I need to have a high availability database system but I can't do it through database cluster or master/slave. I need a jdbc proxy that knows to update multiple data source for a single update statement. I found HA-JDBC project http://ha-jdbc.github.com/ that does that but I was wondering if there are similar or better library than HA-JDBC.
You can see other project around HA features: Sequoia/C-JDBC (http://c-jdbc.ow2.org).
But some links on page are broken ;(
Related
In one of the use cases in my application there is a requirement to publish neo4j transaction data to oracle database in real time. I did google on it, but couldn't find a tool or plug-in which can help. Everywhere on internet talks about rdbms to neo4j sync. So I am planning to do this by manually invoking jdbc commands.
Can you please suggest something?
Had to write my own jdbc code.
I'm trying to connect to my on-premise Oracle database in order to migrate and copy some tables over to Azure SQL, but am not able to do so despite making sure all the connection parameters match the provided values in tnsnames.
Am I missing something? The error says the socket is closed but haven't gotten any useful information other than this prior issue, but doesn't contain any solution. I currently use Oracle 11.2.0.3 so the ADF connector should support this version.
Not sure what else I need to check. Any thoughts would be greatly appreciated!
Your screenprint shows you are using the AutoResolveIntegrationRuntime but as you say your Oracle db is on-premises you would need a Self-hosted Integration runtime (SHIR) as per this article. You would still need an SHIR for an IaaS Oracle db. Ideally the SHIR should be 'close' to the datasource so probably on-premises in the same network.
Do you have any proxy or firewall configured?
Have you tried creating the linked service and then testing the connection? Sometimes it occurred to me that I failed to test the connection of a new linked service but when creating it and retesting the connection is successful ...
I could see there are two ways to index a database records in GSA.
Content Sources > Databases
Using DB connector
As per my understanding, Content Sources > Databases does not support automatic recrawl. We have to manually sync after any changes occured in DB records. Is that correct?
Also, Would using DB connectors help in automatic recrawl?
I would like to check DB in every 15 minutes for the changes and update the index accordingly. Please suggest the viable apporach to achieve this.
Thanks in advance.
You are correct that Content Sources > Databases does not support any sort of automated recrawl.
Using either the 3.x Connector or 4.x Adaptor for Databases supports automatic recrawls. If you are looking to index the rows of databases only and not using it to feed a list of URLs to index then I would go with the 4.x Database Adaptor as it is new.
The Content Sources > Databases approach is good for data that doesn't change often where a manual sync is acceptable. That said though, it's easy enough to write a simple client that logs in to the admin console and hits the 'Sync' link periodically.
However, if you want frequent updates like every 15m I'd definitely go with the 4.x plexi-based adaptor, not because it's newer but because it's better. Older versions of the 3.x connector were a bit flaky (although the most recent versions are much better).
What flavour DB are you looking to index?
I have to load two Derby instances that reference same data Derby file in a remote repository.
Is it possible? How can I achieve this? Would you give detailed information?
Two different connections in the same Derby app is ok because Derby uses Java synchronization features to coordinate their access to the db.
Two different apps cannot use Java synchronization because that only works inside a single Jvm class loader so instead Derby locks the db and only allows one app at a time to access the db.
That is the benefit of the Derby network server because many apps can access the db simultaneously via the network server.
There is a hybrid mode by which one embedded app can also serve as the network server for other apps but it is complex and usually not the best choice.
Update: You can find more information about embedded servers in these docs: http://db.apache.org/derby/docs/10.11/adminguide/cadminov825149.html and http://db.apache.org/derby/docs/10.11/adminguide/radminembeddedserverex.html
or by searching the Internet for "Derby embedded server".
As descriped in the documentation it seems to be possible to replicate packages (spec and body) beside other database objects. But what will happen if the packages of the remote database (slave) are under permanent access? Is a replication under such circumstance possible too? I didn't find any hint in the documentation.
Many thanks in advance