Re: [h2] help debugging h2 issue (big join is too slow)

2015-03-28 Thread Noel Grandin
Also have a look at the performance tuning section of our documentation. Would be interesting to have a stack profile of that query. On Friday, 27 March 2015, Noel Grandin noelgran...@gmail.com wrote: What does explain plan say about the query? On Friday, 27 March 2015, Rob Audenaerde

Re: [h2] help debugging h2 issue (big join is too slow)

2015-03-27 Thread Noel Grandin
What does explain plan say about the query? On Friday, 27 March 2015, Rob Audenaerde rob.audenae...@gmail.com wrote: Hi all, I have a big join with some regexp_replace which seems to take 100% cpu without printing any debug info. It takes over 30 minutes to run (then I killed the process)

Re: [h2] H2 seems to create a huge number of orphan lobs since 1.4.183.

2015-03-26 Thread Noel Grandin
On 2015-03-26 08:48 AM, Thomas Mueller wrote: I didn't know that. I made some tests (see below), and at least Apache Derby seems to do that. I'm not sure about other In Oracle, you may only read a BLOB column in the same order as the SELECT statement, and the BLOB ceases to be valid as

Re: [h2] H2 seems to create a huge number of orphan lobs since 1.4.183.

2015-03-25 Thread Noel Grandin
On 2015-03-23 10:43 AM, Thomas Mueller wrote: This is a problem, and I don't have a good solution for that I'm afraid. Closing the database or closing the connection from time to time are not good solutions I think. I can think of a few options (1) add a manual command to clean them up

Re: [h2] cache_size question

2015-03-21 Thread Noel Grandin
I forgot to ask earlier - what does your database URL look like? -- 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 h2-database+unsubscr...@googlegroups.com. To post

Re: [h2] [H2 1.3.169] Problem while recovering a corrupted database.

2015-03-20 Thread Noel Grandin
Sorry but your database is corrupted. The recover tool is not perfect (there is no way it could be) so you will have to edit your recovered script by hand to fix up any remaining errors. -- You received this message because you are subscribed to the Google Groups H2 Database group. To

Re: [h2] Data loss with H2 1.4.186

2015-03-20 Thread Noel Grandin
The only current stable configuration is PageStore without MVCC. If you need more performance you will either need to - do performance tuning of your queries - upgrade your hardware - look at other database options (like PostgreSQL) -- You received this message because you are subscribed to the

Re: [h2] Re: cache_size question

2015-03-20 Thread Noel Grandin
what version are you using, because it works fine for me with the latest version. On 2015-03-20 03:10 PM, sim wrote: No, I am not. I just connect to db from DBeaver client and run set cache_size 32768 command. Then I open information_schema.settings table and show the result. On Friday,

Re: [h2] Performance question - multi column query

2015-03-20 Thread Noel Grandin
It's a bit of a hack, but what I can suggest is this: do your query like this: SELECT * FROM xvalues WHERE time = x and time (x + 50) and timeend = (x + 1 + 50) Where 50 is a value calculated to give you the complete range 99.999% of the time. When you detect that you might possibly not

Re: [h2] Error when adding milliseconds to timestamps

2015-03-18 Thread Noel Grandin
Thanks for reporting this, fixed in trunk. -- 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 h2-database+unsubscr...@googlegroups.com. To post to this group, send

Re: [h2] Has the name of the inline sql function that converts a string to a timestamp been changed since 1.4.180? And if so, why haven't you publicized the fact?

2015-03-18 Thread Noel Grandin
We had to change part of the client-server protocol, see the changelog entry Version 1.4.182 Beta (2014-10-17) The change in JDBC escape processing in version 1.4.181 affects both the parser (which is running on the server) and the JDBC API (which is running on the client). If you (or

Re: [h2] Data loss with H2 1.4.186

2015-03-18 Thread Noel Grandin
On Wed, Mar 18, 2015 at 8:21 PM, Daniele Renda daniele.re...@gmail.com wrote: Just as info, also disabling MVCC and redoing tests I wrote I can break foreign key! If you can create a standalone test case I can look at fixing this, at the moment I have no idea what could be wrong. -- You

Re: [h2] Data loss with H2 1.4.186

2015-03-18 Thread Noel Grandin
On Wed, Mar 18, 2015 at 8:16 PM, Daniele Renda daniele.re...@gmail.com wrote: In short, if I need row level lock and a good reliability what version of H2 I should use and with which settings? Are you really sure you need MVCC mode, because H2 is plenty fast and concurrent in it's normal

Re: [h2] Data loss with H2 1.4.186

2015-03-18 Thread Noel Grandin
This is probably an issue with mvcc mode. Unfortunately we are unlikely to fix it because mvcc mode is going to be replaced by the MVStore engine. For now I would suggest that you run without mvcc mode. On Wednesday, 18 March 2015, Daniele Renda daniele.re...@gmail.com wrote: The db url was

Re: [h2] PoolableConnectionFactory not supported

2015-03-17 Thread Noel Grandin
this combination of features is not supported: autoServerMode (readOnly || fileLockMethod == NO || fileLockMethod == SERIALIZED || inMemory) -- You received this message because you are subscribed to the Google Groups H2 Database group. To unsubscribe from this group and stop receiving

Re: [h2] db deadlocked until ui takes focus

2015-03-17 Thread Noel Grandin
That sounds like some kind of weird Microsoft Windows message queue issue. You aren't perhaps running any interesting 3rd party JNI libraries are you? Or some kind of virus checker? -- You received this message because you are subscribed to the Google Groups H2 Database group. To unsubscribe

Re: [h2] PoolableConnectionFactory not supported

2015-03-17 Thread Noel Grandin
Hi OK, that error message is inaccurate. The actual condition appears to be: autoServerMode (readOnly || fileLockMethod == NO || fileLockMethod == SERIALIZED || fileLockMethod == FS || inMemory) But I can't see why you would be ending up file locking set to something other than FILE (the

Re: [h2] Persistent to In-memory Database

2015-03-13 Thread Noel Grandin
You trimmed off part of the exception. We need that part :-) -- 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 h2-database+unsubscr...@googlegroups.com. To post to

Re: [h2] Transaction performance

2015-03-13 Thread Noel Grandin
If you could create us a standalone test-case we could probably fix it. -- 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 h2-database+unsubscr...@googlegroups.com.

Re: [h2] JDBC driver for a Windows based client

2015-03-12 Thread Noel Grandin
H2 supports the PostgreSQL ODBC driver, so install that and use it to connect. http://www.h2database.com/html/advanced.html#odbc_driver On 2015-03-11 06:36 PM, Faraz Hasan Rizvi wrote: Hi, I am trying to connect a vendor application to H2 via JDBC. The application has both Linux and Windows

Re: [h2] db deadlocked until ui takes focus

2015-03-12 Thread Noel Grandin
On 2015-03-11 10:45 PM, Brian Craft wrote: I'm at a loss to explain this behavior. Wondering if anyone else has an idea. The app has a swing ui, and h2 with c3po. On occasion the h2 jobs start hanging, and remain stuck until the ui window takes focus. After it takes focus, things are fine

Re: [h2] Query index selection with 'sub types'

2015-03-12 Thread Noel Grandin
On 2015-03-12 01:50 AM, Mike Goodwin wrote: So aside from improving the optimizer what would fix my problem would be the ability to add hints to fix the join ordering. I know this is on the road map and I wouldn't mind doing it and do not think it should be too hard to implement (it amounts

Re: [h2] Re: Insert with multi column constraint issue

2015-03-05 Thread Noel Grandin
On Thu, Mar 5, 2015 at 5:57 PM, Mike Goodwin mike.good...@cantab.net wrote: Ok, I can see that the constraint is specified incorrectly (in the wrong order). It really had to be something like that. Still, should it be possible to create a constraint with mismatched columns like that? You

Re: [h2] Deadlock caused by h2 disk access

2015-03-04 Thread Noel Grandin
, On Wednesday, March 4, 2015 at 3:45:25 PM UTC+7, Noel Grandin wrote: That's not a deadlock, that's just a slow SQL command that is causing your other SQL commands to back up. There is no cure for this beyond performance tuning. Also your thread pool implementation does not seem to be doing it's

Re: [h2] file corrupted

2015-03-04 Thread Noel Grandin
Hi I am sorry but your database is corrupt. All that you can do is to either restore from a backup, or try running the Recover tool, but the Recover tool is not guaranteed to restore all of your data. -- Noel -- You received this message because you are subscribed to the Google Groups H2

Re: [h2] Deadlock caused by h2 disk access

2015-03-04 Thread Noel Grandin
That's not a deadlock, that's just a slow SQL command that is causing your other SQL commands to back up. There is no cure for this beyond performance tuning. Also your thread pool implementation does not seem to be doing it's job of limiting the number of active threads properly, since you

Re: [h2] Re: Connection error - H2 Database Engine: New version 1.4.186

2015-03-03 Thread Noel Grandin
I'm sorry to say that you have a corrupt database. On Monday, 2 March 2015, Antonio Monforte aamonfo...@gmail.com wrote: Hi, *Error report:* URL :jdbc:h2:d:/biigeclipse/data/db;;AUTO_SERVER=TRUE on SQuirrel SQL Client Unexpected Error occurred attempting to open an SQL connection.

Re: [h2] Show indexes

2015-03-02 Thread Noel Grandin
There isn't one. If you want that information you can extract it from the INFORMATION_SCHEMA tables. Although it is general easier to use the MetaData classes and methods from the JDBC driver. -- You received this message because you are subscribed to the Google Groups H2 Database group. To

[h2] Automatic version column

2015-02-25 Thread Noel Grandin
It's not high priority at the moment, unless someone (*hint* *hint*) comes up with a patch :-) Also, most of the ORM libraries will do it for you. -- You received this message because you are subscribed to the Google Groups H2 Database group. To unsubscribe from this group and stop receiving

Re: [h2] Performance question - multi column query

2015-02-25 Thread Noel Grandin
I assume that you meant to write Where time=? And timeend? What does explain analyze say? You could also try adding a condition to your query like time?+x where x is the maximum gap between entries. On Wednesday, 25 February 2015, Uli u.wiel...@gematronik.com wrote: Hi, I have a table

Re: [h2] Is there a command to retrieve the running version of H2

2015-02-25 Thread Noel Grandin
http://h2database.com/html/functions.html#h2version -- 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 h2-database+unsubscr...@googlegroups.com. To post to this

Re: [h2] Automatic version column

2015-02-24 Thread Noel Grandin
It's not high priority at the moment, unless someone (*hint* *hint*) comes up with a patch :-) Also, most of the ORM libraries will do it for you. -- You received this message because you are subscribed to the Google Groups H2 Database group. To unsubscribe from this group and stop receiving

Re: [h2] Need help with [42102-169] Table not found errors.

2015-02-11 Thread Noel Grandin
Yeah sorry, I really don't have any ideas. I've never seen anything like that before. Perhaps you could set the trace level to DEBUG? http://h2database.com/html/features.html#trace_options that would at least capture the stack traces and some more information. -- You received this message

Re: [h2] preserve capitalization of column names in returned meta data

2015-02-09 Thread Noel Grandin
Do you have a short test-case ? On 2015-02-09 03:47 AM, Adam McMahon wrote: Hi, When we get the meta data of a result set, it appears that the capitalization of the column names is not preserved. Is there anyway to preserve it the returned meta results? -Adam -- You received this message

Re: [h2] Need help with [42102-169] Table not found errors.

2015-02-09 Thread Noel Grandin
Are you perhaps dropping and re-creating this table in your code? -- 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 h2-database+unsubscr...@googlegroups.com. To post

Re: [h2] org.h2.util.StringUtils.toUpperEnglish can be a bottleneck

2015-02-09 Thread Noel Grandin
Hi On 2015-02-09 02:21 PM, Steve McLeod wrote: final ResultSet resultSet = conn.prepareStatement(SELECT * FROM foobar).executeQuery(); int rowCount = 0; while (resultSet.next()) { rowCount++; final int columnCount =

Re: [h2] Need help with [42102-169] Table not found errors.

2015-02-09 Thread Noel Grandin
Could you post a chunk of the log around those errors? There might be something else nearby that indicates why that error occurred. I.e. that error might be a consequence of another error. -- You received this message because you are subscribed to the Google Groups H2 Database group. To

Re: [h2] Re: Interfaces instead of direct JTS reference

2015-02-05 Thread Noel Grandin
On 2015-02-05 02:45 PM, bocher wrote: Feel free if I'm wrong but if you're going just to use H2GIS (without modifying it), your application does not need to be a GPL licensed. The aim of the GPL license for H2GIS was to ensure that code was contributed back to the project and help us to

Re: [h2] Missing column type in right-hand parameter in ConditionIn

2015-02-04 Thread Noel Grandin
Hi Sorry this took so long, this patch has been applied. Nice work! Regards, Noel -- 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

Re: [h2] is not null -- not working

2015-02-04 Thread Noel Grandin
Probably because you have stored the string null into your DB instead of an actual NULL value -- 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

Re: [h2] Re: H2 database: slow query although index is used

2015-02-04 Thread Noel Grandin
Hi Sorry this took so long, your patch has been committed. Regards, Noel -- 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

Re: [h2] Using synchronized in user defined functions or not

2015-02-04 Thread Noel Grandin
Synchronized means that only thread at a time can execute the ALIAS, which will slow a multithreaded server down. Whether or not you need synchronized on that method is a different question, and depends on how the method is implemented. -- You received this message because you are subscribed

Re: [h2] PostgreSQL compatibility: generate_series (as an alias for system_range)

2015-02-04 Thread Noel Grandin
Hi Thanks a lot! Very nice unit tests! Your patch has been applied. Regards, Noel. -- 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

Re: [h2] Re: H2 database: slow query although index is used

2015-01-30 Thread Noel Grandin
HI Just to let you know that I am looking at your patch, it's just taking me a while because I don't have much time right now. -- Noel -- 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

Re: [h2] Re: Could data corruption being caused by non-accurate .lock.db information ?

2015-01-30 Thread Noel Grandin
Hi Hmmm, you are correct. I think I am going to let Thomas deal with this one, I obviously do not understand the file locking stuff sufficiently. Regards, Noel -- You received this message because you are subscribed to the Google Groups H2 Database group. To unsubscribe from this group and

Re: [h2] Re: Could data corruption being caused by non-accurate .lock.db information ?

2015-01-30 Thread Noel Grandin
FILE_LOCK=SOCKET would be safer in that situation. On 2015-01-30 10:03 AM, o.vanderm...@xso.nl wrote: Hi Noel, The database file will place on a network share and will be accessed from different computers. What should the FILE_LOCK be in your opinion? Olaf -- You received this message

Re: [h2] Any plan to add distributability for H2 over hazelcast ?

2015-01-28 Thread Noel Grandin
, Noel Grandin wrote: Hi sergi Nice to see h2 being used for something like that! Awesome! I see on that page that grid gain engineers have patched h2. Have those patches been contributed back? Regards, Noel. On Tuesday, 27 January 2015, Sergi Vladykin sergi.v...@gmail.com

Re: [h2] Re: Nested transactions in H2

2015-01-27 Thread Noel Grandin
Instead of spawning a thread and waiting for it, why not just make a method call? Then you'd be using the same connection by default. -- 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,

Re: [h2] Patch to allow declaring triggers as source code

2015-01-27 Thread Noel Grandin
Hi You have convinced me, thank you very much for working on this :-) Patch is committed. Regards, Noel. -- 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

Re: [h2] Any plan to add distributability for H2 over hazelcast ?

2015-01-27 Thread Noel Grandin
Hi sergi Nice to see h2 being used for something like that! Awesome! I see on that page that grid gain engineers have patched h2. Have those patches been contributed back? Regards, Noel. On Tuesday, 27 January 2015, Sergi Vladykin sergi.vlady...@gmail.com wrote: Hi, Apparently H2 is used

Re: [h2] The object is already closed 90007-175 vs 9007-185

2015-01-22 Thread Noel Grandin
It means that you're calling a method on the JDBC Statement object after you have called close() on the object, which is illegal. -- 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,

Re: [h2] Like statement with prefix

2015-01-22 Thread Noel Grandin
It should be doing that automatically, what does EXPLAIN http://h2database.com/html/performance.html#explain_plan say for your query? -- 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

Re: [h2] Patch to allow declaring triggers as source code

2015-01-22 Thread Noel Grandin
On Thu, Jan 22, 2015 at 5:16 PM, ILM ilm.informati...@gmail.com wrote: 1/ Not having to have the Trigger class on the server classpath That doesn't make any sense - the function you call and the object it returns still needs to be on the server classpath. 2/ Allowing to pass parameters, e.g.

Re: [h2] Unique key violation error after disconnecting and reconnecting to db - MVCC related??

2015-01-21 Thread Noel Grandin
If you could create a standalone test case we would be very interested, but the corrupted DB by itself is not that useful. Note that MVCC is still experimental (mostly because of the difficulty of creating test cases that find the corner case bugs) -- You received this message because you are

Re: [h2] Re: Nested transactions in H2

2015-01-21 Thread Noel Grandin
One simple thing you could try is just passing the connection into the child thread so that the parent and child use the same connection. -- 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

Re: [h2] Re: CREATE LOCAL TEMPORARY TABLE FOO TRANSACTIONAL AS ...

2015-01-21 Thread Noel Grandin
Why don't you just do CREATE LOCAL TEMPORARY TABLE IF NOT EXISTS xxx DELETE FROM xxx at the start of your trigger. Then you will have an empty table to work with. -- You received this message because you are subscribed to the Google Groups H2 Database group. To unsubscribe from this group

Re: [h2] Patch to allow declaring triggers as source code

2015-01-21 Thread Noel Grandin
Hi Thanks for the patch! Nice work! I am confused however - what is the benefit of this patch over just creating triggers in the regular fashion? Another way of solving my confusion would be to tell me what is the problem you are trying to solve? Thanks, Noel. On 2015-01-20 01:31 PM, ILM

Re: [h2] Could data corruption being caused by non-accurate .lock.db information ?

2015-01-20 Thread Noel Grandin
It would get an open error, file locked by other user, since it tries to open it in read-write mode. Assuming the fileserver is implementing locking correctly, of course. -- You received this message because you are subscribed to the Google Groups H2 Database group. To unsubscribe from this

Re: [h2] Re: Nested transactions in H2

2015-01-20 Thread Noel Grandin
On Tue, Jan 20, 2015 at 6:35 PM, Aaron Digulla adigu...@gmail.com wrote: What is the correct code for an application where several users can make concurrent changes to the database? Regular concurrent updates are not the problem, that should work just fine. Your original test case was this:

Re: [h2] Re: Recurring data corruption issues

2015-01-19 Thread Noel Grandin
What does your schema look like? (roughly) Are you using LOB's? It might be worth upgrading to the latest 1.4 release and running with MV_STORE=false to use the old storage engine. -- You received this message because you are subscribed to the Google Groups H2 Database group. To unsubscribe

Re: [h2] ALTER TABLE ADD uses too much memory

2015-01-19 Thread Noel Grandin
On 2015-01-18 10:03 AM, Steve McLeod wrote: Noel, you mentioned that you have some suggestions for making the data copying faster? I'm keen to hear them. The quicker the ALTER TABLE ADD ... statements run, the less chance my users experience power outage, or decide to force quit the

Re: [h2] Re: Nested transactions in H2

2015-01-19 Thread Noel Grandin
Well, yes, if you're going to be silly and hold the transaction open for longer than your lock timeout, that is essentially correct. But in the words of a famous doctor joke don't do that -- You received this message because you are subscribed to the Google Groups H2 Database group. To

Re: [h2] Re: Nested transactions in H2

2015-01-16 Thread Noel Grandin
Hi You are running the nested transaction in a second connection, which is never going to work. Regards, Noel -- 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

Re: [h2] Re: Nested transactions in H2

2015-01-16 Thread Noel Grandin
Also, your inner transaction is trying to update a row that does not exist, since it has never been inserted into the table. -- 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

Re: [h2] Trigger compatible definition for H2 and MySQL

2015-01-15 Thread Noel Grandin
Sorry, but there is no common syntax for triggers between MySQL and H2. -- 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 h2-database+unsubscr...@googlegroups.com.

Re: [h2] ArrayIndexOutOfBoundsException with H2 1.4.184

2015-01-14 Thread Noel Grandin
LockException and this is a problem for how my application is designed. Any other idea to try to resolve the problem of the db's size? Thanks 2015-01-14 8:50 GMT+01:00 Noel Grandin noelgran...@gmail.com mailto:noelgran...@gmail.com: On 2015-01-14 09:31 AM, Daniele Renda wrote: Hi Thomas

Re: [h2] ArrayIndexOutOfBoundsException with H2 1.4.184

2015-01-14 Thread Noel Grandin
On 2015-01-14 11:18 AM, Daniele Renda wrote: Well, I'm using org.apache.tomcat.jdbc.pool.DataSource with Hibernate. So I never use Connection.createBlob / createClob but I don't know if then Hibernate do it internally. Make a reproducible test case is quite hard because the application is

Re: [h2] ArrayIndexOutOfBoundsException with H2 1.4.184

2015-01-13 Thread Noel Grandin
On 2015-01-14 09:31 AM, Daniele Renda wrote: Hi Thomas, thanks for your reply and for your time. I'm using H2 1.4.184 from the day you released it. So I'm yet using it, but the db grows every day. Now it is 2.9Gb, Argh!!! I'm using Hibernate 4.3.7.Final in my application. Can you suggest me

Re: [h2] What would be the best update-candidate

2015-01-09 Thread Noel Grandin
what does your database URL look like? -- 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 h2-database+unsubscr...@googlegroups.com. To post to this group, send email

Re: [h2] What would be the best update-candidate

2015-01-09 Thread Noel Grandin
OK, that's good, there is nothing problematic in your settings. Probably the best update-candidate is 1.4.184, but add MV_STORE=false to your URL to keep using the old(stable) storage engine. On 2015-01-09 10:23 AM, Rinse Lemstra wrote: Connecting to H2 database:

Re: [h2] Re: Exception when opening DB after forced shutdown (1.3.176)

2015-01-08 Thread Noel Grandin
what does your DB URL look like? On 2015-01-09 02:52 AM, marcolopes wrote: Since we upgrade to *1.3.176* (from 1.3.169) in our final product, dozens of users are experiencing problems with damaged databases. I currently don't have access to all the logs, but the most recent was this one: --

Re: [h2] H2 Complete Change Log

2015-01-08 Thread Noel Grandin
You can browse back through the revisions of the changelog here: https://code.google.com/p/h2database/source/browse/trunk/h2/src/docsrc/html/changelog.html On 2015-01-09 02:30 AM, marcolopes wrote: Is there a way to access a complete Change Log? http://www.h2database.com/html/changelog.html

Re: [h2] Attempted to run 1.4.184 overnight and failed with ArrayIndexOutOfBoundsException

2015-01-07 Thread Noel Grandin
On 2015-01-07 06:08 PM, Kenton Garner wrote: I am still concerned about the problem that I had with the commit() throwing an exception while actually commiting the changes and then not allowing the rollback ( from the previous reply to this discussion ) Any thoughts on that yet?

Re: [h2] Re: Add support for MEDIAN aggregate function

2015-01-07 Thread Noel Grandin
I had a bash at implementing this about 6 months ago, but it's decidedly non-trivial. It doesn't fit into our existing framework, so We would have to implement it by adding another layer in our processing chain. -- You received this message because you are subscribed to the Google Groups H2

Re: [h2] ALTER TABLE ADD uses too much memory

2015-01-06 Thread Noel Grandin
On 2015-01-06 12:50 PM, Steve McLeod wrote: When our users update to the latest version of our desktop software that uses H2, we need to add several columns to a table with many columns that may have millions of rows. Some people have been getting an OutOfMemoryException during this process.

Re: [h2] does LIMIT improve performane

2015-01-05 Thread Noel Grandin
On 2015-01-04 05:14 PM, Adam McMahon wrote: If there is a limit, does the engine stop searching once the LIMIT has been found? Yes. -- You received this message because you are subscribed to the Google Groups H2 Database group. To unsubscribe from this group and stop receiving emails

Re: [h2] setQueryTimeout on a JDBC Statement seems to effect the entire connection not just the Statement object

2015-01-05 Thread Noel Grandin
On 2015-01-02 06:52 PM, Kenton Garner wrote: Is there a way for H2 to implement the setQueryTimeout() method to only effect the current Statement object. Not easily I'm afraid. I think we'd need to change the protocol to include the timeout on every command. Perhaps your connection pool

Re: [h2] Script tool puts blocksize option in wrong place in generated SQL.

2015-01-05 Thread Noel Grandin
Hi Thank you very much for the excellent problem report, I have committed a fix. Regards, Noel. -- 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

Re: [h2] File corrupted while reading record: 1060 of 43. Possible solution: use the recovery tool [90030-184]

2015-01-05 Thread Noel Grandin
Do you realise that you are using the FILE_LOCK=NO option, which can lead to corruption? http://h2database.com/html/features.html?highlight=FILE_LOCKsearch=file_lock#database_file_locking -- You received this message because you are subscribed to the Google Groups H2 Database group. To

Re: [h2] Re: Performance regression in versions starting 1.165 when using BLOB in temporary table

2015-01-05 Thread Noel Grandin
Hi Thanks for your problem report and investigation. I have committed a fix for this. Regards, Noel. -- 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

Re: [h2] CAST values contain decimals

2014-12-31 Thread Noel Grandin
probably because they are floating point? -- 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 h2-database+unsubscr...@googlegroups.com. To post to this group, send

Re: [h2] Disk space report

2014-12-31 Thread Noel Grandin
http://h2database.com/html/functions.html#disk_space_used -- 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 h2-database+unsubscr...@googlegroups.com. To post to this

Re: [h2] Recurring data corruption issues

2014-12-18 Thread Noel Grandin
On 2014-12-18 09:52 AM, Marc Heinz wrote: - Concurrent access attempts when stored on a network drive ? Probably that. Network drives are notorious for data corruption, especially when concurrently accessing the same file. I would do my best to detect that situation and prevent your users

Re: [h2] Really weird “Data conversion error”

2014-12-12 Thread Noel Grandin
What does the full stack trace look like? On Friday, 12 December 2014, Jules Bertrand jules.brt...@gmail.com wrote: I'm in front of a really weird Data conversion error when i'm trying to insert a row in a table in my H2 database (version 1.4.182). I don't know if i'm missing something or if

Re: [h2] Unable to execute RUNSCRIPT after running the recovery tool

2014-12-11 Thread Noel Grandin
On 2014-12-11 10:17 AM, o.vanderm...@xso.nl wrote: I was curious if you could find out why the recovery tool produces incorrect sql. It's a recovery tool, it's not magic. If your DB is corrupt, then it inherently has bad data and the tool cannot make a decision as to exactly what the

Re: [h2] Is compact possible without shutting down the database

2014-12-11 Thread Noel Grandin
what version are you using and what does your database URL look like? -- 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 h2-database+unsubscr...@googlegroups.com. To

Re: [h2] Can't use h2 database in a thread only with Java 8

2014-12-07 Thread Noel Grandin
what does your database URL look like? -- 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 h2-database+unsubscr...@googlegroups.com. To post to this group, send email

Re: [h2] Wrong results using the BETWEEN operator (H2 1.3.176)

2014-12-05 Thread Noel Grandin
Hmmm, this does not seem to be a BETWEEN issue, since it also fails with ZEILE_RELATIV =16 AND ZEILE_RELATIV = 18 On 2014-12-03 05:21 PM, christoff.schm...@finaris.de wrote: Hi, The following statement returns 0 rows as result, obviously, the result should be one row containing the value

Re: [h2] Connecting remote H2 embedded database over SSH

2014-12-01 Thread Noel Grandin
We don't support that at the moment, but it sounds like a reasonable feature, and one that should not be hard to add (since the JRE already provides most of the building blocks). Patches are welcome :-) On 2014-12-01 02:35 PM, ArchangelAurora wrote: Is there a way to connect to an H2

Re: [h2] like ? escape'' ... not working in 1.4.182 but was in 1.3.153

2014-11-28 Thread Noel Grandin
Can you create a standalone sequence of SQL commands? i.e. CREATE TABLE ... INSERT... SELECT and then I can test this and try to debug it. -- You received this message because you are subscribed to the Google Groups H2 Database group. To unsubscribe from this group and stop

Re: [h2] Re: Why mvstore is activated when I used page store (version 1.4.182)

2014-11-24 Thread Noel Grandin
Hi I just tried this and I'm not seeing that thread running. The only thing I can think of is that somewhere else in your code you are opening another H2 database and not using the MV_STORE=false option. Regards, Noel -- You received this message because you are subscribed to the Google

Re: [h2] USE SCHEMA not supported in MySQL mode?

2014-11-24 Thread Noel Grandin
Hi Thank you for your patch, this has been merged! Regards, Noel On 2014-11-25 02:08 AM, mful...@familysearch.org wrote: I have also tried out 1.4.181 as I have some scripts that specify use /db_name/. I believe the syntax is USE /db_name/ where USE is a reserved word and /db_name/

Re: [h2] Re: java.lang.NegativeArraySizeException (on H2 1.3.176 build)

2014-11-23 Thread Noel Grandin
Hi I have now committed a better fix for this. Regards, Noel -- 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 h2-database+unsubscr...@googlegroups.com. To post to

Re: [h2] Re: java.lang.NegativeArraySizeException (on H2 1.3.176 build)

2014-11-22 Thread Noel Grandin
Can you post your copy of ValueHashMap? -- 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 h2-database+unsubscr...@googlegroups.com. To post to this group, send email

Re: [h2] Re: java.lang.NegativeArraySizeException (on H2 1.3.176 build)

2014-11-22 Thread Noel Grandin
Hi Try changing the line in HashBase maxSize = (int) (len * MAX_LOAD / 100L); to maxSize = (int) (((long)len) * MAX_LOAD / 100L); I'm away from my dev machine so I can't prepare a patch right now, Regards Noel -- You received this message because you are subscribed to the

Re: [h2] Re: Correct way to close an H2 Database

2014-11-22 Thread Noel Grandin
The bottom line is that you should not need to do anything. It should just work fine by default. -- 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

Re: [h2] Correct way to close an H2 Database

2014-11-21 Thread Noel Grandin
Assuming you are not overridig the default behaviour, H2 should close itself cleanly on VM exit. -- 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

Re: [h2] DELETE throws a java.lang.IndexOutOfBoundsException

2014-11-21 Thread Noel Grandin
It's working for me, something like: Statement stat = conn.createStatement(); stat.execute(CREATE TABLE t1 (c1 INT)); stat.execute(INSERT INTO t1 SELECT X FROM SYSTEM_RANGE(1, 10);); prep = conn.prepareStatement(DELETE FROM t1 WHERE c1 IN (SELECT x from TABLE(x

Re: [h2] Why mvstore is activated when I used page store (version 1.4.182)

2014-11-21 Thread Noel Grandin
probably because PAGE_STORE=TRUE is not a vallid setting.​ you need MV_STORE=FALSE -- 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

Re: [h2] ConcurrentModificationException while creating Trigger

2014-11-20 Thread Noel Grandin
What does your database URL look like? -- 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 h2-database+unsubscr...@googlegroups.com. To post to this group, send email

<    5   6   7   8   9   10   11   12   13   14   >