[Michael Vork]

Hi there,

Pretty devious this upload mechanism but it works.

Thanks for the help.

Regards,

Micha

[Michael Vork]

Hi Bob,

Thanks for this. I tested it and it works fine. Some elaboration in the online help would not hurt, I suppose.

Greetings ,

Micha

[Michael Vork]

Responders,

The problem I described occured earlier in the proces.

I prepared the record that I want to insert  as follows:

set rec

and inserted it as follows:

INSERT INTO MSGBUFFER VALUES (‘$MSGID’, ‘$MSG’)

The “set rec

” was the problem.

I changed this to:

set rec [concat ‘$MSGID’, ‘$MSG’]

followed by

INSERT INTO MSGBUFFER VALUES ($rec)

This results in the total MSG saved in the database.

[Michael Vork]

Hello responders,

Yes, I tried blob as well with the same result.

If I echo the $MSG just before the sql statement is executed, it contains the full message. It appears that during the insert action the message it truncated on the r. I also tried to embed the whole message between double quotes, but that did nog work either.

@Ian, could you publish the part of your tcl source in which the insert is executed?  I must be overlooking something.

Thanks in response.

Micha

[Michael Vork]

David,

You might ask your vendor was the point is of supporting HL7 only partially? All or nothing would be better.

Greetings,

Micha

[Michael Vork]

Hi Jennifer,

Did you do a hcidbinit -CAf and drop the monitorShmemFile object within the exec directory?

Then you can be sure that your internal databases are completely reset.

Greetings,

Micha

[Michael Vork]

Good solution, Bill.

[Michael Vork]

Weekly is fine for us, with a prefab script which is scheduled to run on windows every sunday. Works fine.

[Michael Vork]

Hi Kevin,

This depends heavily om the logfiles and the smat-files you want to keep.

Loglevel enable_all will consume lots of space.

Moreover, you should do a hcidbinit after large volumes of messages are stored it the recovery and error database. These db-objects will not shrink automatically.

So no clear answer.

Greetings,

Micha

[Michael Vork]

Charlie,

Thanks again. I think that will do the trick.

Clean and simple.

Greetings,

Micha

[Michael Vork]

Hello Charlie,

Thanks for the quick response.

No doubt sqlite will be my friend. I will try to appreciate the significance of your remark. However, if you would like to elaborate on your original response, that will possibly be a great help.

GRTX,

Micha

[Michael Vork]

Levy,

I tried them all, including the UPOC alternative you suggested.

It is simply is not going to happen.

Thanx for your support.

Greetings,

Micha

[Michael Vork]

Levy and Frank,

I tried both your suggestions. The full network path, as suggested by Frank, does not help.

I tried the suggestions of Levy and have to conclude that there is not an autorization problem.  

The strange thing that is actually happening is that, on saving the netconfig, there is not a error displayed.

However, once the engine is started in which the thread is defined, the processlog indicates the following:

[nci :nci :ERR /0:    FIN_1_cmd:12/29/2011 09:57:37] Check NetConfig error – error list:

FACT_VILA_IN Protocol Fileset/local-tps: OB directory ‘X:’ is not a directory.

Apparently the check on validity of the netconfig, upon saving, is different from the check which is performed upon starting the engine.

I’ll give it a try by using a UPOC.

Thanks for thinking with me.

Grtz,

Micha

[Michael Vork]

Thanks for your responses.

Micha

[Author]

Replies