Summary: | sys-apps/systemd-219-r2 - boot sometimes hangs - systemd-logind Failed to register name: Connection timed out | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Norman Back <gentoo3> |
Component: | [OLD] Core system | Assignee: | Gentoo systemd Team <systemd> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | tpfaff |
Priority: | Normal | Keywords: | PATCH |
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: |
dmesg.txt
journalctl -b -l journalctl -b -l after kill 1 in debug shell and boot completed journalctl -b -l from a successfully boot Patch to revert offending systemd commit. |
Description
Norman Back
2015-03-22 15:56:31 UTC
While attempting to collect log I found 1000's of lines containing 'systemd-udevd[878]: maximum number (24) of children reached' so I added 'udev.children-max=64' to the kernel options. This removed the noise but had no effect on thesystemd-logind issue. I have added logs from 'dmesg', 'journal -b -l' and 'journal -b -l' after using 'kill 1' in the debug shell. Created attachment 399464 [details]
dmesg.txt
Created attachment 399466 [details]
journalctl -b -l
Created attachment 399468 [details]
journalctl -b -l after kill 1 in debug shell and boot completed
This output is not from the same boot as the attached log but 'systemctl list-jobs' typically shows JOB UNIT TYPE STATE 279 xinetd.service start waiting 292 rpc-statd.service start waiting 256 gentoo-local-10-uam-poll-dvd-cd.service start waiting 305 remote-fs-pre.target start waiting 319 kdm.service start waiting 311 distccd.service start waiting 268 mysqld.service start waiting 280 systemd-user-sessions.service start waiting 261 rpcbind.target start waiting 312 nmbd.service start waiting 138 multi-user.target start waiting 283 gentoo-local-91-check-efibootmgr-setings.service start waiting 287 nfs-server.service start waiting 294 rpc-statd-notify.service start waiting 617 systemd-logind.service restart waiting 282 gentoo-local-90-grub2-set-default-0.service start waiting 296 systemd-networkd-wait-online.service start waiting 258 network.target start waiting 313 smbd.service start waiting 289 nfs-mountd.service start waiting 299 remote-fs.target start waiting 304 nfs-client.target start waiting 260 rpcbind.service start waiting 270 vmware.target start waiting 295 network-online.target start waiting 314 apache2.service start waiting 269 postfix.service start waiting 137 graphical.target start waiting 28 jobs listed. Created attachment 399472 [details]
journalctl -b -l from a successfully boot
For comparission I have added 'journalctl -b -l' from a successfully boot I found https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1423811 It looks close to my issue, so I reversed the update referred to ( http://cgit.freedesktop.org/systemd/systemd/patch/?id=13790add4bf648fed816361794d8277a75253410 ), re-emerged systemd with the revesed patch applied and created a new initramfs. With the patch applied I rebooted 40 times without hitting the systemd-logind issue. Without the patch the fail would occurr about every 5th boot. Created attachment 399566 [details, diff]
Patch to revert offending systemd commit.
BTW this issue occurs on most of my boxes, with varing frequency, including the mythtv server. :-( I have also install the revert patch on the mythtv server and will update this bug should the issue reoccur. i had the same problem on 3 different systems, 1 on virtualbox, 2 with real hardware. I solved it by downgrading to systemd-218 for now. The problem seems to occur on several distributions, see http://comments.gmane.org/gmane.comp.sysutils.systemd.devel/28525 and https://bugs.archlinux.org/task/44016 . As far as i can see, this does not happen with systemd-220-r2. Thanks for feedback |