Server Deployments
As always, please refer to the server deployment thread for updates and news.
- On Tuesday, April 2nd, the SLS (Main) channel was updated to server maintenance package 19#19.03.15.525315, comprising the simulator operating system update and EEP.
- On Wednesday, April 3rd, the three RC channels should be updated with at least one server maintenance package (details TBA at the time of writing). Updates should include:
- The removal of UDP Asset message handling across all RC channels – see below.
- Some changes that might help with some disconnect issues older viewers have been experiencing, and which appears to be connected to some TCP message handling between the simulator and the viewer.
Removal of UDP Asset Message Handling
The removal of UDP message handing sees the removal of server-side support for asset (inventory) messaging via UDP. All maintained viewers should be using HTTP for all asset handling – and this has been the case for a good few years for several asset types, with the last few moving to HTTP in 2017; as such these viewers will not be impacted by this deployment.
However, it does mean that anyone using very old viewers still reliant on UDP messaging for assets will not be able to receive any updates for the following: animations, body parts, clothing, gestures, landmarks, meshes and sounds whilst connected to regions on any RC channels; this includes the two legacy viewers provided by Linden Lab, the Linux Spur viewer and the Obsolete platform viewer.
In particular, and since some specific body parts are required to render avatars, user of these older viewers will see their avatar either as a cloud or the “Ruth” avatar, and will be unable to change from it.
SL Viewer
- The Love Me Render viewer, version 6.1.1.525446, was promoted to de facto release status on Tuesday, April 2nd.
- The EEP RC viewer updated to version 6.2.0.525808 on Monday, April 1st.
The rest of the viewer pipelines remain as follows:
- Release channel cohorts (please see my notes on manually installing RC viewer versions if you wish to install any release candidate(s) yourself):
- Bakes on Mesh RC viewer, version 6.1.1.525409, March 26.
- Estate Access Management (EAM) RC viewer, version 6.2.0.525404, March 25.
- Teranino Maintenance RC viewer version 6.1.1.525401, March 20.
- Project viewers:
- 360 snapshot viewer, version 5.1.6.515934, June 6, 2018 (see my hands-on update).
- Linux Spur viewer, version 5.0.9.329906, dated November 17, 2017 and promoted to release status 29 November – offered pending a Linux version of the Alex Ivy viewer code.
- Obsolete platform viewer, version 3.7.28.300847, May 8, 2015 – provided for users on Windows XP and OS X versions below 10.7.
Teleport Disconnects
Teleport disconnects continue to be a problem. Some are pointing to the EEP deployment as the cause, and also citing the use of derendering the sky (Advanced menu > Rendering Types > unchecking Sky or CTRL-ALT-SHIFT-6). This seems to work for at least some of those experiencing the issue. However, Maxidox and Rider Linden have been looking into this, with Rider commenting:
The sky thing is much the same as putting a banana in your ear to keep the alligators away.
In other words, while derendering the sky my offer a workaround for the problem, it doesn’t actually resolve the issue.
Simon Linden added:
It’s definitely possible that any difference in network traffic might make the system behaviour change – but I don’t think changing the sky settings produces that much network usage.
The failures I’ve investigated seem to be falling into a middle-limbo where the first region thinks it’s handed you off successfully, but the 2nd doesn’t get all the data it expects and eventually times out.
One suggestion is to keep examinations of physical region crossing disconnects separate to TP disconnect, on account of the former tending to be more generally unpredictable due to the added data being transferred (vehicle as well as avatar(s)) and the manner in which the hand-offs are made. A further possible conflation of figures is the ongoing Singularity disconnect issue.
At the moment, the Lab believe that have one statistic that may show the issue, but as it is only recorded once a day, they are waiting to gather further data from it before determining if it is revealing a possible issue / cause.