There have been ongoing issues with regions not rendering within the World map. The precise reasons why this is the case is currently unclear; Andrew Linden has been trying to look into the matter since it was reported at the last of the Friday Simulator User Group meetings on the 9th November, but work on Interest Lists has kept him busy.
The problems appear to be twofold: tiles for some regions either entirely fail to generate in the World map, or their appearance is linked to the level of zoom being used.
The first issue is demonstrated with the region Sunny Point, which simply does not appear on the World map at all.
With the second issue, Qie Niangao reports that strips of regions on the World map can effectively vanish at certain zoom levels (see JIRA SVC-8115), with some regions of the Zindra adult content apparently never having been drawn at some zoom levels of the Map.
Regions located close to Harshap demonstrate a part of the problem, in that they will appear in the map when zoomed fully in, but step out once on the zoom, and a strip of regions will vanish.
While the issue has only recently come up for discussion at the Simulator UG meeting, the problem appears to have been persistent for a good while, and some have reported the map is missing strips of region tiles from as many as eight different locations.
The World map is generated via a process which images regions from an altitude of around 350m (hence why builds above this altitude do not appear on the map). The information is then scaled for rendering at a number of levels to represent the different zoom levels within the World map floater. Currently, it appears is if the problem lies with the generation of these different zoom levels, at least as far as the “missing strips” issues is concerned. The data from the process is also used for generating map images at maps.secondlife.com, with the result that issues can also occur when viewing map segments there.
As it stands, the possible causes for the problem are still under investigation by LL personnel. However, anyone encountering problems with their region(s) failing to render properly on the World map should consider raising a bug report and /or attending the Tuesday Simulator UG meeting (the Friday meeting is now discontinued).
Inara Pey: Living in a Modemworld 
I wonder if this is connected in any way to the reports from some people of all objects in their regions becoming Phantom?
LikeLike
I doubt it. The mapping process is a abck-end function which I understand is separate to the simulators themselves. As such, it is more likely to be an issue within the map rendering process, which is failing to produce the images which are displayed within the viewer’s map floater or on the SL map web pages.
LikeLike
I’ve been noticing this a bit recently, but didn’t stop to think that it was unusual. Just looked like the usual problems I’d see on the map now and then – but I guess its not the same issues as I’ve noticed before?
(Not that I can recall specifically what those looked like…)
LikeLike
Could be the same – this has apparently been going on for some time, despite the recent reports and the date on the JIRA. As per the article, Qie reports areas of Zindra haven’t appeared on some of the map zoom renders since that continent came into being.
LikeLike
This happens on grids like OSgrid as well. The causes for it there may definitely be entirely different, but the end effect is definitely the same… tiles fail to load at some zoom levels. I had assumed this was OpenSimulator specific behaviour.
The rather kludgy way that OSgrid solves it at the moment, IIRC, is to simply wipe out all the tiles in the map service’s cache on a regular basis (once a night?), and let simulators repopulate it. A similar fix is almost definitely not as simple for LL.
LikeLike
It’s still interesting to know. A leaves one wondering if both systems use the same underlying approach to tile generation in the first place…
LikeLike
About the map, i use it all days and i can assure ALL, that the problem started less then a week ago!
Before that all continents and regions show att diff levels of zooming!.
But worse for me is the fact that tpv using latest LL code, show a bug that was solved already, the fact that moving surfaces like surf waves, now just look static (So no way to surf using any viewer that updated the new code nor latest LL viewers)!
Until a fix is found (seems LL already fixed it, but not released yet) ill not use any Tpv that moved to this latest code release!
LikeLike
You may have noticed the problem in the last week, but it has in fact been around a lot longer. SVC-8115 was raised at the beginning of August 2012 (3 months ago), and people at the meeting, as indicated in my report, have reported the issue as being present for months at least.
LikeLike
I’ve been experiencing this behavior for as long as I’ve been a member (about 2 years). Although in my case, I simply dismissed it as a Linux thing. Good to know I’m not alone after all. I’ll say one thing: it certainly makes for some challenging flights and sails.
LikeLike