[erlang-questions] Trouble starting mnesia on R14B02/3 node with a R14B01 node connected

[Brian Acton]

Hi,

We are similarly affected by this issue.

We have a cluster of nodes running stock R14B01

We upgraded one node from R14B01 to R14B03.

When we start mnesia, we get the same 'Failed to Merge Schema' error message
that Ferenc got when he tried.

Is there a solution for this? Does someone have a patch somewhere on github?

For us, we can not take the downtime to convert the whole cluster at one
time. We need to be able to restart one server at a time.

Any info, pointers, or help appreciated.

Thx,

--b


On Mon, May 30, 2011 at 6:17 AM, Ferenc Holzhauser <
ferenc.holzhauser@REDACTED> wrote:

>
> Yes, I did try R14B03 first then R14B02 to see if that works.
>
> BR,
> Ferenc
>
>
> On 30 May 2011 15:00, Dan Gudmundsson <dangud@REDACTED> wrote:
>
>> That is a bug in R14B03, I currently fail to understand how it can go
>> wrong on R14B02..
>>
>> Did you try R14B03 first?
>> There is a missing make rule, which doesn't trigger a remake if the .hrl
>> file is updated,
>> old clearcase problem which fixed those kind of build dependencies
>> automagicly.
>>
>> /Dan
>>
>> On Mon, May 30, 2011 at 11:59 AM, Ferenc Holzhauser <
>> ferenc.holzhauser@REDACTED> wrote:
>>
>>> Hi,
>>>
>>> After updating Erlang from R14B01 to R14B02 (or B03, tried both) on one
>>> node of a 2 node setup I had trouble starting mnesia.
>>>
>>> Eshell V5.8.3  (abort with
>>> ^G)
>>>
>>> (node2)1> mnesia:start().
>>>
>>>
>>> =ERROR REPORT==== 26-May-2011::13:54:06
>>> ===
>>> Mnesia('node2'): ** ERROR ** (core dumped to file:
>>> "/home/me/backend/MnesiaCore.node2_1306_410846_451973")
>>>  ** FATAL ** Failed to merge schema:
>>> {aborted,
>>>
>>> {function_clause,
>>>
>>> [{mnesia_schema,cs2list,
>>>
>>> [{cstruct,schema,set,[],
>>>                                            ['node2',
>>>                                             'node1'],
>>>
>>> [],0,read_write,[],[],false,schema,
>>>
>>> [table,cstruct],
>>>
>>> [],[],
>>>
>>> {{1300,896724,827143},
>>>                                             'node1'},
>>>
>>> {{4,0},
>>>                                             {'node2',
>>>
>>> {1302,780884,609311}}}}]},
>>>
>>> {mnesia_schema,do_merge_schema,1},
>>>
>>> {mnesia_tm,apply_fun,3},
>>>
>>> {mnesia_tm,execute_transaction,5},
>>>
>>> {mnesia_schema,schema_coordinator,
>>>
>>> 3}]}}
>>>
>>>
>>>
>>> At the end I stopped both nodes, updated Erlang then I could start both
>>> of them.
>>>
>>> Is it a bug or did I do something wrong when trying to update them one by
>>> one?
>>>
>>> Thanks in advance,
>>> Ferenc
>>>
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> erlang-questions mailing list
>>> erlang-questions@REDACTED
>>> http://erlang.org/mailman/listinfo/erlang-questions
>>>
>>>
>>
>
> _______________________________________________
> erlang-questions mailing list
> erlang-questions@REDACTED
> http://erlang.org/mailman/listinfo/erlang-questions
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://erlang.org/pipermail/erlang-questions/attachments/20110626/ca27c683/attachment.htm>