[Freeipa-devel] [PATCH 0041] Add logging to join command

Martin Kosek mkosek at redhat.com
Mon Mar 25 10:01:48 UTC 2013


On 03/25/2013 10:30 AM, Petr Viktorin wrote:
> On 03/22/2013 05:40 PM, Tomas Babej wrote:
>> On 03/22/2013 05:10 PM, Tomas Babej wrote:
>>> On 03/22/2013 04:51 PM, Petr Viktorin wrote:
>>>> On 03/13/2013 03:05 PM, Tomas Babej wrote:
>>>>> Hi,
>>>>>
>>>>> The following is mentioned in the server log now:
>>>>>    - existence of host entry (if it already does exist)
>>>>>    - missing krbprincipalname and its new value (if there was no
>>>>>      principal name set)
>>>>>
>>>>> https://fedorahosted.org/freeipa/ticket/3481
>>>>>
>>>>> Tomas
>>>>>
>>>>
>>>> Here is what I get first trying to re-enroll a wiped host first using
>>>> admin username/password, and then using a saved keytab.
>>>> The first succeeded, the second didn't.
>>>>
>>> We discussed this with Petr, this is a typo and actually happend in
>>> reversed (and correct) order :)
>>>>
>>>> [Fri Mar 22 16:17:14.338411 2013] [:error] [pid 21335] ipa: INFO:
>>>> Host entry for vm-084.idm.lab.eng.brq.redhat.com already exists
>>>> [Fri Mar 22 16:17:14.367564 2013] [:error] [pid 21335] ipa: INFO:
>>>> admin at IDM.LAB.ENG.BRQ.REDHAT.COM:
>>>> join(u'vm-084.idm.lab.eng.brq.redhat.com',
>>>> nshardwareplatform=u'x86_64', nsosversion=u'3.7.4-204.fc18.x86_64',
>>>> version=u'2.51'): SUCCESS
>>>> [Fri Mar 22 16:17:35.395626 2013] [:error] [pid 21336] ipa: INFO:
>>>> Host entry for vm-084.idm.lab.eng.brq.redhat.com already exists
>>>> [Fri Mar 22 16:17:35.420830 2013] [:error] [pid 21336] ipa: INFO:
>>>> host/vm-084.idm.lab.eng.brq.redhat.com at IDM.LAB.ENG.BRQ.REDHAT.COM:
>>>> join(u'vm-084.idm.lab.eng.brq.redhat.com',
>>>> nshardwareplatform=u'x86_64', nsosversion=u'3.7.4-204.fc18.x86_64',
>>>> version=u'2.51'): SUCCESS
>>>> [Fri Mar 22 16:17:38.729304 2013] [:error] [pid 21335] ipa: INFO:
>>>> host/vm-084.idm.lab.eng.brq.redhat.com at IDM.LAB.ENG.BRQ.REDHAT.COM:
>>>> host_mod(u'vm-084.idm.lab.eng.brq.redhat.com', random=False,
>>>> ipasshpubkey=([...], [...]), rights=False, updatedns=False,
>>>> all=False, raw=False, version=u'2.54'): SUCCESS
>>>> [Fri Mar 22 16:17:41.763080 2013] [:error] [pid 21336] ipa: INFO:
>>>> host/vm-084.idm.lab.eng.brq.redhat.com at IDM.LAB.ENG.BRQ.REDHAT.COM:
>>>> cert_request([...],
>>>> principal=u'host/vm-084.idm.lab.eng.brq.redhat.com at IDM.LAB.ENG.BRQ.REDHAT.COM',
>>>>
>>>> add=True, version=u'2.51'): SUCCESS
>>>>
>>>> I can see what's going on, but I don't think the admin would be much
>>>> wiser seeing this.
>>>> Would we want so say something like "Host entry for X already exists;
>>>> joining may fail on the client side" to explain what's going on?
>>>>
>>> I agree with the proposal. Sending updated patch, here's the diff:
>>>
>>> diff --git a/ipaserver/plugins/join.py b/ipaserver/plugins/join.py
>>> index 86d6674..3b66805 100644
>>> --- a/ipaserver/plugins/join.py
>>> +++ b/ipaserver/plugins/join.py
>>> @@ -101,7 +101,9 @@ class join(Command):
>>>              dn = attrs_list['dn']
>>>
>>>              # No error raised so far means that host entry exists
>>> -            self.log.info('Host entry for %s already exists', hostname)
>>> +            self.log.info('Host entry for %s already exists, '
>>> +                          'joining may fail on the client side '
>>> +                          'if not forced', hostname)
>>>
>>>              # If no principal name is set yet we need to try to add
>>>              # one.
>>>
>>>
>>>> ACK from a technical perspective.
>>>>
>>>
>>> Attached.
>>>
>>> Tomas
>>
>> I sent the wrong patch (the previous one) last time. Sorry for the
>> confusion.
>>
>> Tomas
> 
> ACK
> 

Pushed to master, ipa-3-1.

Martin




More information about the Freeipa-devel mailing list