[Bug 1556175] [NEW] networking.service hangs on shutdown -- ifdown fails to kill dhclient

Martin Pitt martin.pitt at ubuntu.com
Fri Mar 11 16:08:10 UTC 2016 

Public bug reported:

On current xenial, LXD containers take ages to reboot:

lxc launch images:ubuntu/xenial/amd64 x1
lxc exec x1 reboot

"systemctl list-jobs" shows:

117 networking.service         stop  running

root at x1:~# systemctl status networking
● networking.service - Raise network interfaces
   Loaded: loaded (/lib/systemd/system/networking.service; enabled; vendor preset: enabled)
  Drop-In: /run/systemd/generator/networking.service.d
           └─50-insserv.conf-$network.conf
   Active: deactivating (stop-sigterm) since Fri 2016-03-11 16:03:51 UTC; 21s ago
     Docs: man:interfaces(5)
  Process: 187 ExecStop=/sbin/ifdown -a --read-environment (code=exited, status=0/SUCCESS)
  Process: 55 ExecStart=/sbin/ifup -a --read-environment (code=exited, status=0/SUCCESS)
  Process: 49 ExecStartPre=/bin/sh -c [ "$CONFIGURE_INTERFACES" != "no" ] && [ -n "$(ifquery --r
 Main PID: 55 (code=exited, status=0/SUCCESS)
    Tasks: 1 (limit: 512)
   CGroup: /system.slice/networking.service
           └─146 /sbin/dhclient -1 -v -pf /run/dhclient.eth0.pid -lf /var/lib/dhcp/dhclient.eth0

Mar 11 16:03:52 x1 dhclient[207]: For info, please visit https://www.isc.org/software/dhcp/
Mar 11 16:03:52 x1 dhclient[207]: 
Mar 11 16:03:52 x1 dhclient[207]: Listening on LPF/eth0/00:16:3e:5a:f5:97
Mar 11 16:03:52 x1 ifdown[187]: Listening on LPF/eth0/00:16:3e:5a:f5:97
Mar 11 16:03:52 x1 ifdown[187]: Sending on   LPF/eth0/00:16:3e:5a:f5:97
Mar 11 16:03:52 x1 ifdown[187]: Sending on   Socket/fallback
Mar 11 16:03:52 x1 dhclient[207]: Sending on   LPF/eth0/00:16:3e:5a:f5:97
Mar 11 16:03:52 x1 dhclient[207]: Sending on   Socket/fallback
Mar 11 16:03:52 x1 dhclient[207]: DHCPRELEASE on eth0 to 10.0.3.1 port 67 (xid=0x6480ef2b)
Mar 11 16:03:52 x1 ifdown[187]: DHCPRELEASE on eth0 to 10.0.3.1 port 67 (xid=0x6480ef2b)

The above is also the last journal line that I get for the usual 90s
timeout, then the container finally reboots.

Not many processes are actually running any more:

root         1  0.0  0.0  37076  5304 ?        Ss   16:03   0:00 /sbin/init
root        38  0.0  0.0  41572  3316 ?        Ss   16:03   0:00 /lib/systemd/systemd-udevd
root        41  0.0  0.0  35276  5668 ?        Ss   16:03   0:00 /lib/systemd/systemd-journald
root       146  0.0  0.0 181376  2664 ?        Ss   16:03   0:00 /sbin/dhclient -1 -v -pf /run/dhclient.eth0.pid -lf /var/lib/dhcp/dhclient.eth0.leases -I -df /var/lib/dhcp/dhclient6.eth0.leases eth0

but apparently the dhclient instance refuses to die (note that ifdown -a
itself is not running any more).

** Affects: ifupdown (Ubuntu)
     Importance: High
     Assignee: Martin Pitt (pitti)
         Status: Triaged

-- 
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to ifupdown in Ubuntu.
https://bugs.launchpad.net/bugs/1556175

Title:
  networking.service hangs on shutdown -- ifdown fails to kill dhclient

Status in ifupdown package in Ubuntu:
  Triaged

Bug description:
  On current xenial, LXD containers take ages to reboot:

  lxc launch images:ubuntu/xenial/amd64 x1
  lxc exec x1 reboot

  "systemctl list-jobs" shows:

  117 networking.service         stop  running

  root at x1:~# systemctl status networking
  ● networking.service - Raise network interfaces
     Loaded: loaded (/lib/systemd/system/networking.service; enabled; vendor preset: enabled)
    Drop-In: /run/systemd/generator/networking.service.d
             └─50-insserv.conf-$network.conf
     Active: deactivating (stop-sigterm) since Fri 2016-03-11 16:03:51 UTC; 21s ago
       Docs: man:interfaces(5)
    Process: 187 ExecStop=/sbin/ifdown -a --read-environment (code=exited, status=0/SUCCESS)
    Process: 55 ExecStart=/sbin/ifup -a --read-environment (code=exited, status=0/SUCCESS)
    Process: 49 ExecStartPre=/bin/sh -c [ "$CONFIGURE_INTERFACES" != "no" ] && [ -n "$(ifquery --r
   Main PID: 55 (code=exited, status=0/SUCCESS)
      Tasks: 1 (limit: 512)
     CGroup: /system.slice/networking.service
             └─146 /sbin/dhclient -1 -v -pf /run/dhclient.eth0.pid -lf /var/lib/dhcp/dhclient.eth0

  Mar 11 16:03:52 x1 dhclient[207]: For info, please visit https://www.isc.org/software/dhcp/
  Mar 11 16:03:52 x1 dhclient[207]: 
  Mar 11 16:03:52 x1 dhclient[207]: Listening on LPF/eth0/00:16:3e:5a:f5:97
  Mar 11 16:03:52 x1 ifdown[187]: Listening on LPF/eth0/00:16:3e:5a:f5:97
  Mar 11 16:03:52 x1 ifdown[187]: Sending on   LPF/eth0/00:16:3e:5a:f5:97
  Mar 11 16:03:52 x1 ifdown[187]: Sending on   Socket/fallback
  Mar 11 16:03:52 x1 dhclient[207]: Sending on   LPF/eth0/00:16:3e:5a:f5:97
  Mar 11 16:03:52 x1 dhclient[207]: Sending on   Socket/fallback
  Mar 11 16:03:52 x1 dhclient[207]: DHCPRELEASE on eth0 to 10.0.3.1 port 67 (xid=0x6480ef2b)
  Mar 11 16:03:52 x1 ifdown[187]: DHCPRELEASE on eth0 to 10.0.3.1 port 67 (xid=0x6480ef2b)

  The above is also the last journal line that I get for the usual 90s
  timeout, then the container finally reboots.

  Not many processes are actually running any more:

  root         1  0.0  0.0  37076  5304 ?        Ss   16:03   0:00 /sbin/init
  root        38  0.0  0.0  41572  3316 ?        Ss   16:03   0:00 /lib/systemd/systemd-udevd
  root        41  0.0  0.0  35276  5668 ?        Ss   16:03   0:00 /lib/systemd/systemd-journald
  root       146  0.0  0.0 181376  2664 ?        Ss   16:03   0:00 /sbin/dhclient -1 -v -pf /run/dhclient.eth0.pid -lf /var/lib/dhcp/dhclient.eth0.leases -I -df /var/lib/dhcp/dhclient6.eth0.leases eth0

  but apparently the dhclient instance refuses to die (note that ifdown
  -a itself is not running any more).

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/ifupdown/+bug/1556175/+subscriptions

More information about the foundations-bugs mailing list