incubator-couchdb-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chris Ashton <chrisjash...@yahoo.co.uk>
Subject Re: Corrupt database file
Date Fri, 21 Mar 2014 14:04:59 GMT
Yikes, guess that shows just what a beginner I am with it!

Thanks for the tip

Chris




________________________________
 From: Garren Smith <garren@apache.org>
To: user@couchdb.apache.org; Chris Ashton <chrisjashton@yahoo.co.uk> 
Cc: Tim Tisdall <tisdall@gmail.com> 
Sent: Friday, 21 March 2014, 13:53
Subject: Re: Corrupt database file
 

Hi Chris, 

If you have Couchbase, that is different to Couchdb. Its better to ask on the Couchbase mailing
list. They might be able to help you there.

Cheers
Garren


On 21 Mar 2014, at 3:34 PM, Chris Ashton <chrisjashton@yahoo.co.uk> wrote:

> Hi Tim
> 
> Many thanks for this reply. I have tried this dump method but even that fails, complaining
about the database file : "curl: (56) Problem (2) in the Chunked-Encoded data"
> 
> Our files are huge so everything takes ages.
> 
> The only other thought I have right now is to process this enormous text file, add whatever
curly braces and the like are required to turn it back into valid JSON and then, as you suggest,
rewrite it into the database in some way.
> 
> I'm just surprised there is no utility to fix bad data sections, it's a real pity.
> 
> 
> By the way, I probably should have mentioned, we appear to have couchbase single version
1.2.0
> 
> Thanks again
> 
> Chris
> 
> 
> 
> ________________________________
> From: Tim Tisdall <tisdall@gmail.com>
> To: user@couchdb.apache.org; Chris Ashton <chrisjashton@yahoo.co.uk> 
> Sent: Friday, 21 March 2014, 13:19
> Subject: Re: Corrupt database file
> 
> 
> 
> I don't think there's any tool for fixing corrupted db files...  What I'd try doing
is dumping all the content from the DB and reconstructing it.  You can fetch everything in
the database by downloading http://127.0.0.1:5984/my_database/_all_docs?include_docs=true
(where the IP and domain is your server, and 'my_database' is your DB).  You'd then need
to write some sort of script to read that JSON document and then write the values back into
a _new_ database.
> 
> Does anyone know if there's a way to do this same sort of thing with replication?  (I
have to do it the other way because I don't have enough space for 2 copies of my DB on my
system)
> 
> 
> 
> 
> On 21 March 2014 07:46, Chris Ashton <chrisjashton@yahoo.co.uk> wrote:
> 
> Hi
>> 
>> I am a novice with CouchDB and really struggling to support a product that someone
developed before leaving the company. No one else has any skills in Couch so I'm trying to
pick up the pieces.
>> 
>> Our database has stopped responding and we've tried compacting but to no avail.
>> 
>> We are getting errors like the following in the log file, which I presume mean that
we have a corrupt db file for one reason or another:
>> 
>> 
>> [Fri, 21 Mar 2014 10:06:56 GMT] [error] [<0.213.0>] ** Generic server <0.213.0>
terminating
>> ** Last message in was {'EXIT',<0.217.0>,
>>                                {file_corruption,<<"file corruption">>}}
>> 
>> I was wondering if there was a utility which would scan DB files and remove badly
formed parts or anything like that? We are running on Windows.
>> 
>> Many thanks
>> 
>> Chris
Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message