Re: ntp check fails during 11.2.0.4 Install

From: Radoulov, Dimitre <cichomitiko_at_gmail.com>
Date: Wed, 17 Sep 2014 12:05:33 +0200
Message-ID: <54195CED.8030607_at_gmail.com>



You need to run:

./runcluvfy.sh stage -pre crsinst -n node1,node2 -fixup -verbose

Not clocksync!

Compare the file /etc/ntp.conf on the nodes where ntpq works and the nodes where it's not working.
Check if the *restrict* directives are different.

Regards
Dimitre

On 17/09/2014 11:50, Kumar Madduri wrote:
> Hello
> Here is what I get from strace with local host and with 127.0.0.1
> When I try to run runcluvfy I get errors
>
> oracle_at_<removed>:> ./runcluvfy.sh comp clocksync -n <removed>
> [...]
>
> PRVF-9652 : Cluster Time Synchronization Services check failed
>
> Verification of Clock Synchronization across the cluster nodes was
> unsuccessful on all the specified nodes.
>
> Attached is the strace output that i get for localhost and 127.0.0.1
>
> Thank you
> Kumar
>
>
> On Wed, Sep 17, 2014 at 1:13 AM, Radoulov, Dimitre
> <cichomitiko_at_gmail.com <mailto:cichomitiko_at_gmail.com>> wrote:
>
> Yes,
> the configuration seems fine. Could try to strace it and
> analyze/post the output?
>
> strace -f ntpq -p localhost
>
> What happens when you run clufvy manually? What's the exact PRVF
> message?
>
>
> Regards
> Dimitre
>
>
> On 17/09/2014 10:05, Kumar Madduri wrote:
>> Hello
>> Thank you for the reply
>> I checked nslookup localhost and it shows 127.0.0.1
>> This is how I have the entry in my /etc/hosts file
>> 127.0.0.1 localhost localhost.localdomain localhost4
>> localhost4.localdomain4
>> ::1 localhost localhost.localdomain localhost6
>> localhost6.localdomain6
>> There are other older boxes where I have same entries for local
>> hosts in my hosts fiile and I don't have issue there.
>> Only in this new box which is on RHEL 6.5 I am having this issue.
>> Many thanks again
>> Kumar
>>
>> On Wed, Sep 17, 2014 at 12:22 AM, Radoulov, Dimitre
>> <cichomitiko_at_gmail.com <mailto:cichomitiko_at_gmail.com>> wrote:
>>
>> Check if nslookup localhost reports something different than
>> 127.0.0.1.
>> You said that the /etc/hosts content is correct, did you
>> check the localhost entry?
>> It shouldn't be associated with the hostname.
>>
>> Dimitre
>>
>> On 17/09/2014 07:53, Kumar Madduri wrote:
>>> I found this
>>> ntpq -p localhost times out but ntpq -p 127.0.0.1 works.
>>> Our /etc/hosts has right values for localhost entries too.
>>> May not be a firewall issue afterall.
>>>
>>>
>>>
>>> On Tue, Sep 16, 2014 at 8:35 PM, Kumar Madduri
>>> <ksmadduri_at_gmail.com <mailto:ksmadduri_at_gmail.com>> wrote:
>>>
>>> Thanks David for the hint.
>>> ntpq -p timesout on my new servers but works on my
>>> existing servers. May be it is some firewall rule.
>>>
>>> I will pursue with our sysadmins.
>>>
>>> _ Kumar
>>>
>>> On Tue, Sep 16, 2014 at 8:19 PM, David Barbour
>>> <david.barbour1_at_gmail.com
>>> <mailto:david.barbour1_at_gmail.com>> wrote:
>>>
>>> Check the offset. It needs to be less than 1000ms.
>>> Use ntpq -p (I think off the top of my head - you
>>> probably want to check me on this). If your offset
>>> is > 1000, change the ntp server until you find one
>>> that's consistent.
>>>
>>> On Tue, Sep 16, 2014 at 9:56 PM, Kumar Madduri
>>> <ksmadduri_at_gmail.com <mailto:ksmadduri_at_gmail.com>>
>>> wrote:
>>>
>>> Hello
>>> I am trying to install 11.2.0.4 GI (OS is Linux
>>> 6.5) and during the final checks, the installer
>>> complains of ntp check failing on both nodes (2
>>> node rac installation).
>>>
>>> My ntpd file on both nodes is
>>> cat /etc/sysconfig/ntpd
>>> # Drop root to id 'ntp:ntp' by default.
>>> OPTIONS="-x -u ntp:ntp -p /var/run/ntpd.pid -g"
>>>
>>> The ntp daemon is started on both nodes.
>>>
>>> OUI just reports like this ' This task
>>> verifies the cluster time synhronization on
>>> clusters that use Network Time Protocol (NTP)'
>>> Check failed on Nodes 'node2, node1'
>>>
>>> If I stop ntp and then ctssd check is fine and
>>> the installer would let me install. But I have
>>> read that there are several issues with ctssd
>>> and I also want to figure out what is wrong with
>>> the ntp service that it gives this error.
>>>
>>> Any suggestions?
>>>
>>>
>>> Thank you
>>> Kumar
>>>
>>>
>>>
>>>
>>
>>
>
>

--
http://www.freelists.org/webpage/oracle-l
Received on Wed Sep 17 2014 - 12:05:33 CEST

Original text of this message