Whirly Fizzle

Brilliant! Glad you found the problem! I was just writing you an essay when I saw this post haha. I had a look at the logs you sent me. Here is a comparison of the session with a clean cache & the session with a dirty cache. Brief notes - seeing as you already solved the problem. The time from launching the viewer till the login screen appears is disgustingly long for both sessions. The slow launch time wouldn't affect the login timing out though.This indicates something on your system slowing this down dramatically. Now we know what this is From the clean cache session where you managed to login, your FPS was horribly low for the power of your system. Most of the session you were below 5 FPS, average of 2FPS. The viewer was really struggling when it shouldn't have been. The time taken from the viewer sending the login request to receiving the reply is also really slow for both sessions - not really sure why this is - maybe some congestion at LL's end or your own network was under strain. There are other signs in your session that logged in that your network is struggling though. Possibly that app causes network issues too. The time taken to create inventory views from cache should be instant with a clean cache - it still took 10 secs on the clean cache session and a disgusting 61 secs with a dirty cache. Normally a super slow time here indicates a large flat inventory structure but in your case I suspect that app is holding things up. Your total login time, even with a clean cache is 56 secs & this is still rather slow. I don't know what this Wallpaper Engine app is doing but it needs to be burned in the fires of hell. If you want to send me updated logs from a login with a dirty cache now the problem is fixed, I can double check everything looks fine now.

I've sent gifts to Lindens before & they have always been totally cool with it. Usually funny gifts like a mesh bottle of Xanax to the project manager when EEP was released Grumpity sent us all crowns for debugging BUG-139124 - The presence of certain Avatars stops local specular textures from "sticking". Yes really.

As far as I can see, that debug setting doesn't seem to be connected to any code now in Firestorm. I poked the devs to look at this thread though. It's possible that setting is there for OpenSim use.

Just ask Boxy for L$ to help pay your tier & promise to pay him back tomorrow & you'll be instantly connected with a Linden

Just to clarify, you have the same problem with the LL viewer too? Also to clarify, LL support didn't de-flatten your inventory? Can you reproduce a failed login on Firestorm with a dirty cache. Then before relaunching the viewer, zip up the viewer logs folder & either upload the logs.zip to dropbox & give the link (you can PM it to me on the forum if you prefer) or you can email me the logs.zip at whirly.fizzle at phoenixviewer.com. This wiki page tells you where to find the logs folder: https://wiki.firestormviewer.org/file_a_jira#how_to_give_us_your_logs I'll look at the timings for each login stage to see where the problem is.

Do you happen to be using the Vintage skin in Firestorm? This skin actually causes a slower login time then the other skins, especially when you have a large flat inventory - I think it's because of the different layout of the inventory window. If you are using Vintage skin, switch to the Latency skin which looks 99% the same but doesn't have the same problem. If you are fine after switching to a different skin then Vintage, you'll still need your inventory deflattened though because you'll be very close to having the same problem with other Firestorm skins or with other viewers too.

If clearing cache lets you login that session, the problem is most likely to be a large flat inventory structure causing the "Creating in inventory views" part of the login process to take so long the login times out. If this is a flat inventory issue with a certain account, you'll also be able to log your alts in without any problems even with a fully cached inventory. Best thing to do is to contact LL support & ask them to check if your inventory needs de-flattened. LL support can run a script on your inventory to "de-flatten" it, which will fix the problem. If you are a premium member, use Live Chat & they will most likely fix you right away if it just needs the de-flatten script run. If you are a basic member, file a support ticket: https://lindenlab.freshdesk.com/support/solutions/articles/31000131009-contact-support A "flat" inventory is one which has a large number of items in the inventory root folder &/or one or more inventory folders contain a large number of items - where an item is a loose inventory asset or a folder. The more nested your inventory is the better - you can never have too many sub-folders within sub-folders. The de-flatten script LL support run will automatically create sub-folders within any folder that is too flat - ie contains too many items.

Nice breakdown of the problems here: https://blog.cloudflare.com/analysis-of-todays-centurylink-level-3-outage/

This is a LL feature, not something nefarious the person who banned you did. When a banned agent attempts to enter the region, any agents that share the same IP address as the banned account will also be banned for a limited time.

LL will get to it. An hour isn't long to wait....

Bless your heart...

What's the name of the crashing region?

Still makes more sense then the server release notes...

Maintenance-RC viewers have been named after alcoholic drinks for a while now, we've already been through the alphabet once & they are back to A again.

Open Phototools with ALT+P Go to the Light tab. Click the D buttons next to all the shadow & AO settings. Does that fix the crash? Setting shadow res above default is the most usual cause of crashing when taking snapshots with shadows enabled. You need a really powerful graphics card to handle setting shadow resolution at 3 or 4.

Web search blocks the N-word. it returns "Whoops! Your Everything search for "bad word" did not return any matches. Legacy search doesn't block it though & returns some very nasty results under groups.

They already did

Hiya @Mystical Loon Ok, so it's a viewer crash or disconnect. I found some viewer crashes from you in Bugsplat - thanks for enabling crash reporting! There are 5 viewer crashes from you between 06-08-2020 and 07-26-2020. 3 of the crashes were in the 64bit Intel graphics driver ig9icd64. The other 2 were known crashes, Firestorm_Releasex64!`anonymous namespace'::makeMap and Firestorm_Releasex64!LLMeshRepoThread::fetchMeshLOD. However all these crashes appear to be on a different system because it has Intel graphics and the CPU is Intel(R) Core(TM) i5-7200U What happens when Firestorm crashes on the new computer? A) You get a notification saying you have been disconnected from Second Life or that the region is experiencing trouble, with the option to view IM/Quit B) The viewer freezes & never recovers & you have to force quit. C) The viewer closes to desktop with no error message. D) The viewer closes to desktop with an error message from Firestorm - if so, what's the error message. E) The viewer closes & you get an error saying "Firestorm has stopped working" from Windows. F) Something else. Please can you check that you have enabled crash reporting for Firestorm on your new computer. Go to Preferences -> Crash Reports. Make sure that "Send crash reports.." is ticked. Also make sure that "Include Second Life name" is ticked so I can find your crashes in Bugsplat. After changing these settings, relog - if you crash before relogging after changing these settings, the crash won't be sent. Thanks

Oh this will be fun. I love ̶g̶r̶i̶e̶f̶i̶n̶g̶ testing bots. I'm going to quiz Boxy on its knowledge of JIRA

Create a test account. Write an automated action web script thingy to upload the desired image to the test accounts web profile image, which is free. Grab the profile image UUID. The rest is easy.

Objects or mesh or avatars with complex geometry. Or objects with rapid object updates.

Actually... maybe not 💀

Why don't you file this feature request on the Linden Lab JIRA & see what the Lindens have to say about it?

Mazidox just fixed the problem