Search results for query: *

It's hard to say how this affects performance, we do have this option set on most of our queues, but taking into account all the other factors we know impact performance we have tended not to worry about this one, Regards, Philip

Assuming you are using MQSeries, the way Mercator browses the queue before retrieving messages off the queue means that it occasionally misses messages on queues that are fed by other Mercator maps. Adding the option -REFRESH 60 (refresh the queue every minute) to the adapter command solves the...

The only thing I can suggest without seeing your map is to take out the PUT command for now and create a rule to generate the file name only and see how many file names you get. Or upload your map to www.eaiusergroup.com.

What I was trying to say was, that if any of the maps that running are holding a resource that the init pending maps require, then the maps will stay on init pending.

Try: -TO recepient1 -TO recepient2 in your put command or do multiple put commands.

Yes, we had similar problems with version 6.5 and never used it in a production evironment. Upgrade to version 6.7.

Unless I'm missing something, the rule for the detail record would be EXTRACT(Input Record, Debit Amount:Input Record = Credit Amount:Input Record)

Is batch number part of a series? In that case you will get a file for each batch number. Try specifying a specific index, e.g. TEXT(BatchNumber Field:Line[1]:UnidataInput) to just get one file.

Check whether any of the maps require resources held by another map, the easiest way to check this through the Management Console, Regards, Philip

You should have no problem using Oracle 9i client, but you have to rename a Mercator Oracle library (see installation instructions). As for the error, check the mapname.dbl file in your map directory (you might have to add the option -TE+ to your connect string) as you don't get the underlying...

Instead of using the backup feature, try using the PUT command inside the map to backup the file, at which point you can use anything to specify the filename.

As you are not triggering on the source file you are writing to, I can't see it being any different to any other target you might write to. The only thing is that all your maps will single thread because they all need to write to read and write to the same file. Have you checked the Management...

What error do you get? We have quite a few maps that read and write to the same file, although they are all run maps and don't run directly under the Event Server.

Have you tried creating a type tree like this: H1(1:1) record(s) T1(1:1) where record is choice or partitioned group of D1, D2, D3 etc.

If you're running 6.7 you can use the Management Console to see what Mercator is doing and detect any resource conflicts.

Generally when you have a problem like this you're stuck until Mercator get round to fixing it. You could try downgrading to 6.7 no SP, we tried the SP and found it caused more problems than it solved. Otherwise you could try to move your database extracts out of the realm of the event server...

We had endless problems with Mercator 6.5.2 and MQSeries which Mercator blamed on bugs in MQ so we skipped 6.5.2 and went straight up to 6.7 which has been quite stable. One thing to try is setting your maxthreads to 20 as Mercator has a nasty habit affecting the performance of MQ and producing...

MQ has two different ways of connecting, server and client connections (see your MQ documentation on the differences between these types of connections). Mercator simply has 2 different adapters to cater for this. Assuming you have a set of channels between your local and remote queue managers...

The problem is that the table your accessing is no longer in the schema of the user you are logging on as and as far Mercator is concerned you are accessing a different table (lawson.cbcheck instead of cbcheck). You can do one of the two things: - Ask the DBA to setup a public synonym (i.e...