The following notes were taken from the Tuesday, July 20th, 2021 Simulator User Group (SUG) meeting. The meeting was recorded by Pantera Północy, and the video is embedded at the end of this summary.
Please refer to the server deployment thread for updates.
- There was no planned deployment to servers on the SLS Main channel on Tuesday, July 20th.
- Wednesday, July 21st should see all servers on the RC channels re-started. Due to the manner in which simulator deployments are made, this will show as a updated release number (2021-07-16.561609), although there are no changes to the code itself. However, during the restart process, approximately 1/3 of RC servers (not the simulator code) will see changes designed to improve performance “in some situations”. In addition, the deployment notes further state:
As mentioned in some of our recent deploy plans we’ve been investigating the effectiveness of region restarts at the ten day limit. We believe some of the issues Residents are encountering with regions that have long uptimes are due to issues at the machine level, not the simulator level. Our instance refresh for Second Life RC channels this week is an attempt to gather data as to whether this is correct or not.
The two RC updates had been planned for the week have now been altered as follows:
- The ability for estate owners / managers to set a default EEP environment across an entire estate has been pushed back to a future release.
- Opening custom chat ranges to estate managers through the simulator debug console has been put on hold as a result of “privacy concerns” – primarily that people might not be aware how far their local chat is carrying on regions where the range has been increased over the “usual” 20m. As such, consideration is being given to adding a viewer-side change to inform people of the local chat range.
The Fernet Maintenance RC, version 18.104.22.1681414, dated July 14th, was updated to de facto release status on Monday, July 19th. This leaves the rest of the currently available official viewers as:
- Release channel cohorts:
- Project viewers:
- Legacy Profiles viewer, version 22.214.171.1240519, dated October 26.
- Copy / Paste viewer, version 126.96.36.1993365, dated December 9, 2019.
- Project Muscadine (Animesh follow-on) project viewer, version 188.8.131.522999, dated November 22, 2019.
- 360 Snapshot project viewer, version 184.108.40.2069111, dated July 16, 2019.
Agni / Aditi Account Syncing
This has been an issue for some time, causing pain for many creators who wish to use Aditi (the beta grid) for testing, due to significant issues in trying to log-in. The matter has been under investigation for several months, and a solution new appears close, with Simon Linden commenting:
We’ve tossed together some code and stuff and now have the ability to copy your account from production SL (here) to our beta grid. If any of you would like that, send me an IM. We’ve tested it a bit and are ready to try the first volunteers.
There are a few things to note, however: you should clear your viewer cache before logging in to Aditi; when you log in, your outfit may look messed up. Just change it. We’d love to get some feedback on this.
So, if you are interested in regaining access to Aditi, drop an IM to Simon, stating your reason for requiring access and willingness to participate in testing.
To quote Monty Linden (making a long-awaited return to SUG meetings!):
A refresh to a component of the HTTP-Out system, our HTTP proxy, is underway. Changes are meant to be transparent involving operational improvements, but we’re letting everyone know so if you see issues having to do with reachability of external servers or reliability of HTTP access, please speak up. Jira’s always appropriate for reporting issues but be specific in the details. We’re currently serving about 200K requests per minute via LSL so if you tell me “it happened last Thursday,” I’m not going to find anything. Time, place [region], target – details matter. At some point soon, this will go out to Aditi as a non-simulator deploy. [I] Want to encourage reports of issues if anyone sees problems. Especially blockages of things that should not be blocked.
- The meeting includes a general discussion on teleports. teleport failures, region crossings issues and how they might be related and one (physical region regions via forced unseat / teleport) might help in diagnosing force TP failures – and whether or not this is the case.
- There has been a question on whether texture transfer via UDP through the simulator is still / again possible. Commenting on this, Monty Linden stated:
Texture delivery was to be blocked for UDP but not certain that happened. They won’t enjoy it. Would like a capture of a shift-ctrl-3 screen. texture asset delivery via simulator is one of our most pessimal systems. Will check on the UDP status – that’s curious.
- Refer to the video below for additional information.