Problem with Communication Channel reading files

[Former Member]

Hi,

I am currently facing a wierd situtation with my communication channels.

There are two interfaces which reads files with names *C01.xml and *M01.xml.

In Dev system everything was working fine. In production system out of 1000 C01 files some of them are being read by the other interface and they are apparently being throwed into error.

I cross checked several times the naming of files in the channel. There's no error. But how could some files are being read by the second interface where as maximun no.of files are read by correct interface.

Actions i have taken:

We tried refreshing the cache . And tested again. it didnt work.

i stopped channel 2. BUt even then some files are not being read by first interface even though the name is C01**********

Any ideas for this issue?

[Former Member · Accepted solution]

for channel with other interface you may be picking up files with prefix as "*".

please do following,

1. go to that other interface -> channel check and change it there or change in your channel.

2. if you have . extention to pickup file with interval might be eqaul to of what it triggers 2nd interface to pick file C01, you may increase this, but doing so will throw error again.

best advice would be to use 1st option.

[Former Member · Accepted solution]

No i am not using *. i mentioed the naming in first post are those used inthe channels.

Out of 1000 only 2-3 are being picked by wrong interface. where as rest all are processed fine.

[Former Member · Accepted solution]

hi,

You did not answer: what's your polling, and do you have an external triggering with a too quick frequency?

I know you said you refresh cache, but if everything is ok in your config... just to be sure... go to RWB > Cache Monitoring. then cache instance = Central Adapter Engine, and cache object = channel. Put your CCs and check its content, especially the field:

   <Attribute>
      <Name>
      file.sourceFileName
      </Name>
      <Namespace />
      <Value>
      *Here !!!!!!*
      </Value>
      <Flag>
      nonTransportable
      </Flag>
   </Attribute>

After that, if you are sure that your PI Scenario (config) is ok, your cache is ok, then as I suggested you, you should search for an oss notes and write a message to SAP.

Regards.

Mickael

[Former Member · Accepted solution]

Restarting the Server has solved the issue. Now its going fine..

Thanks for ur different suggestions guys

[Former Member]

hi,

1. How this sender CC is triggered: by an external tool or by using a Pooling ? In both cases, do you let enough time to PI to pick-up your > 1000 files ?

Like not have a pooling = 1min, or an external triggering with a "STOP just 1min after a START" repeated with a short frequency.

and do you have a firewall in the middle? if yes, you should increase the polling time or the scheduling...

2. You certainly checked the sender CC, but have you done a "where-used list" on your two Communication Channel ?

maybe by mistake, you use one twice in a Sender Agreement.

3. Are you able to reproduce this case in Pre-prod.

Note: if you really have C01 files treated by the 2nd interface (instead of only the 1st interface)..., an easy solution could be to use two different folders.

Anyway, check if there are OSS notes about such a bad behavior: [FAQ: XI 3.0 / PI 7.0/ PI 7.1 File Adapter|https://service.sap.com/sap/support/notes/821267], and as your issue is in production, open a message to SAP !

regads.

Mickael

[Former Member]

HI

1) Reg FireWall i am not sure of it need to ask

2)Sender agreements are fine. Each channel is used by one SA.

3)No we are not able to reproduce in preprod. In Preprod there's no error. All 1k+files are being processed by coreect interface.