Ayesha Askham

Hmm 9:40pm BST, so 1:40pm PST...where did the post about today's RC Rolls go on the GSP? Has that Ninja editor been active again or has something gone FUBAR....again? :smileyfrustrated:

@Amethyst and Theresa 1) I am well aware of LL ToS. It does not mean that I think they are correct. I only agreed to it in order to access SL. I am far from alone in doing that. 2) Theresa as a (former) member of the (apparently) oppressed Service Sector, I dispute your idiotic statement, no-one is talking about forced labour, though it DOES occur on both sides of the Atlantic and in many other so-called enlightened cultures. 3) Arguing with such entrenched and blinkered thinking was, on my part, a foolish waste of words. As I said before, I wish all who celebrate it in the USA a happy Independence Day.

You may wonder what this has to do with the SecondLife Server thread....read on... Once again services that have a direct bearing on server and simulator support are to be closed due to a US holiday. When are the United States going to realise that they are not "The World"? A global service needs and deserves Global Support, yet once again Linden Lab withdraw many of its services from the Rest of the World due to a local holiday. Many US users of SL will see nothing wrong with this, I humbly suggest that the Rest of the World may think otherwise. No, of course families should be together for what is a very important celebration of US Independence, but to suggest that there would be no-one available to staff Concierge Support is laughably naíve. "World Series" type US arrogance has no place in a Global community. To those who celebrate it I wish you all a Happy Independence Day.

I'm not too sure about the value of this post to many, but irrespective of that this is the wrong forum to post it. The SL Viewer forum would be better.

Yes...well I have my answer, I guess. "UPDATE: Due to a back-end issue discovered in the RC servers after they were rolled out, we are reverting to the previous version. We apologize for any inconvenience." What a pity that the GSP just resorted to pulling the post rather than actually reflecting what was going on. Now the post referrs to it being resolved, which is at best a half-truth.

Caleb I think the Grid Status Page has thrown a shoe. The RC channel restart post appears to have vanished, and the post bookmark throws up an Error 404, ie the post has actually vanished. Any idea why, other than general SL flakiness?

Caleb I don't know if this has anything to do with the latest Main Server roll or not, but on my home sim, Woods of Heaven, currently sim 9053, I see our boat on the mini- and world-maps, despite it having been taken over 24 hours ago! Now I know the maps tend to lag a bit but 24 hours? Surely Not? So is this symptomatic of a problem or is this "just one of those SL things"? :smileyfrustrated:

Wolf, To suggest that LL do not have any control over the Google calendar instance on the Grid Status Page is naive in the extreme. That it might be uncooperative is quite credible but the LL staff are savvy enough to overcome that, were they sufficiently motivated to do it.

And yet again, as if by magic the Google calendar is updated correctly and reflects what the GSP and Caleb tells us. Thing is, that change occurred during my UK night time...and I do sleep, so please, LL, if you are going to put out information about the weekly Maintenance Rolls, make it consistent! This "now you see it, now you don't" routine is tiresome and counter-productive.

Erm...it's me again, Caleb This week the Google calendar still shows no roll for Main Server but a roll for the RC channels, precisely the opposite of what You have posted. What is going on...are they keeping You in the loop? I know in the scheme of things this is not a vital thing but it plays havoc (yes havoc not Havok) with any plans I might have had.

Well, Caleb, Google Calendar, he say no RC Channel rolls this week and none for Main Server next week. This true or does Google calendar speak with forked tongue? Any chance that Linden Lab's various sections could all sing from the same hymn sheet? I guess your post was made prior to a decision to pull the RC code for this week...and it was the code to bring in the new Height adjustment too! (z offset) ETA 12:45am BST: and now it is back on the Google Calendar. Well well!

Interesting...all these sage responses..not one of them has a clue. "[RESOLVED] Unscheduled Inventory Database Maintenance Posted by Status Desk on April 7th, 2015 at 10:38 am PDT [Resolved 10:40 AM PDT, 7 April 2015] The unscheduled inventory database maintenance is now complete. [Posted 9:59 AM PDT, 7 April 2015] We are currently performing unscheduled inventory database maintenance. During this time, some residents may experience login issues, inventory loading issues and an inability to rez. Residents inworld may also be logged off. Please refrain from rezzing no-copy objects and making L$ transactions during this maintenance. Please check back here for updates." When the OP made his post it is probable that the tech teams were just beginning to be aware that the Inventory Server was falling over...again. I hope your issues are now resolved, Farthington. Oh, and Theresa, for goodness sake stop parrotting the advice about cache clearance, it is very rarely wise.

Wow, Caleb Late decision? When I went to bed last night the Main Server roll wasn't event on the Google calendar! Never mind.

Ah...OK, Whirly. But there seems to be so many newly introduced issues that I conflated them. My mistake. However I did not actually say that any specific issue was AIS3 related, but the post sort of implied it, rereading it. So far as I knew many of the inventory issues that are being suffered by FS users and some others ARE related to either AIS3 or the lack of support for it. We are damned if we use it and damned if we don't. Hohum.

This issue or tranche of issues appear to be selectively affecting users in Australia, who are having multiple issues, only some of which can be laid at the door of their ISPs. There are other threads in this forum relating texture issues that seem very much CDN related and much of LL's recent "Scheduled Maintenance" may have ben aimed at improving connections with the CDN. The idea of passing off certain functions to a third party (the Content Delivery Network) sounds fine in principle but it would seem that as with many things the reality is less good than the concept. Many users report improved functions and texture-rezzing with the arrival of CDN but many others find that the converse is true and the advent of CDN has actually degraded performance, while a third group, myself included, are seeing no change at all. But as the OP notes, the problems are not limited to texture transfer and indicate a throttling at another part of the pipeline. This is a matter that needs urgent attention is users in parts of the world other than the USA are to continue to find SL a worthwhile passtime. ETA: Whirly Fizzle's post in the main thread here https://community.secondlife.com/t5/Second-Life-Server/Deploy-for-the-week-of-2015-03-23/td-p/2916641 suggests that many of the current issues facing users are related to the much vaunted but bug-ridden AIS3 code recently introduced by The Lab.

OK Theresa and Qie My information is essentially gleaned from a "reliable source" within The Lab, other than that I'll let your curiosity run wild. Now...despite all the fanfare and hoohaa about the CDN reducing server load, the sim-server still has to communicate with the client, whatever that might be. There is no doubt that increasing client texture memory will help, especially with the "texture thrashing", but that itself is a function of incorrect communication between sim-server and client. If a sim has zero or very little "free time", script run will indeed be hampered, but so also will all the other functions that a sim-server is required to perform every frame. Now whether you like it or not Mainland servers are older and less capable than most newer private region servers, both for full-region and Homestead sims. These problems are, at least in my experience, distinctly more frequent on Mainland than anywhere else and that is circumstantial, but widely observed. It is not that the Mainland servers run badly, if properly loaded, it is simply that when they are stressed they run badly. And Qie, the newer servers DO deliver superior performance, it is just that more regions are crammed on per core than ever so the benefit is largely dissipated.

Qie has it right. The main culprit here is how the server handles the project interesting code actions. It is becoming clear that many (though not all) Mainland servers, which are the oldest in SL, simply cannot handle to workload they are being asked to do. LL are uncomfortably aware of this and are replacing old servers as fast as finance permits but it is a problem for many on Mainland. A simple sim restart might solve the issue but it is quite probably that it will be just as bad. The "Spare Time" reading is quite diagnostic here. If the server has little of no spare time it is not processing information efficiently and texture-related issues WILL ensue. Lots of hi-rez textures make the matter worse.

Wonky links aside, just what does this update DO? There is nothing on The Blog, or in the release notes that gives us any idea just what these improvements to Premium users are! I have noticed nothing. Will someone at The Lab please enlighten us? ETA 18th March: You know I could swear I hear someone singing "La la la, with my fingers in my ears, I can't hear you".

Really? 25 years? SL is only 12 years old! To be brutally frank, very few people are going to have looked at this with any seriousness, and it is VERY unlikely that a Linden of any coat will read it. Certainly no-one is likely to take such a ludicrous headline seriously nor the rant beneath. Wait for SL2.0, or whatever Linden Lab choose to call SL's successor. Maybe your ravings will have some currency there. Oh, and correct your typos.

Deliza On the SecondLife website, log in as you would to post here, then you will see the "Help" tab...use that and go to "Contact Support" and work from there. It sounds like there is a problem with the region you are trying to reach, so you may need to get it restarted. Support should be able to do that for You. One thing to bear in mind is that if You do not already used the Default LL viewer they may ask you to try that first. Don't be put off...insist on a Region restart.

Deliza If You are not able to get onto your own land, try logging in elsewhere. Also if that issue persists, contact LL support, either use Live Chat or raise a trouble ticket over the issue.

Caleb The time would have been about 18:45hr GMT plus or minus 15 mins when I observed these glitches...that is about 12:30 -13:00hr SLT (PST). Two were relatively brief....5-10 seconds and one about 30 seconds' duration. Edit to add more info: Two spells on the sim tonight (UK) at about 2pm and just before 4pm PST, no further instances of the glitch so it must've been transient. Thank Goodness! :smileyhappy:

Caleb Our homestead Woods of Heaven, currently sim 9050, was rolled to 15.01.16.298178 today as expected but in the last hour we have been experiencing massive physics dropouts..we become rooted to the spot, like partial disconnects, for up to 30 seconds. It is happening to multiple users (on this sim) not just me, so I wonder if you are getting other reports of this behaviour?

Erm...Caleb I may be wrong but according to the release thread from last week the project going to Main Server, being 15.01.16.298178, was on Magnum and LeTigre last week, not Bluesteel, which had 15.01.16.298179. So I assume You just got your RCs mixed up, Yes? :smileysurprised: Edited to correct my dates

I suppose the reason the guy in charge of the GSP did not think it was worth mentioning the RC Rolls this week since the repeated "Scheduled Maintenance" covers the "Do Not Rez" warning that they repeatedly trot out. Then again maybe this work is never really finished. I cannot recall a Scheduled RC roll never being mentioned on the GSP before though. To me it sounds like a case of "It wasn't me, I didn't do it...I wasn't even there at the time" :smileyfrustrated: