Hi,
Thank you for your answer, I see.
I tried, and H2 compressed into a 10% size ( 26MB file size to 2.4MB file
size).
I created only one table and estimate the compression rate. Is this correct
way ? Do I have to use file size ?
And another question.
On the other database, I also create only
Hi Thomas,
Thank you for the reply. We moved over to 1.3.175.jar and experienced
several issues with our user databases within days of merging over. This is
with databases created using 1.3.169 but then using 1.75.jar as its source
within the application. These issues were creating similar
This is a fairly generic issue.
Everytime a new release is published it should tell 2 things.
- Can I just update the jar file or do I need to recreate my database
- Do I need to upgrade also the client jar ie. has there been some
changes that affects the network protocol or some other aspects
Hi Thomas,
Thank you for the quick response and the patch!
Is this a one-off fix, or do you plan to include it into future versions?
Thanks,
--
Bartók István
On Wednesday, May 14, 2014 7:44:15 AM UTC+2, Thomas Mueller wrote:
Hi,
Yes, this is a bug. I have a patch:
Index:
I believe, I've found out the cause of the problem.
When H2 seeks beyond the current file-size and tries to write at the new
channel position, it creates a hole in the file. So far, so good.
However, it then erroneously assumes (in its page-allocation internals
somewhere) that this hole
Hi,
Our application uses an embedded H2 database (1.3.173). During execution,
table functions are created. Some of them execute select-statements in the
H2 database.
When not running in in multi-threaded mode, and if multiple threads (each
with separate connection) are accessing the same table
That is a classic example of an ABBA lock, and we are happy to track them down
and fix them.
Can you identify for us a stack-trace where it synchronizes over the database
while holding a lock on a TableView?
If you are using the Eclipse debugger, this is very easy - when it is paused,
Thanks a lot for answers, the MV_STORE=FALSE setting is working !
The does not allow to retrieve the table informations was about select
query and H2 console: select query answers with 42102 error code, and H2
console does not show the table in the left panel.
Using the MV_STORE setting allowed
Hi,
see the stack trace below, I marked the positions where synchronization
occurs over a TableView (red) or Database (blue).
The root cause is, as explained, that the table function implementation
within my class (DataElementTableFunctionManager) triggers a new query
execution to H2, some
Thank you very much! Nice work.
Patch committed.
On 2014-05-13 17:41, Nicolas Fortin (OrbisGIS) wrote:
Hi,
I found an error with Index cost hint computation on spatial index. ScanIndex
give wrong cost if _rowid_ is used, H2 use
the next available index.
Please see this issue for more
Hi
Thanks for the stacktrace.
I think I've fixed this in SVN.
You could either build H2 yourself, or wait until tomorrow and try out the
daily build.
Regards, Noel
On 2014-05-14 16:09, christoff.schm...@finaris.de wrote:
Hi,
see the stack trace below, I marked the positions where
Hi,
That's interesting, I didn't know. I believe the MVStore (the 1.4 version
of H2) no longer relies on the space to be filled with zeroes.
For your file system implementation, what I would do is provide a
compatibility flag that fills the space with zeroes.
Regards,
Thomas
On Wed, May 14,
12 matches
Mail list logo
