[Jerry Sawa]

I do and alert with “prestart”.

Instruction are in this linked topic.          Randomly not receiving ACK’s – Clovertech

[Jerry Sawa]

Following

[Jerry Sawa]

[xlt :thre:ERR /0:dohbeds_out_xlate:03/23/2026 00:53:27] XLATE ICL server open failed, iclErr=1

“XLATE ICL server open failed, iclErr=1” is one of the errors we check for after restarting a process.  If it happens, the error appears in log immediately after starting the process.  If we find that error in the log, we recycle the process.  We’ve had that in place for several years and has worked for us 100% of the time.

• This reply was modified 3 months, 3 weeks ago by Jerry Sawa.

[Jerry Sawa]

Peter, your log reach 2GB because of the error, correct?

When you stopped/started the process then all worked as expected.

Could you send a portion of the process log from when process was started to the point where you started getting the error?

Quite often we get error “ICL initialization failed” when starting a process.  We have code that checks for ICL issues upon startup of process.  If found, we stop and restart the process.

• This reply was modified 3 months, 3 weeks ago by Jerry Sawa.

[Jerry Sawa]

We have a vendor that sometimes does not send the PDF so we have a Pre Proc TCL script that checks the message for “Base64”.

If not found, we kill the message.          return “{KILL $mh}”

[Jerry Sawa]

Or create a variable called @ampersand, value as “=&”, then plug the variable into the URL.

[Jerry Sawa]

Try using “concat”

first parameter will be everything in the URL up to “/mdweb”

second parameter “=&”

third parameter will be everything after”PatientID”

 

• This reply was modified 5 months, 1 week ago by Jerry Sawa.

[Jerry Sawa]

Attached document is kind of old, I’ve modified some settings to match what I have in my previous post.

Attachments:

You must be logged in to view attached files.

[Jerry Sawa]

I handle that issue with an “and” alert.

I see if thread has

• Connection is “up” for over 5 minutes
• and O/B Queue depth >=1 for over 5 minutes
• and Last Receive is >= 300 seconds (5 minutes)

If all those are true, I prestart the thread.   Resolves the issue 99% of the time.

[Jerry Sawa]

They did.  Sorry, I neglected to include.

The mode can be ClientAuth,

SSl Protocol — TLS 1.3.   

Cipher list — TLS_AES_128_GCM_SHA256.

 

[Jerry Sawa]

I would put the URL in the “Host:” field?  See “PDF TCPIP Properties” attachment.

Not sure what to enter in the fields in the “CA Path  CA File” attachment.

Attachments:

You must be logged in to view attached files.

[Jerry Sawa]

Following…..

Infor – Is there any documentation?

[Jerry Sawa]

Not sure what type of logging configurations you have setup, but you can try Infor’s “enable_all“.  Keep in mind that this will add A LOT of entries into your log.  I only use “enable_all” for trouble shooting issues then I remove it when the issue is resolved.  There are other configurations but I don’t know if they’ll give you what you need.

Within the thread on the “Properties” tab, “Engine Log Configuration“.  Click list to get a listing of all Engine Log Configurations available.

[Jerry Sawa]

I’ve receive base64 encoded PDF msgs larger than 10mb without issue.  So your issue may reflect back to what Jeff Dinsmore previously stated.   Unfortunately, I don’t know how to increase the size of the max files configuration.

Are there any other entries in your log other than the Broken Pipe entry?  You may need to increase the logging level of the thread to see if more entries are listed regarding the error that may guide you to the cause/resolution.

[Jerry Sawa]

From my experience with Broken Pipe, thread was unable to write because the connection dropped.  Make sure vendor side has the same settings in Prod as it does in Test.

[Author]

Replies