Link Message Problem

Good day all.

I have a problem that I did find a work around, but that bothers me because I do not understand the root cause of the problem and I can't find any documentation.

I have a script that can be placed in several prims of a linked set. When linked, the script goes out and polls how may other prims have the script in it, and then polls those prims for data in a list format.

Basically the flow is as such:

New prim with script pulls all other prims and finds out how many in the linked set has the script in it. This works well.

New prim then request that each of the other prims sends their data back in CSV format, and the new prim sends its data to the other prims. All scripted prims build their data list, converts into a CSV string (llList2CSV), and sends that string out via a linked message. This works well and testing shows that the data is exact.

The problem is on the listen end of the linked message. When the message is received, it is intact, but another message is also sent. This message is the word "allowed". For the life of me, I can not figured out where this message is orginating from (the word is not even in the script). It is a seperate message from the CSV message and it arrives with each reporting prim. If I just toss out the "allowed" message I can process the CSV string back into a list and go own with the script.

While tossing it works fine, it doesn't give me a warm and fuzzy feeling. Is the message "allowed" an internal message that I just never seen before?

Most appreciated of any help one may offer on this matter.

Bracken

Don't think it's internal since i have never seen it before. Try searching your script with the search and replace (ctrl+f, disable case sensitive)

Disable all scripts but the listener, then reactivate the scripts one by one until you get the message. That should help you narrow down where it's coming from.

Searching produces no word "allowed"

Disabling and renabling always shows up on the listen.

Here is a snippet that might help.

else if (message == "send data"
{
string tempName = llGetObjectName();
string tempDesc = llGetObjectDesc();
vector tempOffset = (vector)tempDesc;
vector tempLoc = llGetPos()+tempOffset;
list tempData = [tempName, tempLoc];
string tempMsg = llList2CSV(tempData);
llMessageLinked(LINK_SET, 784, tempMsg, "";
}

if (number == 784)
{
if (message != "allowed"
{
list tempList = llCSV2List(message);
teleportData = teleportData + tempList;
}
}
it is the message 784 where the allowed comes through.

Hope this helps

Bracken

if the "allowed" message is separate (as I expect it is) there is a script not associated with your others that is returning it.

if ("allowed" = data_string){
llOwnerSay( "errant script return from prim #" + sender_number );
}

case 1 - "allowed" is returned inside of the sub script returns
a script call is updating their data and adding this

case 2 - "allowed" is returned separately once when the main script requests a data dump
errant script is in reported prim, and responding to the data request from the main script

case 3 - separate returns of "allowed" (1 for each sub script)
if your request was broadcast with LINK_ALL_OTHERS, and it says it's coming from the same prim, the errant script is responding to the sub scripts

if the your return data is targeted with something like LINK_ROOT or a number, and the reported prim number is different, it's not responding to the subscripts

this all presumes the subscripts are coded to send on the same target logic.

case 4 - multiple returns of "allowed" exceeding the number of sub script returns by one
errant script is responding to both the main and subscripts

case 5 - even more returns of "allowed" than other scenarios
errent script is responding to subscripts that are reporting with LINK_SET, or LINK_ALL_OTHERS.

Thanks Void,

There are security scripts in the build also, and I have a feeling that I have triggered one of them. Guess it is time to write my own of those too.

Thanks again,

Bracken