Hello all,

I am Bill Torcaso and I have been lurking here for a few months.  And right 
up front, I want to say that H2 is well-supported and good for small to 
medium data set sizes.

We use H2 to hold very large amounts of data - 20GB up to 80GB.  And I have 
seen a problem, multiple times, but not investigated enough to get a 
repeatable case.  

I inherited a legacy Java application with H2 embedded in it.  In its 
prime, the largest DB was much, much smaller.  At a guess, DB's were 
perhaps up to 1GB.

The DB has about 5 tables, and the schema is very straightforward.  It is 
in effect read-only once created; users will only query, not update.  I 
receive the DB with minimal indexes, and as a result, many queries do a 
full table-scan.  I decided to add LOTS of indexes.

I won't take you through all the steps.  But I found it unreliable, with 
the result that the DB could lose ALL of its indexes without reporting an 
error.

I suggest that you do some stress tests with an oversized DB for your 
application.

(And no, I can't be more specific at this time;  creating 20 indexes on 44 
million rows takes a long time to execute, let alone troubleshoot.  But I 
send my thanks to all the H2 developers for their work ....)

On Thursday, July 18, 2013 4:00:45 AM UTC-4, JList wrote:
>
> Hi all,
>
> I've been using H2 in production on a small server without any problems. 
> I'm thinking about using it again in another application. I did a search in 
> the mailing list for database corruption. It looks there there have been 
> some reports of data corruptions, some of which were followed by fixes. I 
> guess some H2 versions might be more reliable then others because they have 
> been better tested and with relatively speaking lower risk changes. I 
> wonder if there are any versions that are considered more reliable and more 
> suited for production use?
>
> Thanks,
> Jack
>

-- 
You received this message because you are subscribed to the Google Groups "H2 
Database" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/h2-database.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to