- openSUSE Support

About trim support on LUKS encrypted devices.
by Carlos E. R. 28 Dec '20

28 Dec '20

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I have, on one Leap 15.1 machine: /etc/crypttab: cr_home /dev/disk/by-id/ata-...-part5 none timeout=300,discard /etc/fstab: /dev/mapper/cr_home /home xfs lazytime,exec,nofail 1 2 Isengard:~ # cat /proc/cmdline BOOT_IMAGE=/boot/vmlinuz-4.12.14-lp151.28.67-default root=UUID=0d457df1-... showopts resume=/dev/disk/by-label/Swap splash=verbose Isengard:~ # Isengard:~ # mount | grep /home /dev/mapper/cr_home on /home type xfs (rw,relatime,lazytime,attr2,inode64,noquota) Isengard:~ # And apparently, trim is working: Isengard:~ # service fstrim status ● fstrim.service - Discard unused blocks on filesystems from /etc/fstab Loaded: loaded (/usr/lib/systemd/system/fstrim.service; static; vendor preset: disabled) Active: inactive (dead) since Mon 2020-12-28 00:01:26 CET; 12h ago Docs: man:fstrim(8) Process: 24826 ExecStart=/usr/sbin/fstrim -Av (code=exited, status=0/SUCCESS) Main PID: 24826 (code=exited, status=0/SUCCESS) Dec 28 00:00:03 Isengard systemd[1]: Starting Discard unused blocks on filesystems from /etc/fstab... Dec 28 00:01:26 Isengard fstrim[24826]: /home: 62.6 GiB (67245150208 bytes) trimmed on /dev/mapper/cr_home <======= Dec 28 00:01:26 Isengard fstrim[24826]: /: 1.7 GiB (1831469056 bytes) trimmed on /dev/sda4 Dec 28 00:01:26 Isengard systemd[1]: Started Discard unused blocks on filesystems from /etc/fstab. Isengard:~ # So I understand from that that trim is working, yet I'm not using "allow_discards" as a kernel option. On another machine, running Leap 15.2 (upgraded from 15.1 on 2020-11-08), it is not working with similar settings. I have: /etc/crypttab: cr_crypta /dev/disk/by-uuid/f9ef602f-... none noauto,discard /etc/fstab: /dev/mapper/cr_crypta /data/cripta xfs defaults,lazytime 1 2 minas-tirith:~ # cat /proc/cmdline BOOT_IMAGE=/vmlinuz-5.3.18-lp152.47-default root=UUID=7d8f7fc8-df18-474b-be96-e43862cf3289 resume=/dev/disk/by-label/Swap splash=verbose console=tty1 loglevel=3 showopts rd.convertfs minas-tirith:~ # minas-tirith:~ # mount | grep data /dev/mapper/cr_crypta on /data/cripta type xfs (rw,relatime,lazytime,attr2,inode64,logbufs=8,logbsize=32k,noquota) minas-tirith:~ # service fstrim status ● fstrim.service - Discard unused blocks on filesystems from /etc/fstab Loaded: loaded (/usr/lib/systemd/system/fstrim.service; static; vendor preset: disabled) Active: inactive (dead) since Mon 2020-12-28 12:48:46 CET; 11min ago Docs: man:fstrim(8) Process: 13106 ExecStart=/usr/sbin/fstrim -Av (code=exited, status=0/SUCCESS) Main PID: 13106 (code=exited, status=0/SUCCESS) minas-tirith:~ # Notice that in this machine the journal does not have information. 15.1 had the same problem, yet the journal is persistent. Syslog does have the entries: minas-tirith:~ # grep -i trim /var/log/messages <3.6> 2020-11-09 00:01:35 minas-tirith fstrim 20497 - - /other: 1.6 GiB (1716981760 bytes) trimmed on /dev/sda9 <3.6> 2020-11-09 00:01:35 minas-tirith fstrim 20497 - - /: 5.1 GiB (5511180288 bytes) trimmed on /dev/sda7 <3.6> 2020-11-09 00:01:35 minas-tirith fstrim 20497 - - /boot: 113.7 MiB (119165952 bytes) trimmed on /dev/sda6 <3.6> 2020-12-01 09:09:20 minas-tirith btrfs-trim.sh 11836 - - Path / is not btrfs, skipping <3.6> 2020-12-21 21:18:43 minas-tirith fstrim 5069 - - /other: 0 B (0 bytes) trimmed on /dev/sda9 <3.6> 2020-12-21 21:18:43 minas-tirith fstrim 5069 - - /: 1.9 GiB (1998217216 bytes) trimmed on /dev/sda7 <3.6> 2020-12-21 21:18:43 minas-tirith fstrim 5069 - - /boot: 9.4 MiB (9889792 bytes) trimmed on /dev/sda6 <3.6> 2020-12-28 12:48:46 minas-tirith fstrim 13106 - - /other: 0 B (0 bytes) trimmed on /dev/sda9 <3.6> 2020-12-28 12:48:46 minas-tirith fstrim 13106 - - /: 781.9 MiB (819888128 bytes) trimmed on /dev/sda7 <3.6> 2020-12-28 12:48:46 minas-tirith fstrim 13106 - - /boot: 4.9 MiB (5072896 bytes) trimmed on /dev/sda6 minas-tirith:~ # Previous 15.1 messages: <3.6> 2020-09-15 18:30:18 minas-tirith fstrim 14184 - - /other: 1.6 GiB (1716981760 bytes) trimmed on /dev/sda9 <3.6> 2020-09-15 18:30:18 minas-tirith fstrim 14184 - - /: 6 GiB (6432956416 bytes) trimmed on /dev/sda7 <3.6> 2020-09-15 18:30:18 minas-tirith fstrim 14184 - - /boot: 123.2 MiB (129189888 bytes) trimmed on /dev/sda6 But observe that /data/cripta does not appear, it is not trimmed. Why not? minas-tirith:~ # fstrim -v /data/cripta fstrim: /data/cripta: the discard operation is not supported minas-tirith:~ # mount | grep /data/cripta /dev/mapper/cr_crypta on /data/cripta type xfs (rw,relatime,lazytime,attr2,inode64,logbufs=8,logbsize=32k,noquota) minas-tirith:~ # On the other machine, it works: Isengard:~ # fstrim -v /home /home: 62.6 GiB (67245076480 bytes) trimmed Isengard:~ # mount | grep /home /dev/mapper/cr_home on /home type xfs (rw,relatime,lazytime,attr2,inode64,noquota) Isengard:~ # minas-tirith:~ # dmsetup info /dev/mapper/cr_crypta Name: cr_crypta State: ACTIVE Read Ahead: 256 Tables present: LIVE Open count: 1 Event number: 0 Major, minor: 254, 0 Number of targets: 1 UUID: CRYPT-LUKS1-...-cr_crypta minas-tirith:~ # Isengard:~ # dmsetup info /dev/mapper/cr_home Name: cr_home State: ACTIVE Read Ahead: 1024 Tables present: LIVE Open count: 1 Event number: 0 Major, minor: 254, 0 Number of targets: 1 UUID: CRYPT-LUKS1-...-cr_home Isengard:~ # I have another laptop running 15.1 where it works: /etc/crypttab: cr_sda8 UUID=1edf494d-d697-40b2-ba00-c7da0a1d5fbe none timeout=300,discard /etc/fstab: /dev/mapper/cr_sda8 /home xfs lazytime 0 1 Legolas:~ # cat /proc/cmdline BOOT_IMAGE=/boot/vmlinuz-4.12.14-lp151.28.67-default root=UUID=8c1f1030-... resume=/dev/disk/by-label/Swap splash=verbose showopts Legolas:~ # Legolas:~ # mount | grep /home /dev/mapper/cr_sda8 on /home type xfs (rw,relatime,lazytime,attr2,inode64,noquota) Legolas:~ # Legolas:~ # service fstrim status ● fstrim.service - Discard unused blocks on filesystems from /etc/fstab Loaded: loaded (/usr/lib/systemd/system/fstrim.service; static; vendor preset: disabled) Active: inactive (dead) since Mon 2020-12-28 13:36:52 CET; 3min 38s ago Docs: man:fstrim(8) Process: 10462 ExecStart=/usr/sbin/fstrim -Av (code=exited, status=0/SUCCESS) Main PID: 10462 (code=exited, status=0/SUCCESS) Dec 28 13:35:33 Legolas.valinor systemd[1]: Starting Discard unused blocks on filesystems from /etc/fstab... Dec 28 13:36:52 Legolas.valinor fstrim[10462]: /home: 69.2 GiB (74332495872 bytes) trimmed on /dev/mapper/cr_sda8 <===== Dec 28 13:36:52 Legolas.valinor fstrim[10462]: /: 2.3 GiB (2484498432 bytes) trimmed on /dev/sda11 Dec 28 13:36:52 Legolas.valinor systemd[1]: Started Discard unused blocks on filesystems from /etc/fstab. Legolas:~ # Information links: <https://superuser.com/questions/1546128/ssd-trim-on-archlinux/1546136#15461…> SSD Trim on Archlinux <https://wiki.archlinux.org/index.php/Solid_state_drive#dm-crypt> <https://wiki.debian.org/SSDOptimization#Mounting_SSD_filesystems> <https://blog.christophersmart.com/2013/06/05/trim-on-lvm-on-luks-on-ssd/> TRIM on LVM on LUKS on SSD - -- Cheers Carlos E. R. (from 15.1 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCX+nYrhwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfVgoQAnAue8M23DsVSZ8/Ahsvp T7e6D1nJAJ4pwLCXnBdHe/5VuVg+VGWHrtnO1Q== =d2w8 -----END PGP SIGNATURE-----

1 0

Tumbleweed Technical Concerns
by Kwame Boateng 27 Dec '20

27 Dec '20

Dear Opensuse Support, I'm concerned about some issues I've encountered while using Opensuse Tumbleweed. Is that something you can help me with? Kind regards, Kwame Boateng

2 1

cron job signatures script
by Marco Calistri 19 Dec '20

19 Dec '20

Hello, (Note: This is a generic Linux/Unix question not specifically related to Opensuse) Here a weird fact I face running a cron job from /etc/cron.hourly which it looks to trim away the final part of the text I aim to get. Below is the signature generated by script I add into the cron job: -- Marco Calistri Build: openSUSE Tumbleweed 20201216 Kernel: 5.9.14-1-default Desktop: Whereas here at the bottom the same script result if I run it from my console: -- Marco Calistri Build: openSUSE Tumbleweed 20201216 Kernel: 5.9.14-1-default Desktop: (Xfce 4.14) <----------------As you can see the script ran by the cron job it is missing this part. I suppose it could be caused by some variables settings which are different from my user environment and the ones used by the cron user. This is the script I use: #!/bin/sh echo "Marco Calistri" > /home/marco/.thunderbird/cy71896h.Marco/signatures/os-release.txt echo "Build: $(head -n 1 /etc/issue|awk -F'to' '{print $2}'|awk -F'-' '{print $1}')" >> /home/marco/.thunderbird/cy71896h.Marco/signatures/os-release.txt echo "Kernel: $(uname -r)" >> /home/marco/.thunderbird/cy71896h.Marco/signatures/os-release.txt echo "Desktop:$(xfce4-session --version |awk -F'xfce4-session' '{print $2}'|awk -F'2' '{print $2}')" >> /home/marco/.thunderbird/cy71896h.Marco/signatures/os-release.txt echo -ne "\n" >> /home/marco/.thunderbird/cy71896h.Marco/signatures/os-release.txt Thanks for any eventual suggestion you could provide to me. Regards, Marco

5 25

What exactly governs the Leap network stack when using network manager?
by Stakanov 19 Dec '20

19 Dec '20

I installed time ago unbound. After having done some adjustment (the dns resolution would be disturbed by settings of "automatic" in /etc/sysconfig "DNS Policy" (which i did set to "", and since then the "outages" I experience with unbound were gone) all was fine. Still rarely(!) it happens that something that seems "unrelated" to unbound "crashes". No error in unbound log, clean connection with the wlan, but no name resolution. A complete system restart does help. Only restarting unbound as service does not. So something that is in the middle between unbound and network does not function as it should (or at least I do not know how to restart it when it blocks). Normally I did sudo rcnetwork restart and that would restart the network. Now it appears that, although the network is restarted, the problem is not solved. Not even restarting unbound and rcnetwork. So what do I have to restart, some systemd thing? Last question: For what is /etc/sysconfig/"netconfig modules", with unbound, what should I insert there in yast, edit /etc/sysconfig as value for unbound? I guess: dns-resolver -dns-bind -dns-dnsmasq nis ntp-runtime (should do) but do I also "-" dns-resolver so all is done by unbound? or does the dns-resolver relies on unbound? What is the value to be set for unbound module: dns-unbound? (if anybody knows). Currently it is simply started by system services at startup, so maybe no extra value for it in etc/sysconfig is needed at all? Thanks in advance.

1 0

latest zypper refreshes repo list
by Peter Suetterlin 18 Dec '20

18 Dec '20

I just tried to look up some packages, as user. It failed because it wanted to refresh the repos, and of course didn't have write permission. Why this change? I've switched off autorefresh *on purpose*.

2 5

spam from lists.opensuse.org ?
by mauro 18 Dec '20

18 Dec '20

Hello, all my outbound emails to <listname>@lists.opensuse.org and certain inbound emails (the last ones are from Dan Čermák <dcermak(a)suse.com>, Gerald Pfeifer <gp(a)suse.com>, etc) end up flagged as spam by my mail provider (libero.it). On my side I've set no filter and blocked nobody. Libero claims responsible for the wrong flagging being the opensuse's servers. I attach their response to my complain. Please check it and let me know. Best Regards, Mauro

11 47

msr: Write to unrecognized MSR 0x17f
by Daniele 18 Dec '20

18 Dec '20

Hello guys, on TW, since kernel 5.9.* I see the error: msr: Write to unrecognized MSR 0x17f by mcelog Please report to x86(a)kernel.org I'm not the only one with this kind of error, it happens with different address and different application to other people. So, it should be a known issue.. Thanks, Daniele

3 3

[opensuse-support] what mechanism makes the preface to a vtty login prompt exist?
by Felix Miata 17 Dec '20

17 Dec '20

Google can't seem to point me to an answer. I did a fresh install of TW a month ago, 20201111 IIRC. On it the vtty prompts are naked, nothing but the prompt on the top line. /etc/issue.net exists, but /etc/issue does not, even after upgrade to 20201214. What else is missing? -- Evolution as taught in public schools, like religion, is based on faith, not on science. Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/

3 5

What is docker0 interface in firewalld configuration?
by Freek de Kruijf 16 Dec '20

16 Dec '20

Hi, I just started a newly installed Tumbleweed system, which now uses nftables instead of iptables for firewalld. The (new?) default zone for firewalld is public, so all incoming traffic on interfaces goes there if an interface is not connected to a zone. This is now the default; on my other system it is external. This zone "public" allows "services: dhcpv6-client ssh". I will remove ssh from this zone and put a number of source addresses in internal, which allows "services: dhcpv6-client mdns samba-client ssh", so also ssh. This restricts ssh access only to these addresses. I am puzzled about a new zone "docker", which shows up when I use "firewall-cmd --get-active-zones" I get zone "docker" and in it: "target: ACCEPT" and "interfaces: docker0". I can't find a /dev/docker0. What is docker0? -- fr.gr. member openSUSE Freek de Kruijf

2 1

[Leap 15] HP All-in-One printer setup
by Bob Williams 15 Dec '20

15 Dec '20

My system is Leap 15.2 and KDE Plasma desktop I am trying to setup a Hewlett-Packard Color Laserjet Pro MFP M477fnw all-in-one printer/scanner. Using YaST, the printer is seen and recognised, and the correct driver for this model is found. Everything works as expected. Using the YaST Scanner module, the scanner is not detected. I have tried using Add, which brings up a long list of scanner models, but not mine. Even if I select the HP generic driver for all-in-one devices, it loads the hpaio driver, but says 'No scanner recognized by this driver' in the Scanner configuration window. Using 'Run hp-setup' gets me to the same list of scanners. Actually, pretty nearly all HP scanners use the hpaio driver (it's part of hplip). What is so frustrating is that both printer and scanner were working perfectly a month ago, so something has changed recently. But what? I have tried connecting the M477 device both wirelessly and through a wired connection, but both give the same result. Something makes me think that YaST's scanner detection has broken. Can anyone confirm this or suggest a workaround? Bob -- Bob Williams System: Linux 5.3.18-lp152.57-default Desktop: KDE Frameworks: 5.71.0, Qt: 5.12.7 and Plasma: 5.18.5 https://useplaintext.email/

4 22