I'm using the Oracle foreign data wrapper and would like to have local copies of some of my foreign tables locally. Is there another option than having materialized views and refreshing them manually?
Not really, unless you want to add functionality in Oracle:
If you add a trigger on the Oracle table that records all data modifications in another table, you could define a foreign table on that table. Then you can regularly run a function in PostgreSQL that takes the changes since you checked last time and applies them to a PostgreSQL table.
If you understand how “materialized view logs” work in Oracle (I don't, and I think the documentation doesn't tell), you could define a foreign table on that and use it like above. That might be cheaper.
Both of these ideas would still require you to regularly run something in PostgreSQL, but you might be cheaper. Perhaps (if you have the money) you could use Oracle Heterogenous Services to modify a PostgreSQL table whenever something changes in an Oracle table.
Related
Apologies if this is answered in the docs or somewhere else obvious, but I have just gone through the rigmorale of having to use postgres_fdw to get data from another database (quite different to the ease of MS SQL!)
I ran an IMPORT SCHEMA statement to get the foreign table/databse data into my working database. Will the data in this schema automatially update whenever the data in the foreign database updates or is this IMPORT SCHEMA a copy of the data?
A foreign table does not contain any data. Querying it will yield the current data in the remote table. Think of it as something like a view.
If you are talking about metadata changes (ALTER TABLE), the analogy to views is also helpful: such changes will not be reflected in the foreign table, and you need to run an ALTER FOREIGN TABLE statement.
Perhaps it would make sense to put the data for these databases in a single database in two different schemas.
I use postgresql. I have many databases in a server. There is one database which I use the most say 'main'. This 'main' has many tables inside it. And also other databases have many tables inside them.
What I want to do is, whenever a new row is inserted into 'main.users' table I wish to insert the same data into 'users' table of other databases. How shall I do it in postgresql? Similarly I wish to do the same for all actions like UPDATE, DELETE etc.,
I had gone through the "logical replication" concept as suggested by you. In my case I know the source db name up front and I will come to know the target db name as part of the query. So it is going to be dynamic.
How to achieve this? is there any db concept available in postgresql? Or I welcome all other possible ways as well. Please share me some idea on this.
If this is all on the same Postgres instance (aka "cluster"), then I would recommend to use a foreign table to access the tables from the "main" database in the other databases.
Those foreign tables look like "local" tables inside each database, but access the original data in the source database directly, so there is no need to synchronize anything.
Upgrade to a recent PostgreSQL release and use logical replication.
Add a trigger on the table in the master database that uses dblink to access and write the other databases.
Be sure to consider what should be done if the row alreasdy exists remotely, or if the rome server is unreachable.
Also not that updates propogated usign dblink are not rolled back if the inboking transaction is rolled back
I'm currently working on dumping one of our customer's database in a way that allows us to create new databases from this customer's basic structure, but without bringing along their private data.
So far, I've had success with pg_dump combined with the --exclude_table and exclude-table-data commands, which allowed me to bring only the data I'll effectively need for this task.
However, there are a few tables that mix lines which references some of the data I left behind with other lines that references data that I had to bring, and this is causing me a few issues during the restore operation. Specifically, when the dump tries to enforce FOREIGN KEY constraints for certain columns on these tables, it fails because there are some lines with keys that have no matching data on the respective foreign table - because I chose to not bring this table's data!
I know I can log into the database after the dump is complete, delete any rows that reference data that no longer exists and create the constraint myself, but I'd like to automate the process as much as possible. Is there a way to tell pg_dump or pg_restore (or any other program) to not bring rows from table A if they reference table B if and table B's data was excluded from the backup? Or to tell Postgres that I'd like to have that specific foreign key to be active before importing the table's data?
For reference, I'm working with PostgreSQL 9.2 on a HREL 7 server.
What if you disable foreign key checking when you restore your database dump? And after that remove lonely rows from the referring table.
By the way, I recommend you to fix you database schema so there is no chance wrong tuples being inserted into your database.
I have a report in SQL server and I am migrating this to Oracle.
The approach I used in SQL server is load sum(sales) , person for given month into temporary tables (hash tables) and use this table to join with other transaction tables show the details, but when it comes to oracle I am not sure if I can use the same method here, because hash tables (temporary tables in SQL server) are specific to session and might not create any problem with output, please advise if there is anything in oracle which is analogous to that.
I came to know there are global temp tables in oracle, do they work in the manner I mentinoed above, also
If a user has no create/drop table privileges can they still use gloabal temp tables?
please help me.
You'll have to show some code or atleast some pseudo-code of how your process runs for anyone to help you. Having said that...
One thing that is different in oracle compared to temporary tables in other databases is that you do not create them each time you need them. You create them once and the data in the table is present either until you commit/rollback (transaction based) or until you end your session (session-based global temporary tables). Also, The data in a temporary table is visible only to the session that inserts the data into the table..
If you are generating the output files once and you don't need that data later, then Global temporary tables would probably fit in cleanly, with some minor changes.
Since you do not create the temporary tables each time you use them, you don't need the create/drop privilege. All you'd need is the insert/read privilege. Just read will not help because you cannot read another session's data anyways, so there is no use for it.
What options do I have for importing a MySQL table using the InnoDB engine, where the table already exists, without losing read access to the existing table? Generally I am referring to situations where ALTER TABLE is really slow, and I instead want to dump the table, modify the dump, then import it.
If I use mysqldump to make the dump, then when I reinsert it drops the table before inserting, so read access is of course lost until the import is complete.
Can I simply change the tablename in the dump, import that, then when that's done, drop the old table and rename the new one? Is there a better way? Thanks in advance.
If your table is of the same structure, I don't see any problems. You need just to skip "DROP TABLE" operator. Almost all dumpers have the option to not include DROP TABLE/CREATE TABLE in the dumps. Personally I'd recommend Sypex Dumper.
If you want to change table structure without locking table, I think the best way is to use
pt-online-schema-change - ALTER tables without locking them
use MySQL Enterprise backup for this http://www.mysql.com/products/enterprise/backup.html