Re: [systemd-devel] hinting systemd about what to do to make a device appear
also sprach Lennart Poettering[2015-09-11 14:16 +0200]: > I figure the next step is to attach gdb to PID, set a break point > in device_process_new() and figure out why systemd doesn't care > for the symlink. Frankly — and you won't like this — but I lack the resources to get involved at this level. My gdb days are a while in the past and in order to figure out what's going on, I'd need to spend a lot more time with the code, I'm afraid. I don't have the time for this. > I still have the suspicion that drbd does something weird with > /dev, and breaks things there... To detect that one option might > be to use Martin Pitt's fatrace tool on /dev and see what accesses > are made to /dev when the device is supposed to show up. Regarding drbd touching /dev, all I can say is that it /looks/ like it's doing everything you're asking it to do, via udev: http://slexy.org/raw/s20OEktMXW Using "fatrace -c" in /dev and starting drbd returns absolutely nothing, not a single line, which I find a bit suspcicious, or does fatrace not include udev/systemd-generated events? -- @martinkrafft | http://madduck.net/ | http://two.sentenc.es/ "da haben wir es also: eine kirchliche ordnung mit priesterschaft, theologie, kultus, sakrament; kurz, alles das, was jesus von nazareth bekämpft hatte..." - friedrich nietzsche spamtraps: madduck.bo...@madduck.net digital_signature_gpg.asc Description: Digital signature (see http://martin-krafft.net/gpg/sig-policy/999bbcc4/current) ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] hinting systemd about what to do to make a device appear
Отправлено с iPhone > 10 сент. 2015 г., в 20:21, martin f krafftнаписал(а): > > also sprach Michael Biebl [2015-09-10 19:13 +0200]: >> Could you run udevadm info on the device? >> Does it have a SYSTEMD_READY flag set? > > No: http://slexy.org/raw/s2VEGDwICN Missing SYSTEMD_READY means it is ready (by default) so it is ok. > -- > @martinkrafft | http://madduck.net/ | http://two.sentenc.es/ > > murphy's law is recursive. > washing your car to make it rain doesn't work. > > spamtraps: madduck.bo...@madduck.net > ___ > systemd-devel mailing list > systemd-devel@lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/systemd-devel ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] systemd 225: cannot override nspawn service
Hi Ruben, Thank you for your reply. I will test when systemd 226 will be available on archlinux. Best regards Christophe 2015-09-09 20:43 GMT+02:00 Ruben Kerkhof: > On Wed, Sep 9, 2015 at 11:48 AM, crequill wrote: > > 1/ I cannot login anymore on containers (/sbin/getatty exist in > container): > > Failed to get login PTY: No such file or directory > > No idea about your other issues, but I hit this one too and it is > fixed in systemd 2.26. > > Kind regards, > > Ruben > ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] hinting systemd about what to do to make a device appear
also sprach Andrei Borzenkov[2015-09-11 10:28 +0200]: > Please run "udevadm monitor --env", start drbd service to trigger > device creation and make output available. http://slexy.org/raw/s21alRk3ZD -- @martinkrafft | http://madduck.net/ | http://two.sentenc.es/ "if you have built castles in the air, your work need not be lost; that is where they should be. now put the foundations under them." -- henry david thoreau spamtraps: madduck.bo...@madduck.net digital_signature_gpg.asc Description: Digital signature (see http://martin-krafft.net/gpg/sig-policy/999bbcc4/current) ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
[systemd-devel] NetDev Vlan device copy parent mac
Hi, I have a working patch to copy the mac address from parent link to the netdev vlan device. Currently today when using the IP command this is the behavior seen and for our projects it is the behavior we would like to see for systemd-networkd also. In the systemd-networkd solution a mac address is generated using netdev_get_mac. I have implemented a patch the mimic this behavior, but as far as the matching goes, which solution would be best accepted for the device comparison? By providing the below key options, users can opt for this functionality or not. A key in the below format? Is the CopyParentMac option superfluous? [NetDev] Name=enp0s20f0.4094 Kind=vlan CopyParentMac=True Parent=enp0s20f0 [VLAN] Id=4094 Thanks Dave. smime.p7s Description: S/MIME cryptographic signature -- Intel Shannon Limited Registered in Ireland Registered Office: Collinstown Industrial Park, Leixlip, County Kildare Registered Number: 308263 Business address: Dromore House, East Park, Shannon, Co. Clare This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies. ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] hinting systemd about what to do to make a device appear
also sprach Lennart Poettering[2015-09-11 12:30 +0200]: > Do I see this right, this is luks on lvm on drbd? Yuck! luks on drbd, and the local volume is on lvm, yes. > Which symlink are you looking for? the drbd one? Right. /dev/drbd/by-res/backup -- @martinkrafft | http://madduck.net/ | http://two.sentenc.es/ "i doubt larry wall ever uses strict." -- frederick heckel spamtraps: madduck.bo...@madduck.net digital_signature_gpg.asc Description: Digital signature (see http://martin-krafft.net/gpg/sig-policy/999bbcc4/current) ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] hinting systemd about what to do to make a device appear
On Fri, 11.09.15 12:45, martin f krafft (madd...@madduck.net) wrote: > also sprach Lennart Poettering[2015-09-11 12:30 > +0200]: > > Do I see this right, this is luks on lvm on drbd? Yuck! > > luks on drbd, and the local volume is on lvm, yes. > > > Which symlink are you looking for? the drbd one? > > Right. /dev/drbd/by-res/backup Hmm, not sure what is going on there. I don't know drbd. I figure the next step is to attach gdb to PID, set a break point in device_process_new() and figure out why systemd doesn't care for the symlink. There's a loop after calling udev_device_get_devlinks_list_entry() and it would be interesting to see if the symlinks pop up there and why systemd doesn't process them... Maybe the stat() verification step in the loop is actually what fails here? I still have the suspicion that drbd does something weird with /dev, and breaks things there... To detect that one option might be to use Martin Pitt's fatrace tool on /dev and see what accesses are made to /dev when the device is supposed to show up. Lennart -- Lennart Poettering, Red Hat ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] hinting systemd about what to do to make a device appear
On Fri, 11.09.15 10:41, martin f krafft (madd...@madduck.net) wrote: > also sprach Andrei Borzenkov[2015-09-11 10:28 +0200]: > > Please run "udevadm monitor --env", start drbd service to trigger > > device creation and make output available. > > http://slexy.org/raw/s21alRk3ZD Do I see this right, this is luks on lvm on drbd? Yuck! Which symlink are you looking for? the drbd one? Lennart -- Lennart Poettering, Red Hat ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] hostname can be changed without permission checks
On Sat, 12 Sep 2015, Michał Zegan wrote: Hello. It seems that I am able to change a hostname with hostnamectl set-hostname name without any problems, even logged in as unprivileged user, and I did not get any authentication requests. I did not modify polkit rules to allow this, not sure about the default ones, but they probably shouldn't allow that, just checked that implicit rules are auth_admin_keep, arch does not have vendor rules and I also do not have my own.. Did you check both /etc/polkit-1/rules.d/ and /usr/share/polkit-1/rules.d/? On my system (Fedora), gnome-control-center has added a rule to the latter directory to allow a local user set the hostname, locale, etc., if they are in the "wheel" group. Perhaps you have something similar? You can test whether PolicyKit is allowing the action with: pkcheck --action-id org.freedesktop.hostname1.set-hostname \ --process $$ --allow-user-interaction If this exits successfully, then it's something in your PolicyKit configuration allowing the action, not systemd. - Michael___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
[systemd-devel] hostname can be changed without permission checks
Hello. It seems that I am able to change a hostname with hostnamectl set-hostname name without any problems, even logged in as unprivileged user, and I did not get any authentication requests. I did not modify polkit rules to allow this, not sure about the default ones, but they probably shouldn't allow that, just checked that implicit rules are auth_admin_keep, arch does not have vendor rules and I also do not have my own.. ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
[systemd-devel] tmp.mount inactive => /tmp not mounted
I recently switched to using systemd in my initrd, and nearly everything works fine, expect now the system comes up without /tmp being mounted correctly. I'm not sure where to start looking. Can anyone nudge me in the right direction? The tmp.mount unit seems to be inactive. It didn't do that before I started using systemd in initrd. ~ # systemctl status tmp.mount ● tmp.mount - Temporary Directory Loaded: loaded (/usr/lib/systemd/system/tmp.mount; static; vendor preset: enabled) Active: inactive (dead) since Sat 2015-09-12 00:28:11 UTC; 33s ago Where: /tmp What: tmpfs Docs: man:hier(7) http://www.freedesktop.org/wiki/Software/systemd/APIFileSystems The local-fs.target unit is active and happy. ~ # systemctl status local-fs.target ● local-fs.target - Local File Systems Loaded: loaded (/usr/lib/systemd/system/local-fs.target; static; vendor preset: enabled) Active: active since Sat 2015-09-12 00:28:12 UTC; 1min 24s ago Docs: man:systemd.special(7) But its dependencies list tmp.mount as not active. ~ # systemctl list-dependencies local-fs.target local-fs.target ● ├─-.mount ● ├─systemd-remount-fs.service ● ├─tmp.mount ● └─var.mount The list of mounts show that tmpfs is mounted on /tmp, but it isn't really. I assume this is a /tmp in initrd that is masked by switching the root to /sysroot (similar to the rootfs on the first line), because it is read-only like the root file system (can't make any files in it), and when I manually start tmp.mount that problem is fixed. ~ # cat /proc/mounts rootfs / rootfs rw 0 0 sysfs /sys sysfs rw,nosuid,nodev,noexec,relatime 0 0 proc /proc proc rw,nosuid,nodev,noexec,relatime 0 0 devtmpfs /dev devtmpfs rw,nosuid,size=954032k,nr_inodes=238508,mode=755 0 0 tmpfs /dev/shm tmpfs rw,nosuid,nodev 0 0 devpts /dev/pts devpts rw,nosuid,noexec,relatime,gid=5,mode=620 0 0 tmpfs /run tmpfs rw,nosuid,nodev,mode=755 0 0 tmpfs /sys/fs/cgroup tmpfs ro,nosuid,nodev,noexec,mode=755 0 0 cgroup /sys/fs/cgroup/systemd cgroup rw,nosuid,nodev,noexec,relatime,xattr,release_agent=/usr/lib/systemd/systemd-cgroups-agent,name=systemd 0 0 tmpfs /tmp tmpfs rw 0 0 /dev/disk/by-partlabel/rootfs / ext4 ro,relatime,data=ordered 0 0 /dev/disk/by-partlabel/varfs /var ext4 rw,relatime,discard,nodelalloc,data=journal 0 0 /etc/fstab is pretty tame. Not sure if it would be causing an issue or not. ~ # cat /etc/fstab PARTLABEL=rootfs / ext4 ro 0 1 PARTLABEL=varfs /var ext4 rw,data=journal,discard 0 2 Some journal output for reference. The root file system is read-only, so the errors with chmod are expected. These errors don't occur when /tmp is mounted properly. systemd[1]: systemd 219 running in system mode. (-PAM -AUDIT -SELINUX -IMA -APPARMOR -SMACK -SYSVINIT -UTMP -LIBCRYPTSETUP -GCRYPT -GNUTLS -ACL -XZ -LZ4 -SECCOMP +BLKID -ELFUTILS -KMOD -IDN) systemd[1]: Started Remount Root and Kernel File Systems. systemd[1]: Reached target Local File Systems (Pre). systemd[1]: Starting Local File Systems (Pre). systemd[1]: Found device SILICONSYSTEMS_INC_8GB varfs. systemd[1]: Mounting /var... systemd[1]: Mounted /var. systemd[1]: Starting Flush Journal to Persistent Storage... systemd[1]: Started Flush Journal to Persistent Storage. systemd[1]: Reached target Local File Systems. systemd[1]: Starting Local File Systems. systemd[1]: Starting Create Volatile Files and Directories... systemd-tmpfiles[161]: chmod(/tmp) failed: Read-only file system systemd[1]: systemd-tmpfiles-setup.service: main process exited, code=exited, status=1/FAILURE systemd[1]: Failed to start Create Volatile Files and Directories. systemd[1]: Unit systemd-tmpfiles-setup.service entered failed state. systemd[1]: systemd-tmpfiles-setup.service failed. systemd[1]: Reached target System Initialization. ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
[systemd-devel] systemctl -M foo operates on host not container foo
This sounds silly but I do think this used to work for me (current: v225 on Arch, perhaps something broke?) > systemctl -M foo enable sshd.service creates the symlink on the host, not in the container. Other sub-commands also work on the host, not the container. machine foo doesn’t actually exist. If it exists, the behavior is still the same. The machine name argument is completely ignored. What am I missing? Actually just found an old v222. That produces a dbus error if machine foo does not exist. That sounds more reasonable. Cheers, Johannes. ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] systemd and intltool
On Thu, 10.09.15 19:10, Michael Biebl (mbi...@gmail.com) wrote: > Hi, > > reading https://wiki.gnome.org/Projects/GnomeCommon/Migration, it says > that intltool is practically dead and one should use gettext directly. > > Do we still need intltool in systemd? Does gettext have support for > translating PolicyKit policy files? Happy to take a patch that removes the intltool hookup if it replaces it with the right gettext hookup instead. Lennart -- Lennart Poettering, Red Hat ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel