RLV Relay Testing

This was prompted by two things that happened last week, both related (well, really, both pretty much the same thing).  During my short holiday in the Z&A main store I encountered a problem where, when I logged off and logged back again later, I wasn't properly caught back onto the doormat. The restrictions applied by the mat were being reapplied, but I wasn't getting caught back onto it. More to the point, I was caught back on the first time, but I wasn't subsequent times.

As you'll imagine, this was kind of worrying. Forcing victims back onto RLV items when they log back in is kind of important to much of what I make in Second Life. So I made a mental note to look into this as soon as I had time.

And then, just a day later, I coincidentally got a message from a customer with the exact same problem. The first relog was fine, but subsequent ones weren't. The one thing we both had in common is we were both using the OpenCollar relays in our collars.

So, this weekend, I've been hunting down relays and testing each one in turn. The tests were done on Firestorm 4.3.1.31155, the RLV version being reported as RestrainedLove viewer v2.7.0 (RLVa 1.4.7). In each case I made an effort to get the latest copy of the relay. I tested each one with the same Z&A product (our doormat) and the test went like this:

1. Use the grab facility to grab myself onto the welcome mat.
2. Lock myself on.
3. Log out.
4. Log in again.
5. If grabbed back, wait a minute to check the lock held.
6. Log out.
7. Log in again.
8. If grabbed back, wait a minute to check the lock held.
9. Unlock and hop off.

Here are the results:


I did initially have a problem with the first regrab with Susan's. While I was grabbed back and all restrictions were reapplied, I got a message shortly afterwards that there'd been no response from the device (in this case the welcome mat) and so I was being released. I tested again and never had a problem after that so I'm thinking this was just down to an unfortunate lag spike or something (a couple of avatars did happen to enter the region at about the same time I did).

The next test is for me to try a different item of RLV furniture -- one that isn't of my own design. I've already tested the worst failure shown above with against a Cerdita FS Post and got the same result. If I get the same results it might be time for me to document the findings and report them to the relay makers (it's worth noting that the apparent failure of Satomi's relay seems to be the same as one I've reported before).