PDA

View Full Version : NTP Server issue while setting up SUSE Openstack Cloud 6



snivas
19-Apr-2016, 12:35
Hi All,

As a part of study, i am setting up SUSE Openstack Cloud 6 on Virtual Box. The SUSE administration server installation went fine. Now i need to add a node to it.

When a new node is added, it is properly getting PXE booted but failing at NTP synchronization stage. Due to that, the newly booted node is not getting registered with crowbar and not shown in the UI.

Log at SLES Admin node : /var/log/crowbar/sledgehammer/d08-00-27-eb-2e-85.localcloud.transcript

DHCP_STATE=discovery^M
19 Apr 10:55:15 ntpdate[719]: no server suitable for synchronization found
Waiting for NTP server
19 Apr 10:55:22 ntpdate[721]: no server suitable for synchronization found
Waiting for NTP server
19 Apr 10:55:29 ntpdate[723]: no server suitable for synchronization found
Waiting for NTP server
19 Apr 10:55:36 ntpdate[726]: no server suitable for synchronization found
Waiting for NTP server
19 Apr 10:55:43 ntpdate[728]: no server suitable for synchronization found

TCP Dump at admin server

16:51:43.768103 IP (tos 0x0, ttl 64, id 27812, offset 0, flags [DF], proto UDP (17), length 76)
192.168.124.28.ntp > sles-admin6.localcloud.ntp: [udp sum ok] NTPv4, length 48
Client, Leap indicator: clock unsynchronized (192), Stratum 0 (unspecified), poll 3 (8s), precision -6
Root Delay: 1.000000, Root dispersion: 1.000000, Reference-ID: (unspec)
Reference Timestamp: 0.000000000
Originator Timestamp: 0.000000000
Receive Timestamp: 0.000000000
Transmit Timestamp: 3670053723.146002963 (2016/04/19 16:52:03)
Originator - Receive Timestamp: 0.000000000
Originator - Transmit Timestamp: 3670053723.146002963 (2016/04/19 16:52:03)
16:51:43.768312 IP (tos 0xb8, ttl 64, id 4838, offset 0, flags [DF], proto UDP (17), length 76)
sles-admin6.localcloud.ntp > 192.168.124.28.ntp: [udp sum ok] NTPv4, length 48
Server, Leap indicator: clock unsynchronized (192), Stratum 0 (unspecified), poll 3 (8s), precision -22
Root Delay: 0.000000, Root dispersion: 0.067794, Reference-ID: (unspec)
Reference Timestamp: 0.000000000
Originator Timestamp: 3670053723.146002963 (2016/04/19 16:52:03)
Receive Timestamp: 3670053703.768103957 (2016/04/19 16:51:43)
Transmit Timestamp: 3670053703.768283903 (2016/04/19 16:51:43)
Originator - Receive Timestamp: -19.377898991
Originator - Transmit Timestamp: -19.377719044

May i know how to overcome this issue? Since the networks configurations are handled by crowbar, the SLES Admin server cannot access internet.

Thanks in advance!

ab
19-Apr-2016, 12:56
Wildly guessing:

Despite not being able to get time information via NTP from an outside
source, does your admin server's NTP service think it has time
synchronized properly? If not, for example if it does not see a source
itself, or if it thinks it has not been up long enough to be stable, I do
not know how it will respond to a client.

--
Good luck.

If you find this post helpful and are logged into the web interface,
show your appreciation and click on the star below...

smflood
19-Apr-2016, 13:52
On 19/04/16 12:44, snivas wrote:

> As a part of study, i am setting up SUSE Openstack Cloud 6 on Virtual
> Box. The SUSE administration server installation went fine. Now i need
> to add a node to it.
>
> When a new node is added, it is properly getting PXE booted but failing
> at NTP synchronization stage. Due to that, the newly booted node is not
> getting registered with crowbar and not shown in the UI.
>
> Log at SLES Admin node :
> /var/log/crowbar/sledgehammer/d08-00-27-eb-2e-85.localcloud.transcript
>
> DHCP_STATE=discovery^M
> 19 Apr 10:55:15 ntpdate[719]: no server suitable for synchronization
> found
> Waiting for NTP server
> 19 Apr 10:55:22 ntpdate[721]: no server suitable for synchronization
> found
> Waiting for NTP server
> 19 Apr 10:55:29 ntpdate[723]: no server suitable for synchronization
> found
> Waiting for NTP server
> 19 Apr 10:55:36 ntpdate[726]: no server suitable for synchronization
> found
> Waiting for NTP server
> 19 Apr 10:55:43 ntpdate[728]: no server suitable for synchronization
> found
>
> TCP DUMP AT ADMIN SERVER
>
> 16:51:43.768103 IP (tos 0x0, ttl 64, id 27812, offset 0, flags [DF],
> proto UDP (17), length 76)
> 192.168.124.28.ntp > sles-admin6.localcloud.ntp: [udp sum ok] NTPv4,
> length 48
> Client, Leap indicator: clock unsynchronized (192), Stratum 0
> (unspecified), poll 3 (8s), precision -6
> Root Delay: 1.000000, Root dispersion: 1.000000, Reference-ID:
> (unspec)
> Reference Timestamp: 0.000000000
> Originator Timestamp: 0.000000000
> Receive Timestamp: 0.000000000
> Transmit Timestamp: 3670053723.146002963 (2016/04/19
> 16:52:03)
> Originator - Receive Timestamp: 0.000000000
> Originator - Transmit Timestamp: 3670053723.146002963
> (2016/04/19 16:52:03)
> 16:51:43.768312 IP (tos 0xb8, ttl 64, id 4838, offset 0, flags [DF],
> proto UDP (17), length 76)
> sles-admin6.localcloud.ntp > 192.168.124.28.ntp: [udp sum ok] NTPv4,
> length 48
> Server, Leap indicator: clock unsynchronized (192), Stratum 0
> (unspecified), poll 3 (8s), precision -22
> Root Delay: 0.000000, Root dispersion: 0.067794, Reference-ID:
> (unspec)
> Reference Timestamp: 0.000000000
> Originator Timestamp: 3670053723.146002963 (2016/04/19
> 16:52:03)
> Receive Timestamp: 3670053703.768103957 (2016/04/19
> 16:51:43)
> Transmit Timestamp: 3670053703.768283903 (2016/04/19
> 16:51:43)
> Originator - Receive Timestamp: -19.377898991
> Originator - Transmit Timestamp: -19.377719044
>
> May i know how to overcome this issue? Since the networks configurations
> are handled by crowbar, the SLES Admin server cannot access internet.

I've recently worked through exactly the same setup and situation you
reference here. Whilst the other nodes would boot they wouldn't then
register with the Admin node. Checking the .transcript file on a failed
node showed the same time errors you report. Running ntpdate by hand
gave more information indicating strata was 16 and thus not suitable.

I found that although my VirtualBox host synchronised time my Admin node
VM did not. In the end, given I am running VirtualBox on openSUSE Leap,
I edited my ntp.conf file to allow the virtual Cloud network I'd created
to receive time.

HTH.
--
Simon
SUSE Knowledge Partner

------------------------------------------------------------------------
If you find this post helpful and are logged into the web interface,
please show your appreciation and click on the star below. Thanks.
------------------------------------------------------------------------

snivas
19-Apr-2016, 16:39
Thanks AB, smflood.

@smflood :

In the end, given I am running VirtualBox on openSUSE Leap,
I edited my ntp.conf file to allow the virtual Cloud network I'd created
to receive time.

could you please explain it in detail? ntp.conf of host machine? I have HostOnly network where all the VMs were connected. My virtualbox is running on Ubuntu 14.04

smflood
19-Apr-2016, 17:40
On 19/04/16 16:44, snivas wrote:

> Thanks AB, smflood.
>
> @smflood :
>
> - In the end, given I am running VirtualBox on openSUSE Leap,
> I edited my ntp.conf file to allow the virtual Cloud network I'd
> created
> to receive time.-
>
> could you please explain it in detail? My virtualbox is running on
> Ubuntu 14.04

So in VirtualBox I created a new host-only network so my SUSE OpenStack
Cloud set up is completely private. I set the the IP address to
192.168.124.1 to stick with the documented defaults for SUSE OpenStack
Cloud.

Given that new nodes will provision and run from 192.168.124.0/24 I
edited my /etc/ntp.conf file to add "restrict 192.168.124.0 mask
255.255.255.0" so that they could all synchronise time with my
VirtualBox host.

That reminds me, I also edited the NTP Barclamp (within the Admin
Dashboard) and entered 192.168.124.1 as the External server. I believe
this latter step makes sure the nodes know of a suitable server and
provided this isn't a high strata should suffice.

HTH.
--
Simon
SUSE Knowledge Partner

------------------------------------------------------------------------
If you find this post helpful and are logged into the web interface,
please show your appreciation and click on the star below. Thanks.
------------------------------------------------------------------------

snivas
20-Apr-2016, 06:13
Simon,

Thanks a ton! It fixed my issue.

eblock
21-Apr-2016, 13:52
Hi,

although your issue has been resolved, I would like to leave a comment. I tested both cloud 5 and 6, and since cloud 6 I had the same problem (no suitable ntp server). In my case it was sufficient to edit the NTP barclamp and add an external NTP server, I didn't have to edit any config files myself. I don't know if the ntp server can be the admin server itself, I have an external server available, so I didn't need to test it.

Regards,
Eugen

snivas
22-Apr-2016, 07:35
Thanks Eugen. Good to know

smflood
22-Apr-2016, 11:20
On 21/04/16 13:54, eblock wrote:

> although your issue has been resolved, I would like to leave a comment.
> I tested both cloud 5 and 6, and since cloud 6 I had the same problem
> (no suitable ntp server). In my case it was sufficient to edit the NTP
> barclamp and add an external NTP server, I didn't have to edit any
> config files myself. I don't know if the ntp server can be the admin
> server itself, I have an external server available, so I didn't need to
> test it.

Thanks for your comments. Note that in my case it's not the Admin node
that's acting as the NTP server but my VirtualBox host machine.

HTH.
--
Simon
SUSE Knowledge Partner

------------------------------------------------------------------------
If you find this post helpful and are logged into the web interface,
please show your appreciation and click on the star below. Thanks.
------------------------------------------------------------------------