Showing entries 1 to 8
Displaying posts with tag: sync manager (reset)
MaxDB Synchronization Manager User Report

United Drugs had occasion to use the Synchronization Manager in production at this year’s convention. Mark Thomas (of United Drugs) and the MaxDB team at MySQL have been getting a use case document together to present the experiences. We had a great deal of help from the SyncMan dev team at SAP Berlin and Chris Hall of Just 4 Dental (who, by the way, also uses the Synchronization Manager in production).

I think the document has come together well and does a good job of describing what the industry can expect out of the Syncronization Manager in its current form. It should also be noted that this document has been presented to SAP Berlin and that many of the issues that have been brought to light have been addressed in the most …

[Read more]
Interaction

Hey all. Please comment on this entry. I’m trying to get an idea of who’s listening to what I say.

Why do you read this blog? Are you interested in community wireless networks? OpenGL development on Linux? Family? Business Intelligence? Synchronizaton Manager news? Freecycle™ development? Random Perl bits? Something else I’ve forgotten about?

I’m finding myself over-extended and I need to cut back on some of the things I focus on. Please let me know what is important to you so I know how best to re-organize my time, energy and other resources.

© cjcollier for C.J.'s WordPress of studlyness, 2006. | Permalink | No comment

Filed under ewn, …

[Read more]
Moved blog, took long weekend

For those of you who want to hear me read this instead of having to do so yourself (no hyperlinks), click here for .ogg and here for .mp3.

I migrated the blogs’ database and code from moonunit to avenger, both of which run Debian Sarge. Y’all should now notice a vast improvement in throughput. The ISP that Moonunit sits on should also see a large decrease in traffic.

Today, I added a campaign to raise funds for hosting. If you have any extra funds, please help me with colocation costs. It’s expensive and y’all use up a bunch of bandwidth. If you find this site …

[Read more]
MySQL Meetup 2006/05/01

Hannah, Scarlet and I attended the Seattle Meetup, since we heard Arjen was planning on being in town. We went to the normal location, but the owner told me they were closed due to their entire staff being involved in the civil rights

[Read more]
Useful MaxDB commands

To find a list of tables in a schema:

$ sqlcli -u TEST,TEST -d SYNCMANA "SELECT * FROM TABLES WHERE SCHEMANAME = 'TEST'"

To find a list of triggers on a table:

$ sqlcli -u DBSERVICE,SECRET -d SYNCMANA "SELECT * FROM TRIGGERS WHERE TABLENAME = 'T'"

To describe a table’s structure:

$ cat me && echo "-----" && 
   loadercli -d SYNCMANA -u TEST,TEST -b ./me && echo "-----" && 
   cat tabledef.txt
dataextract for dataload table T
outstream file 'tabledef.txt'
outstream file 'idontcare'

-----
Loader protocol: '/home/cjcollier/.sdb/loader/log/loader.prt'
Loader packages: '/home/cjcollier/sdb/loader/packages'

User TEST connected to database SYNCMANA schema TEST on local host.
dataextract for dataload table T
outstream file 'tabledef.txt'
outstream file 'idontcare'
Successfully executed

-----
//
CREATE TABLE "T"
(
        "I"  Integer  NOT NULL,
        "C"  Varchar (32) UNICODE,
        PRIMARY …
[Read more]
Synchronization Manager: MySQL as replication destination

Earlier this week, I wrote about the Synchronization Manager and gave a step-by-step tutorial on getting an initial scenario implemented. In this article, I’ll expand on the previous tutorial by introducing the concept of uni-directional synchronization.

Uni-directional participants act as destinations for replicated data. These participants are sometimes called “slaves,” because they are not capable of replicating changes to other members of the synchronization scenario. Theoretically, any RDBMS that provides an appropriate JDBC interface can participate as a replication destination with little effort on the part of the DBA. As an example of this, I have documented my experience creating a uni-directional MySQL participant.

One situation DBAs might use databases synchronized as uni-directional participants is when they want to provide information to clients …

[Read more]
Creating a Synchronization Management Scenario

This is a follow-up to my earlier article on the MaxDB Synchronization Manager. In the months since my prior article, the Synchronization Manager has proved itself production ready and capable of replicating real-world databases in demanding scenarios.

Take the following as theoretical and emperical examples of such scenarios (not necessarily in that order):

i) For its yearly trade show, a company needs to process user registration. Because of the high volume of attendees, there must be more than one system through which registration takes place. Therefore, the registration software is to run on multiple separate laptops. At the end of each day, the laptops synchronize their registration information with each other and a master registration database.

ii) In order to reduce human error and increase efficiency, a grocery store …

[Read more]
Customer?s experience with Sync Manager
14:34 < MThomas> Time to install and configure Sync Manager in development
                 environment: 4 days, one of which was 15 hours long.  Time to
                 install and configure in production: 30 minutes.

I’ve written a bit of documentation that might help you get things running in the same way:

© cjcollier for C.J.'s WordPress of …

[Read more]
Showing entries 1 to 8