Linux OS update to servers a cause of SL TP issues?

As we’re all (probably painfully) aware, the last few months have seen Second Life plague by region crossing issues, with users frequently disconnected (with teleports – being the most common form of region crossing – in particular being affected). One of the pains in dealing with these issues has been identifying the root cause – with most thinking being around it being a timing issue with communications between the region receiving and incoming avatar and the user’s viewer.

However, speaking at the Content Creation User Group meeting on Thursday, April 18th, Vir Linden indicated that the problem might be related to the server Linux operating system update the Lab recently rolled out.

That update – was initially deployed to a small cluster of regions on a release candidate channel called Cake, and it has been reported by those using Cake regions for testing in April, that it was those regions that first demonstrated the teleport issues – although at the time, they were thought to be local connection issues, rather than indicative of a deeper potential issue.

Commenting on the situation at the CCUG meeting, Vir said:

We’ve been having some issues on the simulator side where people tend to get disconnected during teleports … it’s been common enough that shows up as a significant blip on our stats … and that issue seems to have come along … basically when we upgraded the version of Linux that we’re using on our simulators. so we’ve had to do some roll-backs there, just to try to get that issue to go away.

[But] that pushes out the time-line for [deploying] all the things that are based on … the later version [of Linux] that we’re trying to update to … Hopefully we can get those out soon, but I can’t tell you anything about the time-line.

This might explain the scheduled maintenance witnessed on April 18th, with large number of regions going off-line and restarted. If this is the reason, whether it does see a reduction in the teleport issues with those regions rolled-back remains to be seen. But if data does indicate the region crossing issues have been reduced, then this can only be good news and potentially worth the disruption of the maintenance and restarts.

In the meantime, the audio of Vir’s comments is provided below.

 

4 thoughts on “Linux OS update to servers a cause of SL TP issues?

  1. OMG Inara! You must be more careful with your headlines, really.:/ I almost threw up when I read it, thinking I must delete Linux and crawl back to Windows. 😮 And that must never happen! Fortunately you meant only the servers, so my life is good again.
    Although, no, not really. 😮 Linux powers approximately 163% of all interwerbz servers, and Debian is a legend for it’s stability. If we can’t even trust our own Linux servers anymore, who will we trust with anything?

    Like

    1. Fair point… I should have clarified with a “server” is the title. Reworded now to prevent you having further heart attacks 😉

      Liked by 1 person

  2. Hmm. This is a good idea and using actual region-to-region teleports as the test medium is a practical way of finding out. It will take some time for the results to show us a “yes-or-no” answer response, but it is a step in the right direction. One might also note that Windows has also released their new buggy OP update for the Windows 10 system…that might also be a cause. I guess the only way to be sure is to watch the numbers and maybe ask those with Apple OS systems to comment on their teleport problems. There is an obvious cause but so far it isn’t obvious enough name and deal with. At least we can easily get back inworld when our teleport takes us back to our desktop. There is that small saving.

    Like

    1. I’m Win 7, and I’m routinely plagued by disconnects. The jury (for me) is still out on whether the roll-back has eased the issue or not; I’m spent most of my time in-world at Fantasy Faire & parked giving assistance to another user. Given the volume of traffic into and out of the FF regions, it’s hard to tell if a disconnect is related to the current issues or simply a time-out as the region I’m trying to get to / get out of is already getting hammered with assorted user requests.

      Like

Comments are closed.