What are those and what do they do.
When I was checking out Sybase Central I stumbled on Dbspaces. It mentioned sometimes a pathname like c:\dev\j2db\build\database\empty_crosspplatform\example.db
and in databases I use it mention e:\notariaat\assyst.db
What are those dbspaces doing and what is their purpose?
It seems I cant remove them. Is it harmfull?
A SQL Anywhere database can consist of multiple database files. A
DBspace is an extra database file. Extra database files are typically
used to separate data from different tables onto different hard drives,
giving queries a performance boost through better disk I/O.
Deleting the DBspace files may cause whatever program is using the
database to fail.
david_jonker:
Deleting the DBspace files may cause whatever program is using the database to fail.
What if the filenames mentioned don’t excist, like in my case
Hmm. That’s a good question.
If the application accesses the tables in the
older DBspaces (or if you think they do), then
you will likely want to recreate the tables within
those missing DBspaces. You can create these
tables either in a separate DBspace, or within
the main (SYSTEM) database file.
Either way, if you are not going to recreate the
DBspaces, you can just delete them.
David
david_jonker:
Either way, if you are not going to recreate the
DBspaces, you can just delete them.David
That’s just the problem…you can’t delete them
Do you get any sort of error message?
It’s just greyed out
Okay. I’m pretty certain we can still delete them.
You’ll need to start Interactive SQL.
Within Sybase Central, right-click on the database
in the Folders pane that contains the DB space to be
deleted. This will open a context menu. Choose the
Open Interactive SQL menu item.
Then, type the following SQL into the SQL Statements
pane within Interactive SQL:
DROP DBSPACE oldspace;
Where ‘oldspace’ is the name of the DB space. Execute
the SQL statement.
David
Not allowed while ‘DBA’ is using database…
But dont you need at least one DBSPACE? The DBSPACE named ‘SYSTEM’ just simply has the wrong pathname. I noticed also that you can allocate space for a DBSPACE. Could you use this to prevent fragmenting of the db file?
Hi David,
since I have got your attention
I am also trying to use the index consultant. But everytime it gets to analyse about 164000 queries SYbase Central gets disconnected. And the ASA service quits. Has it something with the page-size beeing to small and/or DBSPACES?
RE: DBspace…
The SYSTEM DBspace is the default/main db space. I’m assuming you
mean that the SYSTEM dbspace has the wrong pathname to the other
DB spaces?! If so, it’s not a big deal. It won’t affect performance at all.
Yes, you can allocate additional space for a DBspace, just like you can
for a “regular” database (i.e. only has the SYSTEM dbspace). This will
definitely help with fragmenting of the DB file, because it will allocate
(attempt to) one continuous block of disk space.
RE: Index Consultant
It shouldn’t be doing this. 164,000 queries is a big number. Are you sure
you’re not running out of memory? That could possibly be causing the
connection to close… If you turn on logging and send the file, we can try
to figure out what the problem is. Turn on logging for the server by
specifying:
-oe templogfile.txt
David
david_jonker:
RE: DBspace…The SYSTEM DBspace is the default/main db space. I’m assuming you
mean that the SYSTEM dbspace has the wrong pathname to the other
DB spaces?! If so, it’s not a big deal. It won’t affect performance at all.
Yeah…that’s what I ment.
david_jonker:
Yes, you can allocate additional space for a DBspace, just like you can
for a “regular” database (i.e. only has the SYSTEM dbspace). This will
definitely help with fragmenting of the DB file, because it will allocate
(attempt to) one continuous block of disk space.
david_jonker:
RE: Index ConsultantIt shouldn’t be doing this. 164,000 queries is a big number.
That was just a workload of 4 hours…At first I had a workload of 350,000 queries, after saving the workload, everything froze and the ASA service quit
david_jonker:
Are you sure
you’re not running out of memory? That could possibly be causing the
connection to close… If you turn on logging and send the file, we can try
to figure out what the problem is. Turn on logging for the server by
specifying:-oe templogfile.txt
David
Where do I do that…in the configuration part of ASA service?Something like
-ti 0 -x tcpip{dobroadcast=no} -qs -qw -o "C:/Program Files/Servoy/sybase_db/sybase_log.txt"
"C:/Program Files/Servoy/database/assyst.db"
-oe "C:/Program Files/Servoy/sybase_db/templogfile.txt"
Another question. Does Sybase give training in the Netherlands. I feel like we are not using the full potential of Sybase and would to spice up our knowledge.
Yes, that is correct. Just add it to the other startup
parameters.
We can/do offer personal training in the Netherlands.
The person to contact is Rob Waywell:
rwaywell_AT_sybase.com
David
Hi David,
I can’t seem to get that logging working (see code below).
I tried to do an analysis again with 37.000 queries. This time I get to actually analyse something, but it stops when it reachs 1031. The connection is dropped. And I have to restart the service.
-ti 0 -x tcpip{dobroadcast=no} -qs -qw -o "C:/Program Files/Servoy/sybase_db/sybase_log.txt"
-oe "C:/Program Files/Servoy/sybase_db/templogfile.txt"
"C:/Program Files/Servoy/database/servoy_repository.db"
"C:/Program Files/Servoy/database/assyst.db"
"C:/Program Files/Servoy/database/pcdb.db"
"C:/Program Files/Servoy/database/log_data.db"
Thankx in advance
OKay I got the templogfile working…but it didn’t mention anything.
I got the following errors in Sybase Central
java.sql.SQLException: [Sybase][ODBC Driver][Adaptive Server Anywhere]Communication error
[Sybase][ODBC Driver][Adaptive Server Anywhere]Connection was terminated
SQLCODE: -308
SQLSTATE: 08S01
Tried the following:
-deleted all data from all the tables
-ran index consultant and performed a new analysis based on a previous workload
-this the analysis ran all the way through this time(Except it did an analysis twice the amount of queries)
-When it finished it generated some errors (Some queries were discarded after cauzing parser/optimizer errors)
-Then a Sybase Central Internal Error occured (see attachment for Stack Trace)
-When you want to see the indexes in script-form…Sybase Central doesn’t show them
Is it usefull to send the db file with the workload and without the data to you?
Thanx in advance
scj_errors.txt (38.6 KB)
david_jonker:
RE: DBspace…The SYSTEM DBspace is the default/main db space. I’m assuming you
mean that the SYSTEM dbspace has the wrong pathname to the other
DB spaces?! If so, it’s not a big deal. It won’t affect performance at all.
The third page of the Sybase Central “Create a new index” wizard presents the following question:
“In which dbspace do you want to store the index?”
Directly below this question is a list that contains one line that is identified with a name of “SYSTEM” and a path of “c:\temp\holders.db”.
I assume it is showing that path because at some time in the past I probably unloaded the database to that path. However, the actual database is located in the database subfolder of the Servoy folder.
There does not appear to be any way for me to change the dbspace location that is shown by the wizard. Why does it ask where to store the index and then not give me the ability to specify the location? How can I get around this problem?
Is it safe for me to ignore the path that it is showing and assume that the new index will be saved within the holders.db file in the Servoy database folder?
Dean Westover
Choices Software, Inc.
Sybase Central version 4.3.0.2184
Adaptive Server Anywhere version 9.0.1.1751