[Samba] CDTB On Samba 4.1.12 As Member files server.

[Min Wai Chan]

Dear Steve,

Sorry for the incremental reply...
Try and error are what I'm doing

It seem that I've to leave the option EMPTY and not comment it... (ODD
though, as they say there no default value)

Ok I think both node are ok now...
I just have a big question on the LOCK...

So what happen now, we don't need the lock files anymore?

And split brain how they fix that?


Thank You.



On Thu, Oct 9, 2014 at 9:04 PM, Min Wai Chan <dcmwai at gmail.com> wrote:

> Dear Steve,
>
> Something like this...
> --> I've no idea where the /some/place/on/shared/storage come form...
>
> 2014/10/09 21:03:35.174892 [18102]: CTDB_WAIT_UNTIL_RECOVERED
> 2014/10/09 21:03:35.235889 [recoverd:18300]: server/ctdb_recoverd.c:1797
> Starting do_recovery
> 2014/10/09 21:03:35.235943 [recoverd:18300]: Taking out recovery lock from
> recovery daemon
> 2014/10/09 21:03:35.235963 [recoverd:18300]: Take the recovery lock
> 2014/10/09 21:03:35.235995 [recoverd:18300]: ctdb_recovery_lock: Unable to
> open /some/place/on/shared/storage - (No such file or directory)
> 2014/10/09 21:03:35.236012 [recoverd:18300]: Unable to get recovery lock -
> retrying recovery
> 2014/10/09 21:03:36.175680 [18102]: CTDB_WAIT_UNTIL_RECOVERED
> 2014/10/09 21:03:36.237854 [recoverd:18300]: server/ctdb_recoverd.c:1797
> Starting do_recovery
> 2014/10/09 21:03:36.237930 [recoverd:18300]: Taking out recovery lock from
> recovery daemon
> 2014/10/09 21:03:36.237950 [recoverd:18300]: Take the recovery lock
> 2014/10/09 21:03:36.238039 [recoverd:18300]: ctdb_recovery_lock: Unable to
> open /some/place/on/shared/storage - (No such file or directory)
> 2014/10/09 21:03:36.238058 [recoverd:18300]: Unable to get recovery lock -
> retrying recovery
>
>
> On Thu, Oct 9, 2014 at 8:57 PM, Min Wai Chan <dcmwai at gmail.com> wrote:
>
>> Dear Steve,
>>
>> If I disable the lock,
>>
>> Both node will not start and kept complaining on there is no lock files...
>>
>> Which is odd...
>>
>> On Thu, Oct 9, 2014 at 6:17 PM, steve <steve at steve-ss.com> wrote:
>>
>>> On 08/10/14 18:50, Min Wai Chan wrote:
>>>
>>>> Hi Amitay
>>>>
>>>> Somehow, I think the 2 node are conflicting each other.
>>>>
>>>> I'm not too sure what happen
>>>>
>>>> If I start node 1 only
>>>>
>>>> the ctdb status will be
>>>>
>>>> Number of nodes:2
>>>> pnn:0 192.168.20.30    OK (THIS NODE)
>>>> pnn:1 192.168.20.34    DISCONNECTED|DISABLED|UNHEALTHY|INACTIVE
>>>> Generation:514739632
>>>> Size:1
>>>> hash:0 lmaster:0
>>>> Recovery mode:NORMAL (0)
>>>> Recovery master:0
>>>>
>>>> And Later
>>>>
>>>> Number of nodes:2
>>>> pnn:0 192.168.20.30    UNHEALTHY (THIS NODE)
>>>> pnn:1 192.168.20.34    DISCONNECTED|DISABLED|UNHEALTHY|INACTIVE
>>>> Generation:514739632
>>>> Size:1
>>>> hash:0 lmaster:0
>>>> Recovery mode:NORMAL (0)
>>>> Recovery master:0
>>>>
>>>
>>> Neither node can get the lock. Disable the cluster lock and let drbd
>>> handle it instead.
>>>
>>> HTH,
>>> Steve
>>>
>>>
>>> --
>>> To unsubscribe from this list go to the following URL and read the
>>> instructions:  https://lists.samba.org/mailman/options/samba
>>>
>>
>>
>