Aishagain

@bigmoe Whitfield and @Gabriele Graves One recent issue I helped resolve with my SL partner involved people appearing "in bits" and that turned out to be because my partner habitually puts the PC into sleep rather than shutting it down at night. Once the PC was shut down and restarted the "avatars in bits" issue went away. (BTW cache clearance made no difference as did a reinstall). Also this issue of avatars remaining clouds for longer seems to be related to the advent of EEP in the viewer, but I have noticed that FS 6.4.13 is MUCH better in that regard than 6.4.12, and I must assume that it is at least in part due to the improvements in the LL code used in the latest FS. While I do use the LL viewer for testing purposes I do not use it much, I cannot abide its UI.

@bigmoe Whitfield ahh right fair enough you know more than I obviously. I assumed that if it got to region capabilities it must be at the LL end but your reference to port blocking by a Firewall makes sense.

If the viewer progress bar sticks at "requesting region capabilities" it is already within the LL network and so you must contact SL support about this. Have you tried creating an Alt account to see if you can log in with that? Also it would help if you posted your system spec, from Help> "about Second Life" which you can access from the viewer start up screen without having to log in.

Ironically the only TP crash I had yesterday was leaving an RC channel region for my Main Server home. 😕

@Profaitchikenz HaikuI have noticed that if I open a group IM window and then look at the online members list, as it populates I see a massive increase in udp data flow, so I guess that answers one of your points, the data is held server-side and not on your PC. This, I suggest, is one of the reasons that the group IM servers are being taxed, having to send out great wodges of data. To redesign group chat so this data is not sent would be a pretty major reworking job and would, I assume, involve changes both server- and client-side.

All very odd, @bigmoe Whitfield but I noted that a number of places were unusually quiet tonight (UK time) so perhaps one of the ISPs is having issues. I did not notice anything on my connection though it seemed that general lag was a bit higher and profiles and similar seemed slow to load. No packet loss and ping was normal.

@M1SSZEEIn all the above I did not see any mention of "whitelisting" of the viewer and its cache folder. I assume you did know about this and have done it If not: https://wiki.firestormviewer.org/antivirus_whitelisting may have some relevance.

@Ardy Lay Indeed the restarts that Linden Datacentre servers needed on a fairly short timescale do not appear to be anything like as necessary on AWS cloud servers. My own region ran quite adequately for 21 days recently and showed none of the memory leakage and time dilation issues that were all to familiar in the past. Indeed the only issue I encountrered on one occasion after a run of about 18 days was a small packet-loss issue (0.1%), which I at first attributed to my own connection but which vanished as soon as I TP'd to another region and returned when I came back. This time dilation stability is one of the best aspects, from a user's viewpoint, of the uplift. The irritation I expressed above was not connected to the need for a restart (as region manager I am able to do that myself at will), but the fact that new code was being rolled and the region had been missed out on two seperate occasions recently. That oversight was quickly remedied by LL engineering and my thanks are due to them for a prompt response.

@Ardy LayRubbish. I sent a ticket in to point out that my region had for the second time in short order been "missed out" of the roll. I also noted that Ll replied to the ticket vewry quickly. Unless You are trolling me that is the end of it.

How so, @Ardy Lay?

To LL's credit the ticket was responded to almost immediately and Woods has been rolled to 556255. I noticed at least one other region had not been rolled and as I waited out the Restart of Woods I got the restart warning there, so looks like loose ends are being tidied up promptly.

Ahh well congratulations LL, for the second week in a fortnight you missed my home, Woods of Heaven, in today's restarts. I am still on 555570! and conventiently the "End of Roll" announcement coincided with the closing of Live Chat. WTG

@arabellajones: see the thread pertaining to this week's Main Server roll...Miller Thor has had an update from Vix Linden of SL Support - the roll will happen but it is delayed.

@Miller Thor Since the GSP announced the start of the roll at 3am SLT, I am fairly sure Maestro needs to adjust his watch! That said, I have seen no signs of the new server version on regions I regularly visit which were on Main Server so your guess, right now is as good as mine.

Here we go again. Grid Status Page tells me that there will be a roll on Main Server tomorrow. But WHAT is to be rolled?? 55168, 556225, or yet another variant that has been given the "fruit" test here on an RC channel? It really is too bad that we in Europe won't be told until the roll us upon us. Yes I know it will come no matter what we are or are not told, but that's not the point. I am getting that mushroom feeling again.

@Karly Kiyori: the issue of the texture cache in SL viewers is one about which there are many "old wives' tales still told. The short-term pain of re-caching txture is well worth it if you have texture issues such as yours or similar ones, in my experience, though precisely WHY the issues occir I still half no real idea. There are many SL issue for which a cache clearance will have absulutely no effect but there are some which seems to work, so if it works use it! One thing that does also occur is I assume you "whitelist" the cache and several other elements of the viewer in your antivirus. You seem reasonably smart so I'd be surprised if you don't but just in case have a look at this: https://wiki.firestormviewer.org/antivirus_whitelisting

@Karly Kiyori Go to: https://www.nvidia.co.uk/Download/index.aspx?lang=en-uk type in you details and it will provide you with the driver download you need. If there IS a problem, to roll back you will need to go to Device Manager and find your graphics adapter in there roll back is one of the options. I am in the UK if you are in another country you may need to navigate to the appropriate page for your locale.

@Karly KiyoriTwo elements of your settings as you give them stand out to me: 1) the draw distance of 272m...that is more than the width of a region, so your viewer will be trying to draw the textures for a portion of all the surrounding 8 regions. Frankly it is too high unless you are sailing. Try 128m or even 64m 2) The Nvidia GPU driver is listed as 441.93...in terms of released drivers that is months if not years old and quite out of date. You might go to the Nvidia site and download and install an updated driver. I my opinion you should chance option 1) before attempting a driver update. Beyond that there are other more knowledgeable folk that frequent this forum that may offer help.

@TruBleuz from the Grid Status Page: Reports of Texture Corruption in the Viewer Investigating - Some residents may be seeing texture corruption in the viewer (rainbow colors everywhere). This is caused by the viewer rollback yesterday. Please clear the viewer's cache (Me> Preferences->Advanced->Clear Cache) to fix the issue. We’re very sorry for this extra colorful side effect! Mar 4, 08:01 PST

@GJ Hennesythere was this on the Grid Status Page: Reports of Texture Corruption in the Viewer Investigating - Some residents may be seeing texture corruption in the viewer (rainbow colors everywhere). This is caused by the viewer rollback yesterday. Please clear the viewer's cache (Me> Preferences->Advanced->Clear Cache) to fix the issue. We’re very sorry for this extra colorful side effect! Mar 4, 08:01 PST Whether this is relevant or not I do not know but it seems there are significant issues with LL's viewer at the moment.

This IS a server problem. Clearing the viewer Cache will do NOTHING. To be clear, clearing the viewer cache WILL trigger a refresh of connections, but in order for the issue to be fixed, the server issue needs to be corrected first. That error message is clearly NOT originating at the client, it is originating at the server. The fact that the problem is observed in multiple viewers, most of which did not change in the last few days absolves the client. That LL broke their own viewer is THEIR problem.

This thread was from three months ago. I just opened a new thread to bring it to the top.

I don't know if this is the correct forum for this, but since group IM is hosted on a Server, I guess this is the right forum? I don't know what is going on but Firestorm Support English and several other group chats have been completely down all day today (Thursday March 4th in the UK). There has been no official comment on this on the Grid Status page, or any of the Blogs. Does anyone know what is going on and when this is likely to be fixed? The issue is not like other times when the chat sent is either not accepted by the server or simply does not appear either to the group or the sender. Today the group simply will not open and on trying to open it one gets an immediate error message. So, LL what is going on?

@Aeirifawn26 see: https://status.secondlifegrid.net/incidents/vqbm8q3tg98r Dan Linden covers it in the post above.

And this roll tomorrow... Scheduled Maintenance Scheduled Rolling Restarts Mar 4, 07:00-10:00 PST We will be performing rolling restarts for approximately 5% of regions on Thursday as part of our ongoing effort to optimize region performance, March 4th, beginning at approximately 7:00 AM PDT. Please refrain from rezzing no copy objects and remember to save all builds. Please check this blog for updates. Posted on Mar 3, 13:03 PST any details? Will it be a random (to us) set of regions or all RC Channel or all Main Server? Any fresh fruit involved? 😁 To me this sounds like all of us need to be on our guard for restarts and forget about any building for the duration.