murphy
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org


Re: OSM & GPS??

2019-02-11 Thread Clifford Snow
On Mon, Feb 11, 2019 at 11:48 AM Beartooth  wrote:

> On Sat, 09 Feb 2019 14:59:57 -0800, Clifford Snow wrote:
>
> > I would suggest using QGIS. It run great under Fedora. I use Dani's copr
> > repo for QGIS. (copy of repo below) It's has the latest version, 3.4
> > which is very stable. QGIS will natively open GPX tracks. Then you'll
> > want to get some backgrounds. I would add the QuickMapServices plugin.
> > Once the plugin is installed, Go to Web, QuickMapServices and open
> > settings. Under More service, select Get contributed pack. It will load
> > in more than a dozen backgrounds you can use.
>
> What does QuickMapServices plug into?  Dnf installed QGIS, and
> also mapnik, which I take to be relevant; but having had my share of
> dependency hell back in the day, I'm reluctant to update anything except
> via dnf. Is that going to make trouble?
>

It shouldn't. Plugins are fetched and I believe stored in your home
directory under ~/.local/share/QGIS/QGIS3

>
> There may be trouble already:
>
> $ qgis &
> [1] 12521
> [btth@localhost ~]$ Warning: loading of qgis translation failed [/usr/
> share/qgis/i18n//qgis_en_US]
> Warning: loading of qt translation failed [/usr/share/qt4/translations/
> qt_en_US]
> Warning: Object::connect: No such signal
> QgsMergedBookmarksTableModel::&QgsMergedBookmarksTableModel::selectItem(
> const
> QModelIndex &index )
> Warning: Object::connect:  (receiver name: 'QgsBookmarksBase')
> Warning: QCss::Parser - Failed to load file  "/style.qss"
> QInotifyFileSystemWatcherEngine::addPaths: inotify_add_watch failed: No
> such file or directory
> Warning: QFileSystemWatcher: failed to add paths: /home/btth/.qgis2//
> project_templates
> loaded the Generic plugin
> Warning: QLayout: Attempting to add QLayout "" to QgsPanelWidgetStack
> "mWidgetStack", which already has a layout
>

Its possibly a conflict with python installations. I've been down that
road. Now when I install from pip/pip3 I use the --user option.

>
> > As a personal note from an active OSM contributor, please at least
> > consider uploading your tracks. Just go to osm.org and select GPX Tracks
> > to upload yours. If you are willing to put some extra effort, once
> > you've added your traces, please add your trail to OSM.
>
> I obviously need to learn the difference between a trail and a
> trace. Maybe it's in the matter I read yesterday; absent-mindedness,
> alas!, gets worse with age.
>

Use the gpx trace along with imagery to add the trail in OSM. Once the
trace is uploaded, it can be used as a background to add a line feature to
OSM. Using the browser based editor, iD, select the line feature to add
your trail. Tag the line as highway=path + a name= if its named. A gpx
trace from a consumer grade device can easily be off by 3 meters - if not
more.  By using imagery along with the gpx the trail can easily be more
accurate.

Good luck,
Clifford
-- 
@osm_seattle
osm_seattle.snowandsnow.us
OpenStreetMap: Maps with a human touch
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org


libvirtd, pihole, dnsmasq complaints

2019-02-11 Thread Chris Murphy
I think this is just noise in the journal; not a real problem (it's an
annoyance).

Fedora Server 29 running two docker containers: pihole and battery
historian. When I do 'sysmtectl start libvirtd' I immediately start
getting the following in the journal every 5s.

Feb 11 13:53:23 fnuc.local dnsmasq[32033]: failed to create listening
socket for 192.168.122.1: Address already in use

That continues until I both stop libvirtd and also remove the virbr0
link. I'm kinda wondering if this is suboptimal behavior that
constitutes a bug? I don't know why it's failing to create a socket
for 192.168.122.1, maybe docker owns it. But then why doesn't libvirtd
just fallback to some other address and notify in the journal, rather
than spamming the journal? In any case I'm still able to ssh into and
out of any VM I create. I'm also able to connect via virt-manager and
spice to any VM. So its seems like it's just noise and not an actual
problem.

  fnuc.local  chris  ~  ip a
1: lo:  mtu 65536 qdisc noqueue state UNKNOWN
group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
   valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
   valid_lft forever preferred_lft forever
2: enp3s0:  mtu 1500 qdisc fq_codel
state UP group default qlen 1000
link/ether b8:ae:ed:77:ea:51 brd ff:ff:ff:ff:ff:ff
inet 10.0.0.250/24 brd 10.0.0.255 scope global dynamic noprefixroute enp3s0
   valid_lft 377745sec preferred_lft 377745sec
inet6 2601:282:700:8c78:3cef:4672:2da9:af30/64 scope global
dynamic noprefixroute
   valid_lft 252404sec preferred_lft 252404sec
inet6 fe80::91f1:7594:6bae:99ea/64 scope link noprefixroute
   valid_lft forever preferred_lft forever
3: wlp2s0:  mtu 1500 qdisc noop state DOWN group
default qlen 1000
link/ether fe:b0:a1:f2:be:c4 brd ff:ff:ff:ff:ff:ff
4: docker0:  mtu 1500 qdisc noqueue
state UP group default
link/ether 02:42:7a:65:7b:b6 brd ff:ff:ff:ff:ff:ff
inet 172.17.0.1/16 scope global docker0
   valid_lft forever preferred_lft forever
inet6 fe80::42:7aff:fe65:7bb6/64 scope link
   valid_lft forever preferred_lft forever
6: vethe8a3f14@if5:  mtu 1500 qdisc
noqueue master docker0 state UP group default
link/ether d2:b4:ca:37:f7:3a brd ff:ff:ff:ff:ff:ff link-netnsid 0
inet6 fe80::d0b4:caff:fe37:f73a/64 scope link
   valid_lft forever preferred_lft forever
8: veth1b171f3@if7:  mtu 1500 qdisc
noqueue master docker0 state UP group default
link/ether 9a:f4:98:f3:08:1c brd ff:ff:ff:ff:ff:ff link-netnsid 1
inet6 fe80::98f4:98ff:fef3:81c/64 scope link
   valid_lft forever preferred_lft forever
9: virbr0:  mtu 1500 qdisc noqueue
state DOWN group default qlen 1000
link/ether 52:54:00:24:e3:cf brd ff:ff:ff:ff:ff:ff
inet 192.168.122.1/24 brd 192.168.122.255 scope global virbr0
   valid_lft forever preferred_lft forever
10: virbr0-nic:  mtu 1500 qdisc fq_codel master
virbr0 state DOWN group default qlen 1000
link/ether 52:54:00:24:e3:cf brd ff:ff:ff:ff:ff:ff
  fnuc.local  chris  ~ 




-- 
Chris Murphy
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org


Re: OSM & GPS??

2019-02-11 Thread Beartooth
On Sat, 09 Feb 2019 14:59:57 -0800, Clifford Snow wrote:

> I would suggest using QGIS. It run great under Fedora. I use Dani's copr
> repo for QGIS. (copy of repo below) It's has the latest version, 3.4
> which is very stable. QGIS will natively open GPX tracks. Then you'll
> want to get some backgrounds. I would add the QuickMapServices plugin.
> Once the plugin is installed, Go to Web, QuickMapServices and open
> settings. Under More service, select Get contributed pack. It will load
> in more than a dozen backgrounds you can use.

What does QuickMapServices plug into?  Dnf installed QGIS, and 
also mapnik, which I take to be relevant; but having had my share of 
dependency hell back in the day, I'm reluctant to update anything except 
via dnf. Is that going to make trouble?

There may be trouble already:

$ qgis &
[1] 12521
[btth@localhost ~]$ Warning: loading of qgis translation failed [/usr/
share/qgis/i18n//qgis_en_US]
Warning: loading of qt translation failed [/usr/share/qt4/translations/
qt_en_US]
Warning: Object::connect: No such signal 
QgsMergedBookmarksTableModel::&QgsMergedBookmarksTableModel::selectItem( const 
QModelIndex &index )
Warning: Object::connect:  (receiver name: 'QgsBookmarksBase')
Warning: QCss::Parser - Failed to load file  "/style.qss" 
QInotifyFileSystemWatcherEngine::addPaths: inotify_add_watch failed: No 
such file or directory
Warning: QFileSystemWatcher: failed to add paths: /home/btth/.qgis2//
project_templates
loaded the Generic plugin 
Warning: QLayout: Attempting to add QLayout "" to QgsPanelWidgetStack 
"mWidgetStack", which already has a layout

> As a personal note from an active OSM contributor, please at least
> consider uploading your tracks. Just go to osm.org and select GPX Tracks
> to upload yours. If you are willing to put some extra effort, once
> you've added your traces, please add your trail to OSM.

I obviously need to learn the difference between a trail and a 
trace. Maybe it's in the matter I read yesterday; absent-mindedness, 
alas!, gets worse with age.

> When I'm speaking to a group about OSM, I'm usually asked about quality,
> which is at least as good, if not better, than the others in large
> cities. But for trails, OSM has the most trails of any map service. 99%
> of those trails are added from gpx traces from people just like you.
> 
> If you need help with OSM or QGIS, please contact me directly.

First of all, thanks a million! That's vastly more than I had 
found, and very apposite. I've been beavering into it.

Second, goodgoddlemityWOW! I hadn't the faintest notion of the 
incredible vastness of the project. I imagined something like downloading 
a few Coast & Geodetic Service Maps to a GPS a/o a computer -- as I did 
twenty-odd years ago. 

It's going to take me a while just to digest the idea of what is 
out there, and another while to learn the jargon. "Good Lord willin' an' 
the crick don't rise," as people say in these here parts, I'll be back 
with questions, lots of questions, and I hope more understanding of what 
there is to learn.

Again, many, many thanks!
-- 
Beartooth Staffwright, Not Quite Clueless Power User
Remember I know little (precious little!) of where up is.
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org


Very slow DNF update and a workabound

2019-02-11 Thread Barry Scott
I'll been putting up with very slow dnf update for a while.
strace showed that it was reading /var/lib/rpm/Packages 4KiB at a time.
My /var/lib/rpm/Packages is 118MiB so that takes a while on a hard disk
around 60s.

My desktop is i7 4GHz CPU 16GiB of ram.

Unless there is a way to configure DNF to use bigger reads this hack speeds it
up.

# cat /var/lib/rpm/* > /dev/null; dnf update --refresh 

After the cat of the files Packages will be in kernel buffers and then dnf 
runs in 1s-2s.

Do you want a bug report on this issue?

Barry


___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org


Re: Home-directory, NFS and automount

2019-02-11 Thread Gordon Messmer

On 2/11/19 6:30 AM, Paul-Erik Törrönen wrote:

My setup was simple, I had the following line in /etc/fstab:

nfs-server:/path/to/home   /home   nfs defaults  0 0 



That's not "automount."  That's just a normal NFS filesystem. Using the 
word "automount" will confuse people about what you're trying to do.


Replace the word "defaults" with the word "nofail" and the system should 
behave like it used to.


___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org


Re: Home-directory, NFS and automount

2019-02-11 Thread Paul-Erik Törrönen

On 2019-02-11 14:45, George N. White III wrote:

This change might be the new systemd automounter.   If your previous
configuration
was using autofs, that should still be available.  More detail would be
helpful.


My setup was simple, I had the following line in /etc/fstab:

nfs-server:/path/to/home   /home   nfs defaults  0 0

Locally on the laptop there is also /home/ (with the correct uid 
and gid).


When the laptop is not connected to the local network (ie. the 
nfs-server is not available), the user logs on using the local version 
of /home/.
When the laptop is connected to the local network (ie. the nfs-server is 
available), the laptop mounts nfs-server:/path/to/home to /home at boot 
time and the user logs on using the nfs-server version of /home (which 
has matching uid/gid on files and directories).


Does this clarify the case?

Poltsi
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org


Re: UEFI Migrate to new hard drive

2019-02-11 Thread Richard Shaw
On Mon, Feb 11, 2019 at 12:36 AM Łukasz Posadowski 
wrote:

> Data Sun, 10 Feb 2019 15:20:17 -0600
> Richard Shaw  napisał(a):
>
> > I bought an SSD to replace my HDD and I used System Rescue CD to move
> > my EFI and boot partitions over and then used lvm tools to add the
> > new drive to the lvm and then remove the old one.
> >
> > So all of my data is over on the new driver... but I can't boot it...
> >
> > I've looked at efibootmgr and it still has an entry for Fedora, the
> > partition looks correct and per efibootmgr the loader name looks
> > good...
> >
> > But no matter what I do it won't boot.
> >
> > So I figured it out...
> >
> > Somehow the PARTUUID did change... And using system rescue CD the
> > efivars was mounted read only so even though efibootmgr didn't
> > complain, it wasn't actually changing anything.
> >
> > I had to:
> >
> > efibootmgr -B 0001 (which was the fedora entry)
> > and then
> > efibootmgr -c -d /dev/sda -p 1 -l "EFI/fedora/shim.efi" -L Fedora
> >
> > then it worked...
> >
> > NOTHING ABOUT THIS IS INTUITIVE!
>
> Thanks. I have similar problems with Samsung 530U when changing distro.
> I removed the drive from Samsung, plugged to another computer trough
> usb/sata adapter and wiped it clean. I can't even boot from Rescue CD
> if UEFI has valid xyz distro entries and appropriate partitions on disk.
>
> I will try efibootmgr to solve it next time.


I forgot to mention I had to do a remount of efivars so it could get
updated.

efibootmgr is not good about letting you know. It just silently looks like
it completes.

Thanks,
Richard


>
> --
> Łukasz Posadowski
> ___
> users mailing list -- users@lists.fedoraproject.org
> To unsubscribe send an email to users-le...@lists.fedoraproject.org
> Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
> List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
> List Archives:
> https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org
>
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org


Re: Home-directory, NFS and automount

2019-02-11 Thread George N. White III
On Mon, 11 Feb 2019 at 04:46, Paul-Erik Törrönen  wrote:

> I've had for a long time a setup for (Fedora-)laptops where the laptop
> has local home directories for users in /home/
> and in fstab the directory is mounted (when available) to NFS share on
> home network.
>


> This worked nicely until recently (F28/F29) when it seems that automount
> was configured to  attempt to mount /home as the user logs on (GDM).
> Since this automount fails, the login is terminated despite the fact
> that there is a valid /home/ on the fs.
>
> So the question is: Is there anything similar in simplicity to set up as
> a replacement to the current setup, or, alternatively, how does one
> disable the automount from attempting to mount non-reachable NFS mounts
> at login time?
>

This change might be the new systemd automounter.   If your previous
configuration
was using autofs, that should still be available.  More detail would be
helpful.

-- 
George N. White III
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org


Home-directory, NFS and automount

2019-02-11 Thread Paul-Erik Törrönen
I've had for a long time a setup for (Fedora-)laptops where the laptop 
has local home directories for users in /home/
and in fstab the directory is mounted (when available) to NFS share on 
home network.


This worked nicely until recently (F28/F29) when it seems that automount 
was configured to  attempt to mount /home as the user logs on (GDM). 
Since this automount fails, the login is terminated despite the fact 
that there is a valid /home/ on the fs.


So the question is: Is there anything similar in simplicity to set up as 
a replacement to the current setup, or, alternatively, how does one 
disable the automount from attempting to mount non-reachable NFS mounts 
at login time?


Poltsi
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fedora Code of Conduct: https://getfedora.org/code-of-conduct.html
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/users@lists.fedoraproject.org