[Terence Gucwa]

Hi Russ, I’ve read some of your posts here and found them informative.

[Terence Gucwa]

Regarding my post, we’re on 5.7.  Yes, we have a couple of thread names longer than 15 chars, but they don’t seem involved in this issue, and are unique with the first 15.  Still, thank you Bob, for pointing that out and that’s really something we should get rid of.  

I have talked to Infor and yes, they would want us to upgrade before they would pursue this as a bug.  I’d be happy though, just to know how to avoid the problem, rather than having a fix.  It’s pretty rare, happens only in the wee hours of Sunday mornings, but when it happens, it’s a mess.

[Terence Gucwa]

Did anyone find an answer to this?  We’ve had the same problem – messages getting routed to outbound threads in the wrong site.  But this is in Cloverleaf 5.7.2, AIX 5.3.  The messages sit in the wrong recovery database because the outbound threads don’t exist in that site.

[Terence Gucwa]

OpenView Controller.  Don’t have that, but I certainly think that Cerner is not releasing the connection when they say they are.  They aren’t real open about all this.

[Terence Gucwa]

Just before it did start, I see:

[pdl :PDL :ERR /0:     pemrs_ob:09/01/2011 12:24:22] read failed: Connection reset by peer

[pdl :PDL :ERR /0:     pemrs_ob:09/01/2011 12:24:22] read returned error 73 (Connection reset by peer)

[pdl :PDL :ERR /0:     pemrs_ob:09/01/2011 12:24:22] PDL signaled exception: code 1, msg device error (remote side probably shut down)

But not when it I “should” have seen it shut down, some time before.

[Terence Gucwa]

Yes, it looks like:

[0:TEST] Found ‘.’ in subfield list

I believe it has to do with the use of subsubfields.  COPY statements using subsubfields as sources are just ignored, with the XLT continuing on.

[Terence Gucwa]

The “report” doesn’t exist until the translation itselfl; the translation has to make the “report”.  The HRL is a home-made combination of single-value fields and repeating groups extract from the Probation Dept’s case management system.  Simplified, the HRL contains a singly-occuring VRL with identifying and singly occuing info:

lname

fname

fingerprint-based-id

gender

location

dept-of-chidlren-family-servers-case-name

etc.

And several repeating VRLs:

{ family } While family.seg-type = ‘FM’

  name

  relationship

  etc.

{ int_party }  While int_party.seg-type = ‘IP’

  name

  association-type

  etc.

{ arrest }  While arrest.seg-type = ‘AR’

  offense-date

  etc.

So, presumably, the translation would put a report into either an xlt  temporary variable (@report) or into a tcl global variable (global _report), and then place it into OBX segments in an ITERATE.

Could that ITERATE be over a field, or list?

Or, alternatively, I guess each OBX segment could be constructed individually, duplicating the COPY staetments for all the other fields in each OBX segment.

[Terence Gucwa]

No, I can use VRLs with include files in an HRL.  For example:

pclonglist.vrl might refer to an Include file called pcgroups.vrl.

pcgroups.vrl might have in it groupa, groupb, groupc, and groupd.  Each of these groups might have, let’s say, dozens of fields.

pclonglist.vrl might have in it, let’s say, two fields – groupa and groupb.  The individual fields get pulled in from pcgroups.vrl.

HRL pcbigrecord.hrl might have pclonglist.vrl as a component, along with other components.  What I’m saying is the termination character in pcgroups.vrl is the one that counts; the termination character in pclonglist.vrl was just ignored.

Weird, if you ask me, but now that I know, I can cope.

[Terence Gucwa]

Yes, I understood that all along, but I gather the Global Properties in an Include(d) file has precedence over the Global Properties in the VRL file.  The Global Properties in the VRL were ignored in favor of the Include file.

At least that’s what I just experienced.  It’s not something I would have ever guessed.  Well, I guess I did guess, but it took awhile.

[Author]

Replies