Log in Server Issues....again??

[Ayesha Askham]

OK it seems to be time for the Log in server to fall over.  My viewer sits at "Logging In..." and never moves.  Not again?

 

ETA: Must've been the login server having a blonde moment, I managed OK about 10 mins later!  Funny, it's happened a fair number of times lately and yet once I'm in I have no issues.

:smileyfrustrated:

I do know that LL moved over one of the login servers from Agni to Aditi to improve logins there....I wonder if it's connected?

 

[Whirly Fizzle]

Yeah, I've been seeing this problem recently too and so have some friends. Its very sporadic but it happens regularly enough to be annoying :smileyfrustrated:

The login screen just remains stuck forever showing "Logging in....".

When I have had this happen, it appears to be account specific. I can hop on an alt and login with no problems at all but trying to log the affected avatar in on either Firestorm or the official Linden lab viewer or even the old Phoenix viewer or SL 1.23 viewer fails in the same way.

The last time I had this happen I purged my cache and settings folders for both the official viewer and Firestorm, which did not help at all, though I am fairly sure this is a prolem at LL's end, it was worth a try.

Each time its happened, its only lasted a couple of hours max.

I suspect something is going wrong with the login server for certain accounts when this happens.

If I look at my logs when its stuck at "Logging in....", the logs just sit there stuck at this line....

INFO: newview/llxmlrpclistener.cpp(327) : Poller::Poller: login_to_simulator request sent to https://login.agni.lindenlab.com/cgi-bin/login.cgi

If I run the viewer through a debugger and attempt to login the affected account, wait for it to get stuck at logging in and then break to get a stack, I get...

0:000> kV
ntdll!KiFastSystemCallRet
kernel32!Sleep+0xf
Firestorm_Beta!ms_sleep+0xd (FPO: [Non-Fpo]) (CONV: cdecl) [c:\fs\build_fs_windows_firestorm_4_5_1\build\indra\llcommon\lltimer.cpp @ 80]
Firestorm_Beta!LLAppViewer::mainLoop+0xf38 (FPO: [Non-Fpo]) (CONV: thiscall) [c:\fs\build_fs_windows_firestorm_4_5_1\build\indra\newview\llappviewer.cpp @ 1651]
Firestorm_Beta!WinMain+0x1f2 (FPO: [Non-Fpo]) (CONV: stdcall) [c:\fs\build_fs_windows_firestorm_4_5_1\build\indra\newview\llappviewerwin32.cpp @ 304]
Firestorm_Beta!__tmainCRTStartup+0x150 (FPO: [Non-Fpo]) (CONV: cdecl) [f:\dd\vctools\crt_bld\self_x86\crt\src\crtexe.c @ 547]
kernel32!BaseThreadInitThunk+0x12
ntdll!RtlInitializeExceptionChain+0x63
ntdll!RtlInitializeExceptionChain+0x36

[Whirly Fizzle]

Also to note, each time this has happened to me, there has not been any scheduled/unscheduled maintenance ongoing on the grid status page. I made sure to check back a few hours after I could login again and there were still no posts about maintenance at those times.

[Ayesha Askham]

Hi Whirly

Yes, it is irritating to have this happen then a few minutes later sail into SL without a hitch.  To be honest I feel better knowing that my experience is not isolated!

The Log In server problem seems to have started relatively recently, but no-one at The Lab will admit to there being any problem!

I do know that one of the servers was swapped from Agni Log ins to Aditi ones to reduce the apparent bottleneck there so whether that has any relevance I don't know.  ETA: I do like to repeat myself myself myself.....wibble.

As an aside I see that the Mumbai call girls have flooded the site again.

"Nurse...the screens!"