Mailing-List: contact user-help@couchdb.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@couchdb.apache.org
Received-SPF: pass (athena.apache.org: domain of paul.joseph.davis@gmail.com
 designates 209.85.217.170 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type:content-transfer-encoding;
        b=I17fgCKakkK/S7IGDg4k3tRxbedJAMAHJdDoVBCXarxUekWaHVcYaCaeYr1bNXrjnn
         wVi5046swOa7ViG7J5gh80LtVBTCTtwNI4HuNjfGA6iESEiM41SiQRw6wu74COB0lC4k
         7RuBxRBGaAxL+kB2KULO7CccNYxcDUPy6Di6M=
MIME-Version: 1.0
In-Reply-To: <7DC8F4B4-854D-4170-BB69-B7F27B00D8EA@abaca.com>
References: <D68C6A2C-2721-4B82-8703-653BA1CFF8FD@abaca.com>
	 <70CE5E98-5863-482E-82FD-3C38CECD70A5@apache.org>
	 <25E7BE99-AF26-4260-B434-D1AC747ADEA1@abaca.com>
	 <e2111bbb0905121552h5b1bf4b6ve29871cea1ef930f@mail.gmail.com>
	 <7DC8F4B4-854D-4170-BB69-B7F27B00D8EA@abaca.com>
Date: Tue, 12 May 2009 19:54:15 -0400
Message-ID: <e2111bbb0905121654m4233f94tbb171cf813896583@mail.gmail.com>
Subject: Re: corrupted database file
From: Paul Davis <paul.joseph.davis@gmail.com>
To: user@couchdb.apache.org
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

On Tue, May 12, 2009 at 7:45 PM, Peter Hsu <phsu@abaca.com> wrote:
> May I ask where the feature lives in the timeline?
>

I don't have anything more specific than the next couple of months
probably. It really depends on when the first person decides to tackle
the patch.

HTH,
Paul Davis

> On May 12, 2009, at 3:52 PM, Paul Davis wrote:
>
>> On Tue, May 12, 2009 at 6:17 PM, Peter Hsu <phsu@abaca.com> wrote:
>>>
>>> What does "tail-append" headers mean?
>>>
>>> Is there anything that can be done in the current version to avoid this=
?
>>> (aside from ensuring there is always disk space)
>>>
>>
>> Its a planned feature so that instead of overwriting the header at the
>> beginning of the file we just append a new header to the end of the
>> file. In the future when you have a weird file truncation we can just
>> walk backwards through the db to get the last valid version assuming
>> there's no corruption in the middle of the file.
>>
>> HTH,
>> Paul Davis
>>
>>> Thanks,
>>> Peter
>>>
>>> On May 12, 2009, at 12:23 PM, Damien Katz wrote:
>>>
>>>> It looks like the database file got truncated. Until we get tail-
>>>> append headers, there is no way to recover the file contents when this
>>>> happens.
>>>>
>>>> -Damien
>>>>
>>>>
>>>> On May 12, 2009, at 1:25 PM, Peter Hsu wrote:
>>>>
>>>>> I get the following error when visiting futon:
>>>>>
>>>>> Database information could not be retrieved:
>>>>> {{{badmatch,eof},
>>>>> [{couch_file,handle_call,3},
>>>>> =A0{gen_server,handle_msg,5},
>>>>> =A0{proc_lib,init_p_do_apply,3}]},
>>>>> {gen_server,call,[<0.82.0>,
>>>>> {pread_bin,86816267188},infinity]}}
>>>>>
>>>>> This is in couch 0.9.0.
>>>>>
>>>>> The file system had run out of disk space previously. =A0There is
>>>>> sufficient disk space now, but the database file seems to be
>>>>> corrupted. =A0Is there any utility to try to repair the database?
>>>>>
>>>
>