ntp problem after snapshot 20160905

phil524 · September 8, 2016, 6:21pm

Hello,

after the snapchot installation I restarted the PC and received a start error for ntp

# systemctl status ntpd.service
● ntpd.service - NTP Server Daemon
   Loaded: loaded (/usr/lib/systemd/system/ntpd.service; enabled; vendor preset: disabled)
  Drop-In: /run/systemd/generator/ntpd.service.d
           └─50-insserv.conf-$time.conf
   Active: failed (Result: start-limit) since Thu 2016-09-08 17:51:11 CEST; 30s ago
     Docs: man:ntpd(1)
  Process: 3368 ExecStart=/usr/sbin/start-ntpd start (code=exited, status=0/SUCCESS)

Sep 08 17:51:11 hpprol2 systemd[1]: Failed to start NTP Server Daemon.
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Unit entered failed state.
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Failed with result 'resources'.
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Service hold-off time over, scheduling restart.
Sep 08 17:51:11 hpprol2 systemd[1]: Stopped NTP Server Daemon.
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Start request repeated too quickly.
Sep 08 17:51:11 hpprol2 systemd[1]: Failed to start NTP Server Daemon.
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Unit entered failed state.
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Failed with result 'start-limit'.

I have uninstalled ntp package and reinstalled after removing all files/directories related to ntp in /etc /var/lib. I reconfigured the ntp server in Yast but on exit I received an error.
trying to start ntpd via “systemctl start ntpd” gives still the error

journalctl -xe
Sep 08 17:51:11 hpprol2 ntpd[3379]: ntpd 4.2.8p8@1.3265-o Mon Sep  5 20:41:38 UTC 2016 (1): Starting
Sep 08 17:51:11 hpprol2 ntpd[3379]: Command line: /usr/sbin/ntpd -p /var/run/ntp/ntpd.pid -g -u ntp:ntp -i /var/lib/ntp -c /etc/ntp.conf
Sep 08 17:51:11 hpprol2 ntpd[3380]: proto: precision = 0.140 usec (-23)
Sep 08 17:51:11 hpprol2 ntpd[3380]: switching logging to file /var/log/ntp
Sep 08 17:51:11 hpprol2 ntp_intres[3381]: Cannot reopen log file /var/log/ntp, Permission denied
Sep 08 17:51:11 hpprol2 ntp_intres[3381]: blocking_getaddrinfo can not queue response
Sep 08 17:51:11 hpprol2 start-ntpd[3368]: Starting network time protocol daemon (NTPD)
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: PID 3380 read from file /var/run/ntp/ntpd.pid does not exist or is a zombie.
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Daemon never wrote its PID file. Failing.
Sep 08 17:51:11 hpprol2 systemd[1]: Failed to start NTP Server Daemon.
-- Subject: Unit ntpd.service has failed
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit ntpd.service has failed.
-- 
-- The result is failed.                                                                                                                                                                 
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Unit entered failed state.                                                                                                             
Sep 08 17:51:11 hpprol2 audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=ntpd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Failed with result 'resources'.                                                                                                        
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Service hold-off time over, scheduling restart.                                                                                        
Sep 08 17:51:11 hpprol2 systemd[1]: Stopped NTP Server Daemon.                                                                                                                           
-- Subject: Unit ntpd.service has finished shutting down                                                                                                                                 
-- Defined-By: systemd                                                                                                                                                                   
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel                                                                                                                  
--                                                                                                                                                                                       
-- Unit ntpd.service has finished shutting down.                                                                                                                                         
Sep 08 17:51:11 hpprol2 audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=ntpd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Start request repeated too quickly.                                                                                                    
Sep 08 17:51:11 hpprol2 audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=ntpd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=s
Sep 08 17:51:11 hpprol2 systemd[1]: Failed to start NTP Server Daemon.
-- Subject: Unit ntpd.service has failed
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit ntpd.service has failed.
-- 
-- The result is failed.
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Unit entered failed state.
Sep 08 17:51:11 hpprol2 systemd[1]: ntpd.service: Failed with result 'start-limit'.

I tried to solve the permission error about /var/log/ntp

cd /var/log
hpprol2:/var/log # ls -ld ntp
-rw-r--r-- 1 root root 6630  8 sep 17:56 ntp
hpprol2:/var/log # chown ntp:ntp ntp

I restarted ntp via “systemctl start ntpd” but the same error occurred only the problem about /var/log/ntp seems solved

# journalctl -xe         
Sep 08 18:10:38 hpprol2 systemd[1]: Stopped NTP Server Daemon.
-- Subject: Unit ntpd.service has finished shutting down
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit ntpd.service has finished shutting down.
Sep 08 18:10:38 hpprol2 audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=ntpd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=
Sep 08 18:10:38 hpprol2 audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=ntpd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=s
Sep 08 18:10:38 hpprol2 systemd[1]: Starting NTP Server Daemon...
-- Subject: Unit ntpd.service has begun start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit ntpd.service has begun starting up.
Sep 08 18:10:38 hpprol2 ntpd[4277]: ntpd 4.2.8p8@1.3265-o Mon Sep  5 20:41:38 UTC 2016 (1): Starting
Sep 08 18:10:38 hpprol2 ntpd[4277]: Command line: /usr/sbin/ntpd -p /var/run/ntp/ntpd.pid -g -u ntp:ntp -i /var/lib/ntp -c /etc/ntp.conf
Sep 08 18:10:38 hpprol2 ntpd[4278]: proto: precision = 0.140 usec (-23)
Sep 08 18:10:38 hpprol2 ntpd[4278]: switching logging to file /var/log/ntp
Sep 08 18:10:38 hpprol2 start-ntpd[4266]: Starting network time protocol daemon (NTPD)
Sep 08 18:10:38 hpprol2 systemd[1]: ntpd.service: PID 4278 read from file /var/run/ntp/ntpd.pid does not exist or is a zombie.
Sep 08 18:10:38 hpprol2 systemd[1]: ntpd.service: Daemon never wrote its PID file. Failing.
Sep 08 18:10:38 hpprol2 systemd[1]: Failed to start NTP Server Daemon.
-- Subject: Unit ntpd.service has failed
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit ntpd.service has failed.
-- 
-- The result is failed.
Sep 08 18:10:38 hpprol2 systemd[1]: ntpd.service: Unit entered failed state.
Sep 08 18:10:38 hpprol2 audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=ntpd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=
Sep 08 18:10:38 hpprol2 systemd[1]: ntpd.service: Failed with result 'resources'.
Sep 08 18:10:38 hpprol2 systemd[1]: ntpd.service: Service hold-off time over, scheduling restart.
Sep 08 18:10:38 hpprol2 systemd[1]: Stopped NTP Server Daemon.
-- Subject: Unit ntpd.service has finished shutting down
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit ntpd.service has finished shutting down.
Sep 08 18:10:38 hpprol2 audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=ntpd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=
Sep 08 18:10:38 hpprol2 systemd[1]: ntpd.service: Start request repeated too quickly.
Sep 08 18:10:38 hpprol2 audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=ntpd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=s
Sep 08 18:10:38 hpprol2 systemd[1]: Failed to start NTP Server Daemon.
-- Subject: Unit ntpd.service has failed
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit ntpd.service has failed.
-- 
-- The result is failed.
Sep 08 18:10:38 hpprol2 systemd[1]: ntpd.service: Unit entered failed state.

Any Idea?

Many thanks in advance
Philippe

xorbe · September 8, 2016, 6:58pm

Is this what is causing a new ~30 delay before x.org pops up? (using vbox)

wolfi323 · September 8, 2016, 7:23pm

Maybe, this has been mentioned here too:
http://lists.opensuse.org/opensuse-factory/2016-09/msg00181.html

Try to disable the ntpd service and see if it helps…

No idea how to fix the OP’s problem though, i.e. how to get ntp to work again.

eng-int · September 8, 2016, 7:49pm

I am not sure how relevant this is, but …

I just used YaST to update ntp-4.2.8p8-1.3 from TW “20160831” to ntp-4.2.8p8-1.4 then to ntp-4.2.8p8-2.1 (different mirrors)
Both these updates reported time-outs after failing permissions to write the pid and ntp.log.
I then tried ntp-4.2.8p8-24.1 from the Leap-42.1 oss update repository.
The 24.1 (Leap) package started instantly with no errors…

Absolutely NO guarantees – but it might be better than nothing meantime.

phil524 · September 8, 2016, 8:57pm

Thanks it worked for me also. I installed the leap42.2 version ntp-4.2.8p8-1.2.x86_64.rpm

# systemctl status ntpd
● ntpd.service - NTP Server Daemon
   Loaded: loaded (/usr/lib/systemd/system/ntpd.service; enabled; vendor preset: disabled)
  Drop-In: /run/systemd/generator/ntpd.service.d
           └─50-insserv.conf-$time.conf
   Active: active (running) since Thu 2016-09-08 20:55:43 CEST; 10s ago
     Docs: man:ntpd(1)
  Process: 10292 ExecStart=/usr/sbin/start-ntpd start (code=exited, status=0/SUCCESS)
 Main PID: 10305 (ntpd)
    Tasks: 2 (limit: 512)
   CGroup: /system.slice/ntpd.service
           ├─10305 /usr/sbin/ntpd -p /var/run/ntp/ntpd.pid -g -u ntp:ntp -i /var/lib/ntp -c /etc/ntp.conf
           └─10306 ntpd: asynchronous dns resolver

Sep 08 20:55:43 hpprol2 systemd[1]: Starting NTP Server Daemon...
Sep 08 20:55:43 hpprol2 ntpd[10304]: ntpd 4.2.8p8@1.3265-o Sun Aug 28 14:48:12 UTC 2016 (1): Starting
Sep 08 20:55:43 hpprol2 ntpd[10304]: Command line: /usr/sbin/ntpd -p /var/run/ntp/ntpd.pid -g -u ntp:ntp -i /var/lib/ntp -c /etc/ntp.conf
Sep 08 20:55:43 hpprol2 ntpd[10305]: proto: precision = 0.136 usec (-23)
Sep 08 20:55:43 hpprol2 ntpd[10305]: switching logging to file /var/log/ntp
Sep 08 20:55:43 hpprol2 start-ntpd[10292]: Starting network time protocol daemon (NTPD)
Sep 08 20:55:43 hpprol2 systemd[1]: Started NTP Server Daemon.

I think that I’ll open a bug for this problem.

Regards
Philippe

dth2 · September 8, 2016, 9:56pm

Same long ntp delay here. I switched off the auto-start ntpd option in yast services manager and booting is back to normal time. No adverse effects so far.

wolfi323 · September 8, 2016, 10:50pm

The ntp update has been reverted, the next snapshot will again only contain the previous working version.
That one is still in the repo btw, you can just downgrade it manually for now.

See http://lists.opensuse.org/opensuse-factory/2016-09/msg00185.html

Btw, if you just want to keep the time on your system synchronized, you can also configure that to be done every 15 minutes via a cronjob (in YaST->Date and Time or YaST->NTP), there’s no need to run the ntpd service for that.

eng-int · September 8, 2016, 10:59pm

Your reply gave me the confidence to try a “zypper up”. The Leap ntp was retained as it has a higher version number.
The only problem is that all the user passwords were scrambled for reasons unknown. A rescue boot from install media let me reset the passwords before a reboot and so far everything seems normal.

labouie · September 9, 2016, 2:27am

I’m getting the same issue. I’ll simply wait for the next snapshot for a fix.

eng-int · September 9, 2016, 3:26am

Just in case anyone is not sure how to do that:


#  zypper in --oldpackage  http://download.opensuse.org/tumbleweed/repo/oss/suse/x86_64/ntp-4.2.8p8-1.3.x86_64.rpm

jpluimers · September 9, 2016, 10:34am

I had already updated to snapshot 20160905 which failed ntp. Then I zypper upgraded but that was to snapshot 20160907 (as 20160908 isn’t finished in OpenQA yet), so I had to manually downgrade or wait for 2160908 to be finished:
zypper in --oldpackage http://download.opensuse.org/tumbleweed/repo/oss/suse/x86_64/ntp-4.2.8p8-1.3.x86_64.rpm
systemctl restart ntpd.service–jeroen

wolfi323 · September 9, 2016, 2:50pm

Or use YaST, select the ntp package and click on “Versions” below the package list, then choose the working one.

suse_rasputin · September 9, 2016, 3:06pm

Many thanks for the zypper code, as in Yast there is no older version…

phil524 · September 9, 2016, 9:18pm

hello;

just installed snapshot 20160907.
It contained a new kernel and ntp.
After reboot I cannot boot directly to KDE. I receive a terminal and the last message displayed is “Starting terminate Plymouth boot screen”.
Login in as root I could start the GUI (KDE) via “init 5”
Reinstalling ntp-4.2.8p8-1.4.x86_64.rpm dis not solve the problem
I tried to disable ntpd via “systemctl disable ntpd” but after reboot still no GUI. >:)

Regards
Philippe

wolfi323 · September 9, 2016, 9:27pm

Hm, if zypper finds the older version, YaST should show it too.
But you need to click on “Versions” below the package list to see all available versions, as I wrote.

And what do these commands say?

systemctl get-default
systemctl status display-manager

suse_rasputin · September 9, 2016, 9:36pm

Hi!

Sorry, but I have a screenshot here, in my Tumbleweeg in Software Management for NTP there is under “Versions” only ONE entry. Was also only ONE entry before I downdated with the Zypper code given above…

No idea what’s going on…

wolfi323 · September 9, 2016, 10:00pm

Well, apparently the older packages are not listed in the repo metadata.
So zypper wouldn’t find them either.

But the mentioned zypper command contains the full URL (I forgot that when I wrote my previous reply), so it could download the package from the repo without knowing that this version exists.

Btw, the link I posted before that contains an rpm command that does the same.

I reverted ntp in openSUSE:Factory - the snapshot 0908 will have this
resolved.

Users can:

lock ntp before doing the update to not run into the issue

If you already DID update, you can revert to the last published version
using:

rpm -Uhv --oldpackage http://download.opensuse.org/tumbleweed/repo/oss/suse/x86_64/ntp-4.2.8p8-1.4.x86_64.rpm

(all on one line, in case the mails do some nasty breaks)

This should get you back going.

The snapshot 0908 will have this corrected version of ntp in the repo
again.

eng-int · September 9, 2016, 10:38pm

YaST does not “see“ the other versions because they are not listed in the repository metadata. Presumably they will come back when the repository is reconstructed again.


┌─────────────────────────────────────────┐                       
│     │Name│     │Version    │Repository│Size        │Architecture    │                       
│  i  │ntp │     │4.2.8p8-1.3│@System   │     2.3 MiB│x86_6            │                       
│     │ntp │ x   │4.2.8p8-2.1│Tumbleweed│     2.3 MiB│x86_6         │                       
│     │ntp │     │4.2.8p8-2.1│Tumbleweed│     2.3 MiB│i586             │
└─────────────────────────────────────────┘

The manual zypper command works because it specifies the particular RPM file as if *** was not in a repository – which strictly it is not, it is just an RPM that happens to be in the same folder as the repository RPMs.

wolfi323 · September 10, 2016, 12:49am

eng-int:

YaST does not “see“ the other versions because they are not listed in the repository metadata. Presumably they will come back when the repository is reconstructed again.
┌─────────────────────────────────────────┐                       
│     │Name│     │Version    │Repository│Size        │Architecture    │                       
│  i  │ntp │     │4.2.8p8-1.3│@System   │     2.3 MiB│x86_6            │                       
│     │ntp │ x   │4.2.8p8-2.1│Tumbleweed│     2.3 MiB│x86_6         │                       
│     │ntp │     │4.2.8p8-2.1│Tumbleweed│     2.3 MiB│i586             │
└─────────────────────────────────────────┘ 
The manual zypper command works because it specifies the particular RPM file as if *** was not in a repository – which strictly it is not, it is just an RPM that happens to be in the same folder as the repository RPMs.

Yeah, I figured that out myself meanwhile. As mentioned, I “forgot” that the specified zypper command specified the actual URL to the package.

Anyway, a new snapshot has been released meanwhile, which includes the ntp “update” 4.2.8p8-3.1 (that actually is a downgrade to the previous version):
http://lists.opensuse.org/opensuse-factory/2016-09/msg00218.html

So updating the system should be safe now, and actually fix ntp if you didn’t manually downgrade it yet.

eng-int · September 10, 2016, 5:08am

It took me a long time to comment, and I did not see your post until it was too late to modify/delete mine.

So updating the system should be safe now, and actually fix ntp if you didn’t manually downgrade it yet.

Confirmed


 ┌────────────────────────────────────────────┐
 │     │Name│      │Version      │Repository    │Size          │Architecture   │
 │  i  │ntp     │      │4.2.8p8-1.3│@System     │     2.3 MiB│x86_64           │
 │     │ntp     │ x   │4.2.8p8-3.1│Tumbleweed│     2.3 MiB│x86_64           │
 │     │ntp     │      │4.2.8p8-3.1│Tumbleweed│     2.3 MiB│i586               │
 │                                                                                                                  │
 └────────────────────────────────────────────┘


y3:~ # l /usr/sbin/ntpd      
-rwxr-xr-x 1 root root 884424 Aug 13 18:26 /usr/sbin/ntpd*
y3:~ # zypper up ntp-4.2.8p8-3.1
y3:~ # l /usr/sbin/ntpd
-rwxr-xr-x 1 root root 884424 Sep  8 21:04 /usr/sbin/ntpd*

And I can confirm that it (build 3.1) works.