[h2] Re: java.lang.RuntimeException - 50000-175

2014-03-18 Thread Vishwanatha Sundarappa
Thank you very much for your suggestion If I remove MULTI_THREADED, my server start up takes in few hours for around 700M record of 150G. Combination of LOG, UNDO_LOG and MULTI_THREADED, takes just 20-30mins. I need this H2 instance read only purpose, any alternative suggestion to MULTI_THREADE

[h2] H2 1.3.174 in MVCC mode causes incorrect results, but only in memory mode

2014-03-18 Thread Wes Clark
I've been able to reliably reproduce a problem with MVCC mode (MVCC=TRUE on the URL) where the results come back in the wrong order, even when there is an ORDER BY in the query. The problem never appears when running in FILE mode, or in TCP mode. I understand the caveats about MVCC mode, but w

Re: [h2] NPE using FILE mode with 1.3.175

2014-03-18 Thread Wes Clark
With 1.3.174, this problem disappears, and I don't need any functionality in 1.3.175, so I consider this closed. On Friday, March 14, 2014 1:00:03 AM UTC-7, Thomas Mueller wrote: > > Hi, > > To find out what version of H2 you are using, could you post the complete > stack trace including error m

[h2] Generated keys not returned if sharing a connection across threads

2014-03-18 Thread u . g . dietrich
If sharing a connection across multiple threads, I sometimes get a "No data is available [2000-175]" if returning a generated key. The same code runs fine if I use one connection per thread. *Question:* Is it known and accepted behavior, that connections may not be used concurrently from separa

Re: [h2] Re: Unique index or primary key violation: "SYS_ID ON PUBLIC.SYS(ID)" [23505-168] (using 1.3.168)

2014-03-18 Thread Dieter Cailliau
URL: jdbc:h2:/tmp/bms;DB_CLOSE_DELAY=-1;LOCK_TIMEOUT=5000 According to your query, the create_build was 168. I'm not using temporary tables (afaik) just regular JPA (hibernate inside jboss7). I didn't see OutOfMemory in my logs, but they don't go back to the first occurence of the problem, so i

Fwd: Private message regarding: [h2] java.lang.RuntimeException - 50000-175

2014-03-18 Thread Thomas Mueller
Hi, I forward the message Vishwa sent to me in private. As documentd, the options LOG=0 and UNDO_LOG=0 can lead to corruption; specially when used together with MULTI_THREADED=1. Regards, Thomas -- Forwarded message -- From: Vishwanatha Sundarappa Date: Tue, Mar 18, 2014 at 6:

Re: [h2] State of DDL compatibility for MySQL

2014-03-18 Thread Noel Grandin
Thank you, patch is committed. On 2014-03-18 09:38, Peter Jentsch wrote: Hi Noel, ""I wrote the code, it's mine, and I'm contributing it to H2 for distribution multiple-licensed under the H2 License, version 1.0, and under the Eclipse Public License, version 1.0 (http://h2database.com/html/

Re: [h2] State of DDL compatibility for MySQL

2014-03-18 Thread peter . jentsch
Once again... ""I wrote the code, it's mine, and I'm contributing it to H2 for distribution multiple-licensed under the H2 License, version 1.0, and under the Eclipse Public License, version 1.0 (http://h2database.com/html/license.html)." ATB, Peter Am Montag, 17. März 2014 08:02:25 UTC+1 s

Re: [h2] State of DDL compatibility for MySQL

2014-03-18 Thread Peter Jentsch
Hi Noel, Am Montag, 17. März 2014 08:02:25 UTC+1 schrieb Noel Grandin: > > Wow, fast work! > > The patch seems fine, please can I get a license statement as per: > http://h2database.com/html/build.html#providing_patches > > ""I wrote the code, it's mine, and I'm contributing it to H2 for dist