I'm fighting with some memory leak. From time to time the vsize of the
process goes up with 100k. I have several function like this, and I
call them in each second. My memory growth happens every 10 minutes or
so.
int dbSqliteSelectSensorsToReport(sdmd_t *sdmd, sensor_t *sensor, int
deviceType,
On Fri, 3 Nov 2017 21:40:52 +
Simon Slavin wrote:
> Those two are in conclict. You might like to try putting BEGIN … END
> around your inserts just for testing.
That is a good idea, thanks. So it might look like this?
BEGIN TRANSACTION;
I'm implementing an application that is on a flash filesystem (JFFS).
The flash is connected to the CPU with SPI.
I experience about 1s INSERT time. My DB isn't big, just about 200k. I
implement a fifo like operation, so i have a fixed length row count. I
DELETE old data from time to time.
I
On Sat, 10 Sep 2016 13:31:51 +0100
Simon Slavin wrote:
> You do it after opening a connection to the database. Probably the
> first thing you do after sqlite3_open().
Thanks!
--
73 de HA5OGL
Op.: Levente
enough to set this once as the connection created? Or shall I set this
every time a do sqlite3_step()?
Thanks,
Lev
--
73 de HA5OGL
Op.: Levente
___
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin
On Sat, 20 Aug 2016 18:01:46 +0100
Simon Slavin wrote:
> Yes and yes, but do it this way.
>
> 1) Create the database file by opening it.
> 2) Do something that makes the file non-blank, like creating a table.
> 3) Issue "PRAGMA journal_mode=WAL"
On Sat, 6 Aug 2016 17:03:30 -0400
Richard Hipp wrote:
> Doing "PRAGMA journal_mode=WAL;" on your database (just once, perhaps
> from a command-line shell) will fix this for you.
I read in the documentation:
The WAL journaling mode uses a write-ahead
I only have a single thread, so
no other entity is accessing the database at the same time.
Other solution would be to pass the filename of the database, but then I have
to close the connection. Must I?
I don't know if it is possible.
Thanks for your hints in advance.
Lev
--
73 de HA5OGL
Op
JOIN B ON B.KeyField = A.KeyField
>JOIN C ON C.KeyField = A.KeyField
> ;
>
>
> You could of course add lots of other columns, this just the minimum
> to make the idea work.
Okay. Thanks for the hint. I think I go for 1.
Lev
--
73 de HA5OGL
Op.: Levente
the syntax.
Any help are welcome.
Thanks,
Lev
--
73 de HA5OGL
Op.: Levente
On Sat, 24 Jan 2015 20:59:22 +
Simon Slavin wrote:
> and set it to 6 (60 seconds) or so.
Okay, I try that, but I still don't understand how can a single threaded
application get a locked error.
Levente
Hi list,
I'm currently working on a www frontend that uses SQLite.
I sometimes get a database locked error when I access the database by
calling the execute() call. This is on PHP.
Reading the document
http://www.sqlite.org/rescode.html#locked
I learned that locked error occurs only on the
On Tue, 6 Jan 2015 10:45:40 +0100
Stephan Beal wrote:
> PHP's general-purpose mechanism for warning squelching is to prepend
> the command which is warning with an @ sign:
Thanks. This is what I need.
Thank you again.
Levente
___
sqlite-users
On Mon, 05 Jan 2015 18:29:52 -0800
Roger Binns wrote:
[...]
> Or in short, you squelch the diagnostics by providing acceptable SQL.
> You need to log/trace queries to find out which ones are the
> problems.
I know what this error/warning is. I
I'm using the PHP bindings for SQLite3. How can I squelch error and
warning messages? I do error checking, but the failing call emits the
messages.
Messages like:
Warning: SQLite3::prepare(): Unable to prepare statement: 1, no such table:
Thanks,
Levente
15 matches
Mail list logo