openSUSE Users November 2021

users@lists.opensuse.org

52 participants
50 discussions

gtk2 application start delays 25 seconds
by Istvan Gabor 06 Dec '21

06 Dec '21

Hello: This occurs in openSUSE Leap 15.3 (64 bit). Gtk2 applications start with a 25 seconds delay. That is, it takes exactly 25 seconds the application window to appear. This behavior has been reported previously in other distros, eg: In 20.04.1 all gtk2 apps starts 25 seconds https://bugs.launchpad.net/ubuntu/+bug/1891245 I installed appmenu-gtk2-module and libcanberra-gtk-module but it did not solve the problem. How can I fix this in Leap 15.3? Thanks, Istvan

5 7

zypper and certificate issue
by Roger Oberholtzer 30 Nov '21

30 Nov '21

I have trouble refreshing the status of a couple repositories on my pretty much up-to-date Tumbleweed system. The problem is with packman, as well as M17N:/fonts/openSUSE_Tumbleweed and windows:/mingw:/win64/openSUSE_Tumbleweed/. So it is both for some openSUSE and some non-openSUSE repos. Some external repos work, and some openSUSE work. The message is generally this: Repository 'MinGW64' is invalid. [mingw64|https://download.opensuse.org/repositories/windows:/mingw:/win6 /openSUSE_Tumbleweed/] Valid metadata not found at specified URL History: - Download (curl) error for 'https://download.opensuse.org/repositories/windows:/mingw:/win64/openSUSE_T…': Error code: Curl error 60 Error message: SSL certificate problem: unable to get local issuer certificate - Can't provide ./repodata/526857fc123d7384947e093f8293170c026ad9f736e350bedeb65fd292475760-primary.xml.gz So I have some sort of a certificate problem. If I run zypper search ca-certificates, It says: S | Name | Summary | Type --+----------------------------------+-------------------------------------------------------+-------- i | ca-certificates | Utilities for system wide CA certificate installation | package | ca-certificates-cacert | CAcert root certificates | package i | ca-certificates-mozilla | CA certificates for OpenSSL | package | ca-certificates-mozilla-prebuilt | Pre-built CA certificates for OpenSSL | package | ca-certificates-steamtricks | Provides /etc/ssl/certs/ca-certificates.crt | package It does not help to delete and redefine the repository. Do I need to install ca-certificates-cacer? I Googled this, and the only solution I saw was for SUSE, using the SUSEConnect --cleanup command. Which I do not have on Tumbleweed. -- Roger Oberholtzer

5 16

Re: KWallet--how?
by Carlos E. R. 30 Nov '21

30 Nov '21

On 30/11/2021 03.55, dmcgarrett(a)optonline.net wrote: I had to edit your post to know who said what. That mail client you use is terrible. >> Don't use instructions for windows. > The instructions I found--this was a couple of months ago, and different > of course, said it only works > in Windows, and it did. Once you get T'Bird working in one spot, it > works everywhere. > >> I don't think you will be happier with KMail. > > I was happy with T]Bird until it lost its mind again. > >> Look. Open Thunderbird. >> Edit / Preferences. >> Search for "passw" (there is a search box, top-right). > >> You will see a button for "Saved passwords", click that. Search then >> for the one or ones you want and simply delete them. Or remove all. You >> can also display them. Once done, close that dialog, and verify that you >> have defined a Master Password. If not, do create it now. > > >> Or is it the master password the one you forgot? > > Up to now, I didn't use a master password. > I erased the only password it showed. I was not using a master p/w, but > now I have added one. It still asks for > a password (after entering the master password) and it still refuses to > use a password I was using, and it still asks for and ignores any new > password I enter. And one of its complaints along the way--it didn't > know my > mail server, altho it is in the program somewhere and showed it along > the way. > >> I don't know if there is a way to regenerate a forgotten master >> password, but my guess is, you have to first erase all existing >> passwords, then try disable "use a master password". > >> Otherwise, I would create a new Thunderbird profile and start over. > Well, I'm certainly willing to do that. How can I get rid of everything > that's there and start over? Is there an omnibus command like zap > thunderbird? > Probably not! (I don > t want to start the OS over! > Thanx for the input--still beating on this terrible optimum routine. I highly suspect that your problem is not with Thunderbird, but with that stupid mail server / provider of yours. Thunderbird should: - Ask ONCE for the master password once you open the program the first time. - Ask ONCE in life for the password of your mail account the first time it tries to use it. Once for receiving, once for sending. - If the SERVER says the password is wrong, Thunderbird will ask for that server or account password, and will ask a thousand times till you give it right - ie, the server says it is right. You can hit cancel or ESC, probably. - Once the SERVER accepts the password you typed, Thunderbird will save it and never ask for it again in life. Ok, certainly there are easy ways to delete the entire Thunderbird thing, but you don't need to do that. Just create a new mail account for your mail provider. Just give it a different name than the one it has now. Thunderbird can have dozens of accounts defined. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)

1 0

Re: Swapping ThinkPad drive
by Carlos E. R. 28 Nov '21

28 Nov '21

On 26/11/2021 04.50, James Knott wrote: > On 2021-11-25 9:39 p.m., Carlos E. R. wrote: >> That's the one. On BIOS it lists disks, but on UEFI it lists "bootable >> oses", not partitions, that have entries in the EFI partition(s). > > No, even when I set for UEFI only, I still only see drives Weird. >> I don't know when UEFI machines started, but 10 years ago it would be >> a primitive form, if any. > > There was an update in 2013. > >> Well, you have to add Windows there. Look at the entries in your old >> disk EFI partition. > > The old disk was legacy, not UEFI. Ah! I forgot that. Sure, you need to install Windows using UEFI. Or ask in a Windows forum how to migrate a Windows install from traditional booting to UEFI booting, whether that is possible. It might. > > I guess I'll have to forget about UEFI on this computer. :-( > I don't know. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)

3 5

[opensuse] Kernel compilation the openSUSE way?
by Bjoern Voigt 28 Nov '21

28 Nov '21

I usually take the newest openSUSE kernel (from repository Kernel_stable). But since some month I have a DVB T2 HD card, which requires a patch (see https://bugzilla.kernel.org/show_bug.cgi?id=194171). So I need to compile kernels myself. I have a working compilation path. But the current path has 12 steps! It takes some time. I wonder, why it is so complicated to configure, compile and install a Kernel under openSUSE. Is there a better/easier/automated way to configure, compile and install a Kernel under openSUSE? I know this blog article. But the article uses Vanilla kernels and also needs 6 steps. With additional patches (4 steps), third-party modules (2 steps) and cleanup (1 step) the article would also need 13 steps: Compiling the Linux Kernel, the SUSE way https://www.suse.com/communities/blog/compiling-de-linux-kernel-suse-way/ This is my Kernel configuration, compilation and installation path: 1. Update kernel-source and kernel-default from Kernel_stable repository 2. Save kernel-source: cp -a /usr/src/linux-4.x.y-... /usr/src/linux-4.x.y-my 3. Apply my patches cd /usr/src/linux-4.x.y-my patch -b -p1 < ~myuser/.../my-kernel-patch.patch 4. Copy standard configuration: cd /usr/src/linux-4.x.y-my cp -v /boot/config-4.11.4-1.gcba98ee-default .config 5. Edit .config: remove CONFIG_LOCALVERSION, CONFIG_DEBUG_KERNEL, CONFIG_EXPERT remove some hardware-dependent settings 6. edit configuration interactively: Disable CONFIG_EXPERT and CONFIG_DEBUG_KERNEL, because otherwise "make binrpm-pkg" creates huge kernel binaries. make oldconfig [...] Local version - append to kernel release (LOCALVERSION) [] (NEW) -my1 * * Configure standard kernel features (expert users) * Configure standard kernel features (expert users) (EXPERT) [N/y/?] (NEW) Kernel debugging (DEBUG_KERNEL) [N/y/?] (NEW) make menuconfig Do some additional configuration. 7. Compile kernel (-j4 for four processor cores, ionice and nice to reduce the load, binrpm-pkg for RPM without source-RPM) make clean nice ionice -c idle make -j4 binrpm-pkg 8. Install the kernel RPM package rpm -Uvh /usr/src/packages/RPMS/x86_64/kernel-4.x.y_my1-....x86_64.rpm 9. Install necessary symbolic links for "dkms" cd /lib/modules/4.x.y-my1 ln -sv /usr/src/linux-4.x.y-my source ln -sv /usr/src/linux-4.x.y-my build 10. Update third-party modules (e.g. Nvidia driver): dkms install -m nvidia -v 375.66 -k 4.x.y-my1 11. Cleanup cd /usr/src/linux-4.x.y-my make clean 12. Select Grub configuration and reboot grub2-once --list | less grub2-once 2 reboot Greetings, Björn -- To unsubscribe, e-mail: opensuse+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse+owner(a)opensuse.org

7 13

[oS-en] eth0 trouble
by Carlos E. R. 28 Nov '21

28 Nov '21

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, Yesterday I wanted to update my small laptop (on Leap 15.2), and I connected it to the ethernet cable, but the laptop did not recognize it. If run "rcnetwork restart" it would appear, but would not connect to anything. At first I thought "dns trouble", but then I found that a ping 8.8.8.8 failed as well. I left it at that and switched ti WiFi, and performed the update, hoping that would cure the problem. Today I rebooted, with a newer kernel, and the cable connected from the start. After opening the desktop (I'm using Network Manager), I observed ini gkrellm that the interface eth0 was appearing and disapearing every minute or so. After a while, it became stable. It is not the cable, moving it has no effect. Message log: <https://susepaste.org/51821193> (I see the interface is going on and off again, in the log) Obviously, I have tried another cable, connected to another port of the switch. Same trouble. The switch led is solid green, but the computer does not see it. "rcnetwork restart" makes it appear with the correct IP, but a ping to the router (by number) fails. In fact, after removing the cable completely and issuing the command again, ifconfig reports that eth0 keeps the same address, but wlan has no address, and I can no longer connect to it. Ping to router from machine fails. Manually telling NM to dissconnect the ethernet (which cable was removed minutes ago) allows the wifi to work again. Second part of messages log: <https://susepaste.org/63312588> What is going on? This laptop worked previously, days ago. With the same cables. Another laptop works fine with one of those cables. - -- Cheers Carlos E. R. (from 15.2 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCYZ4p/Bwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfV3pAAn3ji0olpKPHiklGVVp/5 sPFsVbO9AKCV/fG9rlt9K0khax47qv2dswWz4g== =5vrF -----END PGP SIGNATURE-----

6 19

[oS-en] time rarity
by Carlos E. R. 27 Nov '21

27 Nov '21

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I was running a backup job, and I timed it. The output is weird: (here some other times out of the screen, about 3 minute total) real 161m35.208s user 216m2.220s sys 36m30.599s real 10m27.555s user 14m33.090s sys 2m37.550s real 11m7.647s user 14m39.255s sys 1m55.313s total real 184m27.589s user 247m7.060s sys 41m20.225s Now, how can be the "real" time less than the user time? I don't remember seeing it before. It is happening on a new 15.2 install. Perhaps the sum of times on the two cores of the machine? In case the job is relevant, it is this: function hacer() { echo echo "Haciendo la particion $1 ($2) en $3" echo "copiando, comprimiendo, y calculando md5..." mkfifo mdpipe dd if=/dev/$1 status=progress bs=16M | tee mdpipe | pigz > $3.gz & md5sum -b mdpipe | tee -a md5checksum_expanded wait rm mdpipe echo "$3" >> md5checksum_expanded echo "Verificando..." pigz --test $3.gz echo echo "·········" } echo time hacer sda3 "161M" "sda3__Windows" echo time hacer sda4 "10M" "sda4__LENOVO" echo time hacer sda6 "11M" "sda6__LENOVO_PART" - -- Cheers Carlos E. R. (from 15.2 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCYaFeeRwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfVZN8An3+T7XOnkPIYam64elLm zJIX5os5AJ9KGI+o0AZluPSZH1PhNsq5ds+JiQ== =MLSm -----END PGP SIGNATURE-----

2 2

preserving a kernel
by George from the tribe 27 Nov '21

27 Nov '21

So I updated tumbleweed and now my bluetooth doesn't work. After some investigation, I plan to file a bug report. It seems the firmware worked with kernel 5.14, but now it doesn't work with 5.15. But for now, as long as I boot into the old kernel, I still can use my bluetooth headphones, and this is important for multiple applications. The kernel I want to use is the 2nd kernel in the list, which is 5.14. I want to preserve this kernel so that on future updates, if they have not solved the bug, the 15.4 kernel will be preserved, but the default kernel, which is 5.15, will be replaced by the next kernel update. Is there a way to do this? At one time in the past I expanded the number of kernels that were in the list of available kernels so that I kept like 6 kernels. I don't remember why I did that, and I don't need that many in any case. However, I don't know how to preserve just one kernel, so that that particular kernel won't be erased, but that updates for other kernels will still load and I will be able to try them. Any ideas? -- George Box: 15.2 | Plasma 5 | AMD Phenom IIX4 | 64 | 32GB Laptop #1: 15.2 | Plasma 5 | AMD FX 7TH GEN | 64 | 32GB Laptop #2: TW | Plasma 5 | Core i7 | 64 | 16GB

3 3

puzzled by opensuse repo paths
by Istvan Gabor 26 Nov '21

26 Nov '21

Hello: In openSUSE 15.3 I see a lot of packages indicated by red text in Yast software manager. This lead me to check my configured repos and I am puzzled by what I found. For example I have configured openSUSE Leap 15.3 OSS update repo as: https://download.opensuse.org/update/leap/15.3/oss I checked the above path and found that is does not have the update packages. For example it does not have any kernel updates. But I have updated kernel package installed: kernel-default-5.3.18-59.27.1.x86_64 which is shown red in Yast software manager, indicating that I have no repo configured that has that package. The updated kernel package was installed by Yast Online Update Tool before I manually adjusted the repos (as I did before in earlier openSUSE versions). I am now puzzled by where the updated packages indicated by red text came from. Why not the update/leap/15.3/oss path contains these updates? Where can I find the Leap 15.3 specific updated repo? I see under opensuse/update/ path there are leap/ openSUSE-current/ openSUSE-non-oss-current/ openSUSE-stable/ tumbleweed/ tumbleweed-non-oss/ paths, but there are no version numbers given. What is stable, what is current? Where are they defined? Thanks, Istvan

2 1

Swapping ThinkPad drive
by James Knott 26 Nov '21

26 Nov '21

I've bought a 1 TB SSD drive to replace the 750 GB drive in my ThinkPad. I was considering using the opportunity to change to UEFI. I'll be mounting the old drive in an external case so that I can copy over the partitions, one of them NTFS. Anything I should worry about in converting to UEFI? Any suggestions? tnx jk

4 18

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

openSUSE Users November 2021