problem with vpn connection

Post questions here relative to DataStage Server Edition for such areas as Server job design, DS Basic, Routines, Job Sequences, etc.

Moderators: chulett, rschirm, roy

Post Reply
tom
Participant
Posts: 46
Joined: Fri Oct 14, 2005 12:38 am

problem with vpn connection

Post by tom »

Hi,
Am trying to connect to server which is located in far location.but am facing problem in the response ie.., if am trying to open a job,it takes more than half hour for the job to pop up.the connectivity is purely VPN,after that i map with the server.and open the Datastage manager window,to open the job bcos am not able to open the job using the Datastage designer window.when i try to open a job using designer i get a error msg as"(40531)The object manager call to open an exsiting job could not be made,or failed within the object manager".can any one help me in the issues above.
Devlopers corner
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

Tom,

if you open up a cmd window on your PC and execute the command "tracert {hostname/host IP}" what kind of average times are you getting? The TCP/IP connection between client and host might not be the most efficient, but we used to use modems at 9600 Baud to work from home and it would suffice. If you average times are less than several seconds then I would also look at your server's performance intead of concentrating on network latency and throughput.

Opening up a job can take a long time, especially in a project category with many entries. Especially the designer/manager options that have to traverse all jobs in all categories can take a long time - my particular one is the pop-up window in the designer that lets you open a job, this can take 10 minutes on one system at my current site, whereas using the tree structure list on the left to locate and open the same job will take only seconds. In this case the project is split into many categories with few jobs in each one, but does have a total of over 2200.
tom
Participant
Posts: 46
Joined: Fri Oct 14, 2005 12:38 am

Post by tom »

Hi ArndW,
i tried tracing using the command ,am getting the time as 1086ms,1035ms,1106ms.and atlast it say trace complete.and i have only 24 jobs in a category.and am facing another problem too,after the job is up when i try to open the transformer stage i get an error msg as "(40503)A call to an OLE server has failed,or a runtime error occured within the OLE server itself",the funny thing is am able to open the properties window of the other stages in the same job.i feel like am in middle of the warfield,with the error msgs as my enemies.












ArndW wrote:Tom,

if you open up a cmd window on your PC and execute the command "tracert {hostname/host IP}" what kind of average times are you getting? The TCP/IP connection between client and host might not be the most efficient, but we used to use modems at 9600 Baud to work from home and it would suffice. If you average times are less than several seconds then I would also look at your server's performance intead of concentrating on network latency and throughput.

Opening up a job can take a long time, especially in a project category with many entries. Especially the designer/manager options that have to traverse all jobs in all categories can take a long time - my particular one is the pop-up window in the designer that lets you open a job, this can take 10 minutes on one system at my current site, whereas using the tree structure list on the left to locate and open the same job will take only seconds. In this case the project is split into many categories with few jobs in each one, but does have a total of over 2200.
Devlopers corner
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

over 1 second is a very bad latency and turnaround - but not enough to cause all your woes. If you search the forum for your second error you will find some threads that seem to point at a DS client issue on XP which can be resolved by a vendor patch.

Have you gone out and done a check for project integrity and a re-index (through the DS.TOOLS TCL command)? If you monitor your server performance does your %idle consistantly stay below [arbitrary amount] 15%? Does the error happen for just this one job or for all jobs?
tom
Participant
Posts: 46
Joined: Fri Oct 14, 2005 12:38 am

Post by tom »

Hi ArndW,
can u plz xplain me in detail how to go with the DS.TOOLS TCL command.and i get the same error msg for all the jobs.

ArndW wrote:over 1 second is a very bad latency and turnaround - but not enough to cause all your woes. If you search the forum for your second error you will find some threads that seem to point at a DS client issue on XP which can be resolved by a vendor patch.

Have you gone out and done a check for project integrity and a re-index (through the DS.TOOLS TCL command)? If you monitor your server performance does your %idle consistantly stay below [arbitrary amount] 15%? Does the error happen for just this one job or for all jobs?
Devlopers corner
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

The DS.TOOLS command can be called from the TCL command line or from the command line tool in the DataStage administrator client. There are 6 options at versions 7 and below, you will need to make sure that nobody else is using the project (including your own designer/manager/director clients) and execute options (4) [Check integrity of job files] and 2) [Rebuild indices].

If you call this menu from the DataStage administrator it will not be formatted properly in the window, but you can still execute the options. If you need information on calling up the DataStage TCL environment then please use the search facility to get detailed instructions.
Post Reply