Question Error messages on the build port

I have seen odd behavior like this as well. Unfortunately it is hard to reproduce and troubleshoot.

My guess is that somewhere in the code it uses a real number (rnum, every object is loaded on bootup and given a sequential number) where it should use a vnum (the # the builder's use).

A reboot will stop the errors. But we need to try to isolate the error so any help you can provide in reproducing it will help us troubleshoot. How long before the errors start? What trigger type was active before the errors?

The data on this is beyond me to isolate. All I can tell is that my triggers work, they do what they're supposed to with no problems and then one day I'll log in and error messages start hitting my screen. From my notes here, I can see it's happened 7 months, 4 months and 3 months ago. It may have happened other times when I wasn't around to see it. Krell suggested that it might be a time thing so I checked the uptime last time it happened and it had been up 38 days. The one on 4d looked like a similar problem but the port had not been running nearly as long.

I think the reason the reboot works is that it seems that the trigger is attached, not edited. A reboot detaches all attachments.

If I find it happening again, I think I can get more info on it but this is only because I'm learning more about it every time it happens. Unfortunately, I don't have any idea when its going to happen again. It doesn't seem to happen often which is good for the mud but bad for information purposes.

This was funny since I logged, chatted with Rumble a bit and the errors started coming. Rumble wondered, in fact, what I had done to start it up, since he'd been idling. I checked my logs and saw that that I had input this:

wiznet afk for maybe 5 to 10 but i'll still hear it all off the screen
wiznet ooh, there it goes!

I have no time stamps on this and forgot to check history but there were
approximately 10 minutes between so I was basically idling also. The only things that wouldn't have shown on my command log were numpad commands like 'look', 'exit' and basic directions. I was not in Valen Park at the time.

I adjusted the triggers in a hit or miss sort of fashion and got some data.

These are two examples:
5729 The Dark Tunnel is actor, 56068 Wandering Through the Gardens is self.
5060 The Main Tent is actor, 56068 Wandering Through the Gardens is self.

The actors are rooms in different zones, self is a room containing the trigger. The rooms with triggers are normal. I could see from the echos that the various park mobs were wandering around as they should. Unfortunately, this output spammed my screen so I didn't get as much data as I would have liked, but I think what I got was sufficient.

It proved that, according to my trigger, various rooms were wandering through my rooms.
It was not, as I had previously thought, a matter of triggers becoming attached in odd places.
The rooms were from various zones but each set of errors would be different rooms in one zone. The next set of errors would be different rooms in another zone.

The zones I noticed where 15: Straight Path, 50: The Great Eastern Desert, 54: New Thalos and probably either 36 or 305, depending on whether the room name was A <black or white> square or The <black or white> square. These zones are all grid (donated and stock, found in the downloads), which was interesting since there are many more non-grid zones than grid zones. However, there may have been non-grid in the ones I missed.

Other errors were:
[ Trigger: (2708) makes shackles loosen and disintegrate, VNum 2708, type: 0. unknown room field: 'eq' ]
from Detta's Memlin Caverns and Dollhouse, also by Detta. I don't know if this means anything since I didn't notice if any of the error messages were from zones that weren't Detta's or mine.

The mud had been up 4 days so length of time doesn't seem to be a factor. I would also say that it probably isn't me since I've been on at various times without this happening.

I now have the syntax I need to pull that information so I should be able to get more data if I see it happening again. I also have to learn to turn on log

I am guessing the errors are from a misuse of vnum and rnum somewhere in the code. Triggers that are error free are then being improperly attached somehow. This causes room triggers to error when they try to use mob or obj variables and vice versa. The problem is figuring out where the bug is in the code.

By code you mean not me, right? Because it's kinda odd that it never seemed to be a problem before I put the trigger in. And when it hits, it spreads to other triggers in other zones. I thought it was the attachments but they seemed fine. The code seems to actually believe that various rooms are wandering around and doing things. The one I caught from Memlin Caverns is a mob greet-all. This suggests that the mob (Keeper of the Temple in room 2743) saw a room go through and tried to 'eval rwrist %actor.eq(rwrist)%' but couldn't since rooms don't wear equipment. I didn't play with the trigger, of course, so I don't know which room was apparently walking through.

There is a possibility that it's happened before but because my triggers fire so often, it happens more with them. Before that, it may have been more like saying, 'That's odd." every once in a while and rebooting the computer.

Correct, it is a code issue not a builder issue. I've seen it happen across several different zones. Like you said when I couldn't figure the issue out a copyover solves the issue.