Wake Up XI?

[Former Member]

We have a process that runs every morning at approx 3AM which sends

benefit data from SAP HR through XI to a SQL DB for online benefits

enrollment.

We have our job scheduling software run this and if it fails, to

immediately retry 3 more times with the same variant.

Several times a week, the initial job fails but a subsequent job

succeeds. It seems that we need to "wake up" the listening process

that's running.

I know when we were on 3.1h and used Taxware as our tax solution, the

daemon that ran at UNIX level had to be "woken up" periodically with a

call to it.

Anyone run into any similar behaviors?

Thanks,

Robert.

[henrique_pinto · Accepted solution]

Please provide more information regarding the technical scenario.

Is the connection between ERP and XI an RFC?

If it is, there may be some issues regarding the recconection after some time of unavailability.

Check this blog: /people/community.user/blog/2006/09/21/xi-rfcadapter-reconnect-issue

Regards,

Henrique.

[Former Member · Accepted solution]

henrique,

the source system is SAP HR ECC 6. We also had this same issue with our 4.7 HR system.

yes, SAP calls an RFC with XI as the destination.

EDIT: ha.. read the blog in that link.. He and I set up this RFC interface together - me on the ABAP side, him on the XI side.

i'll ask him why he's publishing blogs about this error and not fixing it!

[Former Member · Accepted solution]

Underthe JDBC channel settings, at the bottom of the settings page, you will find "advanced mode". Select the check box to make the advanced properties settings visible. Select 1 or 2 for the "Number of retries of Database transaction on SQL error" may be that will fix your issue and it might wake up the listener during the first try and on the second try, it might be successful.(give it a shot).

Also, select the checkbox "Disconnect from database after processing each message" this will ensure that the connection does not become stale after sometime.

Message was edited by:

Anand R Shankar

[henrique_pinto · Accepted solution]

Hey robert,

haha! Ok, then.

Let us know the results.

Regards,

Henrique.

[Former Member]

Hey

>>Several times a week, the initial job fails but a subsequent job

succeeds.

i had a similar kind of situation with HTTP adapter,it was related to network connectivity, a lot of times XI was unable to establish connection with the receiver system in the first try and when it tried next time it waz successful.

you may have a look at the network connection

Thanx

Aamir

[Former Member]

what's your target DB?

[Former Member]

our target is just a SQL database that sits on a network drive.

i'm an ABAPper so i apologize if my replies aren't from an XI view.