named stop to work

pppplus · Mar 12, 2017

Hi

I've a strange and new problem.

named is broken from yesterday (I just a reboot the VPS).
After rebooting, no problem, named is ok.

But, some minuts/hours after, I receive a notification.

Code:

service named status
Redirecting to /bin/systemctl status  named.service
* named.service - Berkeley Internet Name Domain (DNS)
   Loaded: loaded (/etc/systemd/system/named.service; enabled; vendor preset: disabled)
   Active: [COLOR="#FF0000"][B]failed[/B][/COLOR] (Result: timeout) since Sun 2017-03-12 11:03:06 CET; 44s ago
  Process: 16978 ExecStart=/usr/sbin/named -u named $OPTIONS (code=exited, status=0/SUCCESS)
  Process: 16975 ExecStartPre=/usr/sbin/named-checkconf -z /etc/named.conf (code=exited, status=0/SUCCESS)

Mar 12 11:01:36 host1v104.hb50.fr named[16979]: zone kidnet.fr/IN: sending notifies (serial 2016022500)
Mar 12 11:01:36 host1v104.hb50.fr named[16979]: zone presta-modules.fr.cr/IN: sending notifies (serial 2017021701)
Mar 12 11:01:36 host1v104.hb50.fr named[16979]: zone echecs.ovh/IN: sending notifies (serial 2016112403)
Mar 12 11:01:36 host1v104.hb50.fr named[16979]: zone zzzovniz.fr/IN: sending notifies (serial 2015120205)
Mar 12 11:01:36 host1v104.hb50.fr named[16979]: zone monptp.fr.cr/IN: sending notifies (serial 2016052401)
Mar 12 11:01:36 host1v104.hb50.fr named[16979]: zone presta-modules.fr.nf/IN: sending notifies (serial 2017021701)
Mar 12 11:03:06 host1v104.hb50.fr systemd[1]: named.service start operation timed out. Terminating.
Mar 12 11:03:06 host1v104.hb50.fr systemd[1]: Failed to start Berkeley Internet Name Domain (DNS).
Mar 12 11:03:06 host1v104.hb50.fr systemd[1]: Unit named.service entered failed state.
Mar 12 11:03:06 host1v104.hb50.fr systemd[1]: named.service failed.

When I run :
./build rewrite_confs

named is ok :

Code:

service named status
Redirecting to /bin/systemctl status  named.service
* named.service - Berkeley Internet Name Domain (DNS)
   Loaded: loaded (/etc/systemd/system/named.service; enabled; vendor preset: disabled)
   Active: activating (start) since Sun 2017-03-12 11:04:01 CET; 49s ago
   CGroup: /system.slice/named.service
           `-18038 /usr/sbin/named -u named

Mar 12 11:04:01 host1v104.hb50.fr systemd[1]: PID file /run/named/named.pid not readable (yet?) after start.
Mar 12 11:04:01 host1v104.hb50.fr named[18038]: running
Mar 12 11:04:01 host1v104.hb50.fr named[18038]: zone kidnet.fr/IN: sending notifies (serial 2016022500)
Mar 12 11:04:01 host1v104.hb50.fr named[18038]: zone ovniz.fr/IN: sending notifies (serial 2015120900)
Mar 12 11:04:01 host1v104.hb50.fr named[18038]: zone monptp.fr.cr/IN: sending notifies (serial 2016052401)
Mar 12 11:04:01 host1v104.hb50.fr named[18038]: zone host1v104.hb50.fr/IN: sending notifies (serial 2015111300)
Mar 12 11:04:01 host1v104.hb50.fr named[18038]: zone zzzovniz.fr/IN: sending notifies (serial 2015120205)
Mar 12 11:04:01 host1v104.hb50.fr named[18038]: zone presta-modules.fr.cr/IN: sending notifies (serial 2017021701)
Mar 12 11:04:01 host1v104.hb50.fr named[18038]: zone echecs.ovh/IN: sending notifies (serial 2016112403)
Mar 12 11:04:01 host1v104.hb50.fr named[18038]: zone presta-modules.fr.nf/IN: sending notifies (serial 2017021701)

When I run
service named restart

It's very long, and I've the message

Code:

Redirecting to /bin/systemctl restart  named.service
Job for named.service failed because a timeout was exceeded. See "systemctl status named.service" and "journalctl -xe" for details.

systemctl status named.service show named is actually running

Have you an idea about my problem ?
Where to search to find the problem ?

I'm on Centos 7.3

Thanks

Sergey · Mar 12, 2017

Hi,

Try:

Code:

service named configtest

pppplus · Mar 12, 2017

Thanks for reply:

The service command supports only basic LSB actions (start, stop, restart, try-restart, reload, force-reload, status). For other actions, please try to use systemctl.

And with systemctl, not better :
/bin/systemctl configtest named.service
Unknown operation 'configtest'.

pppplus · Mar 12, 2017

in /run/named, there is no file named.pid

So, the problem seems to be the file is never created here.

Is it a permission problem ?

Sergey · Mar 12, 2017

or please try this one:

Code:

named-checkconf /etc/named.conf

and if bind is running in chroot environment try this:

Code:

named-checkconf -t /var/named/chroot /etc/named.conf

pppplus · Mar 12, 2017

I add -z, otherwhile, I've no answer :

Code:

/named-checkconf -z /etc/named.conf
zone localhost.localdomain/IN: loaded serial 0
zone localhost/IN: loaded serial 0
zone 1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa/IN: loaded serial 0
zone 1.0.0.127.in-addr.arpa/IN: loaded serial 0
zone 0.in-addr.arpa/IN: loaded serial 0
zone host1v104.hb50.fr/IN: loaded serial 2015111300
zone ovniz.fr/IN: loaded serial 2015120900
zone kidnet.fr/IN: loaded serial 2016022500
zone monptp.fr.cr/IN: loaded serial 2016052401
zone presta-modules.fr.nf/IN: loaded serial 2017021701
zone echecs.ovh/IN: loaded serial 2016112403
zone presta-modules.fr.cr/IN: loaded serial 2017021701

Sergey · Mar 12, 2017

There is no problem with your configuration.
Maybe other guys could help you.
I found this solution in one forum but i am not sure about it.

I managed to resolve this by changing one line in /etc/named.conf:

named='named'
To -->
named='named -u named'

pppplus · Mar 12, 2017

I've no line named='named' in /etc/named.conf.

When I run :
#named -fg
errors are :

Code:

12-Mar-2017 14:06:38.354 open: /etc/rndc.key: permission denied
12-Mar-2017 14:06:38.354 couldn't add command channel 127.0.0.1#953: permission denied
12-Mar-2017 14:06:38.354 open: /etc/rndc.key: permission denied
12-Mar-2017 14:06:38.354 couldn't add command channel ::1#953: permission denied
12-Mar-2017 14:06:38.354 not using config file logging statement for logging due to -g option
isc_file_isplainfile 'data/named.run' failed: permission denied
12-Mar-2017 14:06:38.354 checking logging configuration failed: permission denied
12-Mar-2017 14:06:38.354 loading configuration: permission denied
12-Mar-2017 14:06:38.354 exiting (due to fatal error)

pppplus · Mar 14, 2017

Anyone for an idea ?

Permissions was root:root for /run/named

I change permissions to :
drwxr-xr-x 2 named root 80 Mar 14 10:58 named

It works (it works also with named / named)

Can it be a problem for something else in the system ?
Is it better named:root or named:named ?

Thanks

zEitEr · Mar 15, 2017

Your changes of permissions on /run/named won't survive a reboot, if /run is mounted as tmpfs:

Code:

mount | grep ' /run '
tmpfs on /run type [B]tmpfs[/B] (rw,nosuid,noexec,relatime,size=404080k,nr_inodes=505099,mode=755)

you might need to change directory in named.conf from /run/named to something different

pppplus · Mar 15, 2017

Yes you're right, permission comes back to root:root after restarting the container.
Do you know why I've this problem ?

Richard G · Mar 15, 2017

I found this was a bug a while ago already in Fedora, but exact the same error notice. Maybe they fixed it, but this was the solution, maybe you can try it, add this to your named.conf:

Code:

pid-file      "/var/run/named/named.pid";

pppplus · Mar 16, 2017

J'ai déjà cette ligne (plus exactement pid-file "/run/named/named.pid"

/var/run renvoie vers /run

=> j'avais déjà testé en mettant /var devant également, mais bien sûr sans changement

zEitEr · Mar 16, 2017

Google Translate:

I have already this line (more precisely pid-file "/run/named/named.pid"
/var/run returns to /run => I already tested by putting /var in front of it, but of course without change

So it seems /var/run/ is a simlink for /run/.

CentOS 7 has this:

Code:

# ls -la /var/run
lrwxrwxrwx. 1 root root 6 Jul  9  2014 /var/run -> ../run

In this case feel free to change it to /var/named/run/ (you might need to create it first and change ownership).

pppplus · Mar 16, 2017

Sorry, I answer previously in french !
My answer was :

I've already this line (exactly : pid-file "/run/named/named.pid"
/var/run is a symlink to /run

=> I'd already test with /var but it doesn't solve the problem.

With /var/run instead /run, always the same errors:

Code:

# service named status
PID file /var/run/named/named.pid not readable (yet?) after start.


# service named restart
Redirecting to /bin/systemctl restart  named.service
Job for named.service failed because a timeout was exceeded. See "systemctl status named.service" and "journalctl -xe" for details.

Note, I've also change /etc/systemd/system/named.service, adding /var in the PID path

named stop to work

pppplus

Verified User

Sergey

Verified User

pppplus

Verified User

pppplus

Verified User

Sergey

Verified User

pppplus

Verified User

Sergey

Verified User

pppplus

Verified User

pppplus

Verified User

zEitEr

Super Moderator

pppplus

Verified User

Richard G

Verified User

pppplus

Verified User

zEitEr

Super Moderator

pppplus

Verified User