Firestorm

I just installed the Firestorm Release as well. I had a few successful restarts then I started getting the exact same error. I completely uninstalled and did another reinstall had a few successful restarts of the viewer and then the error started up again. I am not sure if it is related to making changes to my preferences or not.

Same issue here. Installed new version and after a few customizations I started getting this same error.

Below are my computer configs:

Firestorm 3.2.2 (24336) Nov 27 2011 17:05:37 (Firestorm-Release)
Release Notes

CPU: Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz (2660.04 MHz)
Memory: 9208 MB
OS Version: Microsoft Windows 7 64-bit Service Pack 1 (Build 7601)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce GTX 260/PCI/SSE2

Windows Graphics Driver Version: 8.15.0011.8627
OpenGL Version: 3.0.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 296
Bandwidth : 1000
LOD factor : 2
Built with MSVC version 1600

Can you please provide the logfile from when you crash?

A log file from a crash this morning has been attached.

I am encountering this issue too, under similar conditions as described in the original report (although a different sim). Some additional information that may help:

The message quoted in the subject and report is one of 2 messages I have seen under these conditions. Sometimes I get this message instead:

ERROR: llimage/llimage.cpp(173) : LLImageBase::allocateData: LLImageBase::allocateData called with bad dimensions: 0x0x1

I have it in both 3.2.2 and 3.2.1.

I think the issue already existed in Firestorm Beta Mesh, however it never gave an error message in that version, it just crashed straight to desktop without warning.

System info:

CPU: Intel(R) Core(TM)2 Duo CPU E6750 @ 2.66GHz (2671.66 MHz)
Memory: 2048 MB
OS Version: Microsoft Windows XP Service Pack 3 (Build 2600)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 8600 GT/PCI/SSE2

Windows Graphics Driver Version: 6.14.0012.7533
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: metaharper (appalachian_sky)
Font Used: Deja Vu (96)
Draw distance : 96
Bandwidth : 2000
LOD factor : 4
Built with MSVC version 1600

With regard to the appearance of this problem, it did not happen in the mesh-beta. I had been running it (mesh-beta) for quite some time and the only crashes I had were "normal" for the sim. I could run for about 45 minutes to an hour before a silent crash. (I never reported it because I had the same problem with all viewers I used). This crash happens within 5-10 minutes when there is heavy activity.

@Richard: Thanks for providing the log file. This apparently seems to be a memory issue where the viewer just ran out of memory:

2011-11-28T15:21:26Z WARNING: llcommon/llthread.cpp(204) : LLThread::start: failed to start thread Curl Multi
2011-11-28T15:21:26Z WARNING: llcommon/llapr.cpp(284) : ll_apr_warn_status: APR: Not enough storage is available to process this command.

WorkingSetSize KB:      1624084

Latter indicates the viewer has reached the usual crash area on 32bit Windows systems. There have been reports that some reports the version with the latest OpenGL fixes from LL might use up even more memory. Could you try version 3.2.1 if you get any improvements with it?

Ive had the "bad memory allocation for the image buffer!" popup myself too & Id go with Ansariels diagnosis, that it happens when im out of RAM.
Ive also seen the "failed to start thread Curl Multi" spam in my logs from the Mesh beta onwards. This seemed to get spammed a lot to logs several mins before I would crash. Id see camera lag & viewer freezes at the same time.
On Mesh Beta I would just silently CTD. On 3.2.1/3.2.2 Ill get the "bad memory allocation for the image buffer!" fatal error then crash on clicking Ok. Its just the method of error reporting thats changed as far as I know. This way is much better! You immediately see what the problem is without having to go & check logs.

Firestorm 4.0.1 (24184) Nov 26 2011 04:23:30 (Firestorm-Fizzlefire)
Release Notes

You are at 131,249.0, 251,722.0, 21.6 in Veil of Shadows located at sim5470.agni.lindenlab.com (216.82.22.113:13004)
Second Life Server 11.11.09.244706
Release Notes

CPU: Intel(R) Core(TM)2 Quad CPU Q8300 @ 2.50GHz (2493.75 MHz)
Memory: 3326 MB
OS Version: Microsoft Windows Vista 32-bit Service Pack 2 (Build 6002)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce GT 230/PCI/SSE2

Windows Graphics Driver Version: 8.17.0012.7533
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox 2.1.3010.6270

Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 256
Bandwidth : 800
LOD factor : 3
Built with MSVC version 1600
Packets Lost: 0/23,141 (0.0%)

I used the mesh beta version from the day it was released and aside from some bugs where the panels forget that they were undocked and the contacts list not displaying names immediately, I can count the number of times the viewer crashed on me on one hand. This new release opens cine the first time then errors immediately each time it is opened after the initial installation. This occurs at my home on the mainland sim of keroo. There is no heavy traffic and the error occurs before i even rez. i have added my log file and i have also reverted back to the mesh-beta release and running without error or crash.

Astrea, is that really a log from when you crash? Looks for me like normal exit:

2011-11-28T09:35:34Z WARNING: newview/lltoastalertpanel.cpp(164) : LLToastAlertPanel::LLToastAlertPanel: Alert: Are you sure you want to quit?
2011-11-28T09:35:36Z INFO: newview/llappviewer.cpp(3547) : LLAppViewer::requestQuit: requestQuit

I've added another log file...what I started doing was not clicking the "ok" button and just let the error window sit in the background as I waited to see if I would completely rez, or if the bridge would eventually create (which it never did). I would then close the browser down, leaving the error message untouched, which I believe allowed the normal shut down. I did click on the error message several times, resulting in an immediate crash/shutdown so I believe I have the correct log attached now. I'm pulling these log files from the folders in my trash because I wiped everything each time I did a re-installation. If this log does not give you the information, let me know and I'll reinstall and capture a fresh log. I did full uninstall (including killing the firestorm folder from app/data)/restart/reinstall several times and was able to recreate the problem each time.

Crash from version 3.2.1 (rje.resident) The dump file is empty. Similar conditions on the crash though delayed a little more. (Scene was not as filled with av's) Another quick note, Task manager shows around 2gb of memory in use (out of 3gb installed and usable.)

It's happened twice to me now. First time FS was able to continue until I clicked OK; second time FS froze.

CPU: Intel(R) Core(TM)2 Extreme CPU Q6800 @ 2.93GHz (2933.39 MHz)
Memory: 3070 MB
OS Version: Microsoft Windows XP Service Pack 3 (Build 2600)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 8800 GTX/PCI/SSE2

Windows Graphics Driver Version: 6.14.0012.5721
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Hybrid
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 80
Bandwidth : 1000
LOD factor : 4
Built with MSVC version 1600
Packets Lost: 0/17,459 (0.0%)

Can everyone getting this issue try the following please & let us know the results. Also give your full system information from Help -> About Firestorm

• Reinstall Firestorm as a clean install following the instructions here http://wiki.phoenixviewer.com/doku.php?id=fs_clean_reinstall

• If this doesnt work, then from the login screen, go to Viewer -> Preferences -> Graphics -> Rendering -> Use HTTP Textures -> untick -> Apply/Ok -> Close login screen & relog

• If this doesnt work, then from the login screen, make sure Debug menu is active with CTRL+ALT+D
  Then Debug -> Show debug settings -> UseHTTPInventory -> Set to FALSE -> close login screen -> relog

Use HTTP Textures - on or off? I did do a clean install, by the way; will try the other options tomorrow.

Yesterday I did clean installs as well as some restarts after uninstalling and killing the Firestorm directory out of AppData\Roaming. Today, I did a clean install, following the instructions (which outlined the same procedure I did yesterday, and always do when I reinstall). The only difference from yesterday to today is that I went into my SecondLife folder in AppData\Roaming (where I keep some old chat log files) and killed the user_setting and browser_profile and log folders in that directory. These were old files and did not have any current activity or mod dates. But, I killed them anyway and simply left the plain chat logs. I did a restart, then reinstalled.

I looked at the Viewer -> Preferences -> Graphics -> Rendering -> and the "Use HTTP Textures" is already selected. However, I did uncheck and check them again to apply.

So far, I have silent crashed once (something that does occur right after a fresh install until things settle a bit) but I have not received the error message.

I have since started applying all of my personal preferences and restarted the viewer a number of times and so far I have not received the error.

At this time, my debug settings for "UseHTTPInventory" is still set at the "TRUE" default. Because I have not received the error, I have not done that step to set it to "FALSE".

Even though I have not received the error, I am attaching the log file because I did have the one silent crash. This may help to identify if that crash was related at all to this issue.

Whirly....what is the purpose of that UseHTTPInventory setting? Is there a reason why it is beneficial to have it set at the TRUE default? Will it benefit me at all to set it to FALSE? Will I notice any improvement in performance or lose any features by doing so?

Thank you all for your help. Keeping my fingers crossed....

OOps sorry, I missed part of the instructions out. Edited them now. Its HTTP Textures off we wanted to try.
UseHTTPInventory is best left on TRUE, you should get faster inventory load with it enabled.
The only reason we wanted to see of those two HTTP settings changes made any difference was because it was reported in the SL forums by someone having this issue that it fixed them.

@Whirly and whoever else.. lol.

–
If this doesnt work, then from the login screen, go to Viewer -> Preferences -> Graphics -> Rendering -> Use HTTP Textures -> untick -> Apply/Ok -> Close login screen & relog
–

So far so good. There are 3 other major issues I'm working through. In the past I've had little problems with FS. Now it's becoming a nightmare.

Disregard my last message. So far so good didn't work out so well.. Trying this now -

–
If this doesnt work, then from the login screen, make sure Debug menu is active with CTRL+ALT+D
Then Debug -> Show debug settings -> UseHTTPInventory -> Set to FALSE -> close login screen -> relog
–

I recall that having HTTP Textures off in the past resulted in very slow rezzing, At least it did for me in Phoenix. However, at the moment I seem to be running fine with both HTTP Textures selected and UseHTTPInventory as True. So my only change from last night was deleting those folders from the SecondLife Folder I still had sitting in my AppData\Roaming directory. Curiouser and Curiouser.

Clean install, HTTP textures and inventory disabled. Lots of "Bad CACHED TEXTURE size:" messages in log. In last half minute there were "Warning - allocate image data" messages.

Clean install, works fine with alts that have smaller inventory. Main account has inventory over 165k items. Runs fine until the inventory finishes loading or is almost finished loading, then I get this error every time:

ERROR: llimage/llimage.cpp(244) : LLImageBase::getData: Bad memory allocation for the image buffer!

Had a very similar issue with the beta. Dump file and error marker are empty. Please let me know if I can supply any additional information that might be useful in resolving this problem. Thanks!

Using Develop->Show Info->Show Memory the crash occurs when the memory value is 1670000 to 1699000 KB used. Yes, inventory quantity does affect this. I wonder about aging of objects as the memory usage does not appear to return to a lower value when leaving. I started out in one location (home) at 766000KB went to another site and got to 869000KB and then returned home and was at 1006200KB. I lowered draw distance from 128m to 40m and tp'ed out and back in. Memory went from 1006200 to 951000.

Same issue and error as listed in the lead post. Tried both current versions of FS listed on DL, clean install. clean cache. Tried all fixes listed plus a few more things I disabled. RAM tests as fine and is brand new anyway. PC passes all performance tests and hardware tests I know of. Info as follows:

CPU: Intel(R) Pentium(R) D CPU 2.80GHz (2793.17 MHz)
Memory: 2047 MB
OS Version: Microsoft Windows XP Service Pack 3 (Build 2600) compatibility mode. real ver: 6.0 (Build 2900)
Graphics Card Vendor: ATI Technologies Inc.
Graphics Card: ATI Radeon HD 4300/4500 Series
Windows Graphics Driver Version: 8.892-110914m-125026C-ATI
OpenGL Version: 3.3.11079 Compatibility Profile Context

libcurl Version: libcurl/7.20.1 OpenSSL/0.9.8j zlib/1.2.3
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1

@Whirly

It should be NOTED that if you uncheck the "Use HTTP Textures" you should clean your "CASHE" folder manully before you relog - different texture code base.

Firestorm 3.2.2 (24336) Nov 27 2011 17:05:37 (Firestorm-Release)
Release Notes

CPU: AMD Phenom(tm) II X4 955 Processor (3200.06 MHz)
Memory: 8192 MB
OS Version: Microsoft Windows 7 64-bit Service Pack 1 (Build 7601)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce GTX 460/PCI/SSE2

Windows Graphics Driver Version: 8.17.0012.9036
OpenGL Version: 4.2.0
RestrainedLove API: RLV v2.7.0 / RLVa v1.4.3a
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 200
Bandwidth : 1500
LOD factor : 4
Built with MSVC version 1600

I'm having the same issue...it started on the release version and affected the beta one too, which didn't have this kind of crash and not so frequently anyway.

Firestorm 3.2.1 (24179) Nov 27 2011 15:25:09 (Firestorm-Release)
Release Notes

You are at 113,495.0, 313,547.0, 35.0 in Blessed Land located at sim9494.agni.lindenlab.com (216.82.43.197:13004)
Second Life RC KT 11.11.09.244706
Release Notes

CPU: Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz (2400.02 MHz)
Memory: 3583 MB
OS Version: Microsoft Windows XP Service Pack 3 (Build 2600)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 9800 GT/PCI/SSE2

Windows Graphics Driver Version: 6.14.0012.7533
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 1000
LOD factor : 2
Built with MSVC version 1600
Packets Lost: 222/25,561 (0.9%)

I have the same issue as all the others here. I first tried Firestorm 3.2.2 with a clean install and I crashed within 15 minutes with the error ERROR: llimage/llimage.cpp(244) LLImageBase::getData: Bad memory allocation for the image buffer!.
Relogged and crashed again 3-4 more times.

Miro suggested trying 3.2.1, again on a clean install. And again I crashed.
Tried it twice more, and each time I crashed again with the error:ERROR: llimage/llimage.cpp(244) LLImageBase::getData: Bad memory allocation for the image buffer!

Editing this to add:
Yesterday I was on FS 3.2.1 with an alt all day shopping in busy areas, and had not 1 crash all day. I have since logged in several other alts and tried zipping round various busy shopping areas, and none of them crashed either.
Logged Helli in, and within minutes on my own land, I crashed with that error.
So it looks like my issue is a Helli only (avatar specific) issue.

the same error is happening to me mostly in crowdy place and when I cam around. Funny that as the error pops up, I don't crash as long as I don't click ok on that popup...

I did the clean install as it is said, but no success regarding the crash "llimage/llimage.cpp (244) LLImageBase::getData: "bad memory allocation for the image buffer!"

----------------

Firestorm 3.2.2 (24336) Nov 27 2011 17:05:37 (Firestorm-Release)
Release Notes

CPU: Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz (2400.04 MHz)
Memory: 2047 MB
OS Version: Microsoft Windows XP Service Pack 3 (Build 2600)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 9800 GT/PCI/SSE2

Windows Graphics Driver Version: 6.14.0012.7533
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: V3
Viewer Skin: starlight (original_orange)
Font Used: Deja Vu (96)
Draw distance : 96
Bandwidth : 1000
LOD factor : 2
Built with MSVC version 1600

I get the same error on a Mac as well.

Firestorm 3.2.2 (24336) Nov 27 2011 16:12:10 (Firestorm-Release)
Release Notes

You are at 187276.0, 263937.0, 3314.9 in The Renaissance Galleria located at sim20938.agni.lindenlab.com (216.82.29.217:13000)
Second Life RC KT 11.11.09.244706
Release Notes

CPU: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz (3400 MHz)
Memory: 16384 MB
OS Version: Mac OS X 10.7.2 Darwin 11.2.0 Darwin Kernel Version 11.2.0: Tue Aug 9 20:54:00 PDT 2011; root:xnu-1699.24.8~1/RELEASE_X86_64 x86_64
Graphics Card Vendor: ATI Technologies Inc.
Graphics Card: ATI Radeon HD 6970M OpenGL Engine

OpenGL Version: 2.1 ATI-7.14.5
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Hybrid
Viewer Skin: metaharper (modern)
Font Used: Deja Vu (96)
Draw distance : 192
Bandwidth : 6100
LOD factor : 4
Built with GCC version 40201
Packets Lost: 22/84032 (0.0%)

My memory is topping out too.

I've always manually deleted FS's data between installs(windows too local & roaming).
_
Firestorm runs a helluva lot smoother on linux though, at least until i run out of memory now
_
I was about ready to go back to a 64bit version so it see my other half of ram.. until i saw Ona Waffle with the same issue and twice the amount of memory.
_
Disabling http textures & inventory did seem to help (i just did them both at the same time and Kranked everything up draw distance/depth-of-field n stuf.
_
i did clear cache (through the viewer) before relogging after disabling those options.

Firestorm 3.2.2 (24336) Nov 27 2011 15:24:57 (Firestorm-Release)
Release Notes

You are at 463063.0, 302803.0, 63.7 in Glumdale located at sim9693.agni.lindenlab.com (216.82.45.27:13001)
Second Life RC LeTigre 11.11.30.245889
Error fetching server release notes URL.

CPU: AMD Phenom(tm) II X6 1090T Processor (3200 MHz)
Memory: 3274 MB
OS Version: Linux 2.6.38-9-generic #43-Ubuntu SMP Thu Apr 28 15:25:15 UTC 2011 i686
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce GTX 275/PCI/SSE2/3DNOW!

OpenGL Version: 3.3.0 NVIDIA 270.41.06
RestrainedLove API: RLV v2.7.0 / RLVa v1.4.3a
libcurl Version: libcurl/7.21.1 OpenSSL/1.0.0d zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: OpenAL, version 1.1 ALSOFT 1.11.753 / OpenAL Community / OpenAL Soft: PulseAudio Software
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: V3
Viewer Skin: starlight (original_orange)
Font Used: Deja Vu (96)
Draw distance : 192
Bandwidth : 500
LOD factor : 3
Built with GCC version 40405
Packets Lost: 50/8508 (0.6%)

NOTE: the specs above were copied after i logged back in (to copy them for this). My draw distance was maxed during the "stress test"

The ERROR that i normally would get is the same as this Jira's, though this time it was different "ERROR: llrender/llvertexbuffer.cpp(1441) : mapVertexBuffer: memory allocation for vertex data failed."

***i saved the log as "Alisa menna's firestorm log" i'll try to put it in here.
GOOD LUCK ! !

Update (with 2nd Firestorm log)

Logged back in.. Still in the same sim.. still have the viewer's http textures unchecked and "usehttpinventory..whatever" set false. I stayed in one location without moving much besides double click TPing around the castle some. Chatted for an hour at least.

Though without lights&shadows, depth-of-field, and draw distance was set reasonably for indoors @96 meters. Graphics set as ultra though

The place wasn't that crowded. There may have been 15 - 20 people that was bold within the "nearby" window.

I was watching my memory system monitor in the panel (gnome)as i was chatting. About three quarters was used and one quarter cached, none free. Then as i walked outside, the amount used started to increase and within three to four minutes (guessing), i crashed.

Experiencing the same problem as those listed above, have tried both versions of Release FS with clean installs. Have tried all fixes shown above as well.
Firestorm 3.2.2 (24336) Nov 27 2011 17:05:37 (Firestorm-Release)
Release Notes

CPU: AMD Athlon(tm) II X2 240 Processor (2815.17 MHz)
Memory: 3328 MB
OS Version: Microsoft Windows 7 32-bit Service Pack 1 (Build 7601)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 9600 GT/PCI/SSE2/3DNOW!

Windows Graphics Driver Version: 8.17.0012.8562
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: metaharper (appalachian_sky)
Font Used: Deja Vu (96)
Draw distance : 96
Bandwidth : 2000
LOD factor : 4
Built with MSVC version 1600

I'm getting this error too. I'm using the Mac version of Firestorm.

IMPORTANT

We need to get more data from people experiencing this issue, especially if you can reproduce this error message right after logging in with a clean cache to rule out it just being an out of memory issue.

• Logout of SL on all avatars.
• From the login screen, go to Viewer -> Preferences -> Network & Cache -> Click open next to cache location.
• Delete the folder named texturecache & close that window.
• Close the login screen, relaunch Firestorm & login.
• Wait for the error llimage/llimage.cpp (244) LLImageBase::getData: "bad memory allocation for the image buffer!" to happen & click Ok on the message box. You will crash.
• Create a new folder on your desktop, you will be copying files into here.
• Browse back to your Firestorm cache folder location. If you dont know where to find this, open the Firestorm login screen again, Viewer -> Preferences -> Network & Cache -> Click open next to cache location.

  ---

Default cache locations are:

• Windows XP: C:\Documents and Settings[USERNAME]\Local Settings\Application Data\Firestorm
• Windows Vista: C:\Users[USERNAME]\AppData\Local\Firestorm
• Windows 7: C:\Users[USERNAME]\AppData\Local\Firestorm
• Mac: ~/Library/Caches/Firestorm
• linux: ~/.firestorm/cache

You will need to have show hidden files & folders enabled to browse to these locations.
See http://wiki.phoenixviewer.com/doku.php?id=showing_hidden_folders

• Open the folder named texturecache
• Inside this folder, find the files named texture.cache & texture.entries
• Make a copy of these 2 files & paste them into the new folder you created.

We also need to get a list of your directory structure:

• Browse to your Firestorm cache folder again
• On Windows, hold down SHIFT and right click the texturecache folder. Choose Open a command window here
• On linux, open a terminal/console window, and navigate to the folder: cd ~/.firestorm/cache/texturecache
• On Mac, open a Terminal window, and navigate to the folder: cd ~/Library/Caches/Firestorm/texturecache

• On Windows, at the command prompt, type in dir /s > directorylist.txt and hit enter
• On Linux and Mac, type in the command: ls -R . > directorylist.txt

• You should now see a file called directorylist.txt inside your texturecache folder.
• Copy this file & paste it into the new folder you created previously
• Zip up/compress this folder & attatch to this issue using More Actions -> Attatch file.

Actually got it fixed at least for my windows XP install. This should also work for Windows Vista and Windows 7. For MAC OS or Linux I can't make any suggestions. Go to the following link and DL and install this fix:

http://www.microsoft.com/download/en/details.aspx?displaylang=en&id=35

I can't say it will work for everyone and can't even definitively say it has for me since I have only logged in once. But so far so good.

Attached texture cache after crash. Note that crash occured with a clean cache and "HTTP textures" off.

Folks,,, I read and followed this comment...
Dmitri Polonsky added a comment - Dec 05, 06:39 AM +0100

Actually got it fixed at least for my windows XP install. This should also work for Windows Vista and Windows 7. For MAC OS or Linux I can't make any suggestions. Go to the following link and DL and install this fix:

http://www.microsoft.com/download/en/details.aspx?displaylang=en&id=35

I can't say it will work for everyone and can't even definitively say it has for me since I have only logged in once. But so far so good.
-------------------------------------
I downloaded and installed it,, them opened the viewer to the log in page, emptied my cache, closed it all and reloaded and opened the viewer and logged in....everything was as grey as to be expected,,,I just stood still until everything was loaded and started double clicking all over the place on my own parcel/sim...colours and sculpties were all much improved and the speed was "back-to-normal" for viewer 24336.... I logged out and back 2 more times...it looks as tho my problem has been fixed... I will check it all much deeper tomorrow,, but try that download, It certainly cant hurt and just might fix the problem...My sincere thanks Dimitri,, and of course always to the Phoenix team...Thanks all. . . . . MH

Thank you for all your logs and comments. Please keep them coming, we need as much information on this issue as possible,

Attached zipped cache files while running Firestorm 4.0.1 (24226) built with Zi's image_crash_patch_01
Logged in with clean cache, crashed approx 10 mins later with the llimage error

Cleared cache per instructions. Attached FirestormDebug-rje.zip file also includes log file and debug_info.log (and empty dump file created.)

Another crash. I tried installing the DirectX update from Microsoft as suggested by another poster. Crash still occurs. Included also is the listing of the entire local/firestorm directory.

llimage/llimage.cpp (244) LLImageBase::getData: "bad memory allocation for the image buffer!" over n over n over n over..grrr

This has also happened to me at various times.

I ran ok with the fix for a day then experienced the crash again. A friend suggested rolling back my NVidea drivers to version 260.99. I did this. Once more manually cleared cache... logged in set HTTP Invent back to true and HTTP Textures checked. This time I lasted almost 48 hours before experiencing the error again (albeit in a very crowded laggy sim). I have turned HTTP Textures off again and will see how this runs
Further note: After a couple of days the error returned, and I have since regularly experienced it and CTD's. I hope this gets sorted out soon its driving me crazy!!

Well. Something had to change. Installed the 64bit Ubuntu 11.10.. And after getting settled, in fallback mode (classic gnome), with the "Indicator Applet Ported To GNOME 3" and ia32-libs of course.. I'm running great !

I ran two sessions side be side today, smoothly. With HTTP textures -enabled & httpinventory -true (default settings). I've been logged in on multiple occasions, during several long periods of time with no sudden rise of memory use and no crashes.
I can't blame the viewer for anything now

Still getting the bad memory allocation...my info...

Firestorm 3.2.2 (24336) Nov 27 2011 17:05:37 (Firestorm-Release)
Release Notes

You are at 131,894.0, 190,438.0, 24.9 in Silver Bells located at sim21080.agni.lindenlab.com (216.82.30.114:13005)
Second Life RC KT 11.11.12.245070
Error fetching server release notes URL.

CPU: AMD Phenom(tm) 9600 Quad-Core Processor (2310.54 MHz)
Memory: 3071 MB
OS Version: Microsoft Windows 7 32-bit (Build 7600)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 8800 GT/PCI/SSE2/3DNOW!

Windows Graphics Driver Version: 8.17.0012.6658
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 1000
LOD factor : 2
Built with MSVC version 1600
Packets Lost: 136/5,146 (2.6%)

I'm getting this all the time too. I tried to install Fedora 16 (linux) but its heaps worse. hardly last 2 mins before a crash. so i went back to Fedora 15 (ghosted copy) which was fine before with the Mesh Beta. Its a little better but still happens regularly.

Here's my specs:

Firestorm 3.2.2 (24336) Nov 27 2011 15:24:57 (Firestorm-Release)
Release Notes

You are at 150198.0, 312172.0, 21.6 in Atrum Complexo located at sim9415.agni.lindenlab.com (216.82.43.118:12035)
Second Life RC KT 11.11.12.245070
Retrieving...

CPU: Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz (1600 MHz)
Memory: 4027 MB
OS Version: Linux 2.6.38.8-32.fc15.i686.PAE #1 SMP Mon Jun 13 19:55:27 UTC 2011 i686
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 8600 GT/PCI/SSE2

OpenGL Version: 3.3.0 NVIDIA 275.19
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/1.0.0d zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: OpenAL, version 1.1 ALSOFT 1.11.753 / OpenAL Community / OpenAL Soft: PulseAudio Software
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox 3.2.0002.10426

Settings mode: V3
Viewer Skin: starlight (original_orange)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 900
LOD factor : 3.7
Built with GCC version 40405
Packets Lost: 3105/76198 (4.1%)

this is the error

Thank you,
LisaMarie Freschi

Anything can be done for this?

I have had one more instance of this crop up. My Anti Virus removed SL Voice and quarantined my Firestorm program folder so I had to uninstall FS, Uninstall my AVV Anti Virus, then I decided to try using the microsoft Anti Virus. FS absolutely would not run with the MSE package. IT gave me that error again no matter what I did so I had to completely remove FS again, then MSE, then run the DX fix again, reinstalled AVG, the fix then FS on a clean install and not it is working fine again. NOt sure if that has any bearing on it but it seems that certain anti virus programs cause the issue as well

llimage/llimage.cpp (244) LLImageBase::getData: "bad memory allocation for the image buffer so everyone was telling me to switch to the 3.2.1 to get rid of this problem..it doesnt work. i signed on for a few days,no problem,,atm,ive tried logging on over and over and over,,i cant get online because of this error..has anyone figured it out yet,cuz like,if its gonna keep this up i guess i need to get rid of the viewer..i mean i like the features but if its going to do this features does u no good if u cant get online,lol

Hi everyone. Okay,the last time I commented was on 11/29 which is when I followed Whirly's request for the clean install. I had been running error free (multiple viewers and not clearing my cache up to today. Today, I decided it had been awhile so I cleared my cache through preferences (note that I only clicked the button to clear cache, I did not go in and delete the texturecache folder) and when I restarted the viewer, I got the error again along with a message indicating that my voice server was not reachable. When SLvoice.exe crashes, I always have to do a restart of my computer, so...I manually deleted the texturecache folder and did a restart of my computer. So far I am running without error again.

I am still receiving this error. This never happens on my home sim and many other sims I visit. I do know some of the sims I frequent will ALWAYS prompt the error, so I use Phoenix for those locations to avoid the error. Have not had this error with Phoenix at all.

Any official word on this?

This happens to me a LOT, probably a few times a day. BOOM, I'm dead. Please please fix it!

Can people getting this error, particularly those getting it right at login IM me inworld please. We have a build we would like you to test to see if theres any improvements.
Thanks

Another thing to try please:

Go to Avatar -> Prefs -> Graphics -> Rendering -> Use HTTP Textures -> untick it.
Then go to Avatar -> Prefs -> Network & cache -> Cache location -> open -> logout & delete the folder named texturecache -> relog & see if theres any improvement.

If this does help, please try the following:

Reenable HTTP Textures & set the Max number of concurrent HTTP GET requests to 1.
Delete your texture cache folder again & please let us know the results of this test
Thanks!

Initial crash using Firestorm development test (devFS32) HTTP Textures enabled.(8)
Crashed without error message.

Firestorm 4.0.1 (24267) Dec 14 2011 03:03:14 (Firestorm-private-devFS32)
Release Notes

CPU: Intel(R) Core(TM)2 CPU 6400 @ 2.13GHz (2131.21 MHz)
Memory: 3062 MB
OS Version: Microsoft Windows Vista 32-bit Service Pack 2 (Build 6002)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 7300 LE/PCI/SSE2

Windows Graphics Driver Version: 8.17.0012.6724
OpenGL Version: 2.1.2
RestrainedLove API: RLV v2.7.0 / RLVa v1.4.3a
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: V3
Viewer Skin: starlight (original_orange)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 1000
LOD factor : 2
Built with MSVC version 1600

Crash of devFS32 with HTTP Textures turned off.

Using devFS32 from Whirly - I can not get past Initializing Textture Cache. Eventually Windows says Firestorm not responging, I close the program and that's as far as I get. I waited over 20 minutes to move past Initializing Texture Cache before the not responding message.

CPU: Intel(R) Core(TM)2 Duo CPU P8400 @ 2.26GHz (2261.01 MHz)
Memory: 3037 MB
OS Version: Microsoft Windows 7 32-bit Service Pack 1 (Build 7601)
Graphics Card Vendor: ATI Technologies Inc.
Graphics Card: ATI Mobility Radeon HD 3400 Series

Windows Graphics Driver Version: 8.762-100803a-103697C-ATI
OpenGL Version: 3.3.10151 Compatibility Profile Context
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Hybrid
Viewer Skin: metaharper (appalachian_sky)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 3000
LOD factor : 4
Built with MSVC version 1600

Hi Alessandra,

Youll need to delete your cache manually by the looks of it.
Make sure hidden files & folders are enabled, see http://windows.microsoft.com/en-US/windows7/Show-hidden-files
Then browse to C:\Users[USERNAME]\AppData\Local\Firestorm
Delete the whole Firestorm folder. You should then be able to login.

Cleared the cache and it started. So far, no crashes with HTTP Textures on or off

It happens on 3.22 also. Below are my system specs and error.

Fatal Error
Error: llimage/llimage.cpp(244) : getData: Bad memory allocation for the image buffer!

Firestorm 3.2.1 (24179) Nov 27 2011 14:39:32 (Firestorm-Release)
Release Notes

CPU: Intel(R) Core(TM)2 Duo CPU E8135 @ 2.66GHz (2660 MHz)
Memory: 4096 MB
OS Version: Mac OS X 10.5.8 Darwin 9.8.0 Darwin Kernel Version 9.8.0: Wed Jul 15 16:55:01 PDT 2009; root:xnu-1228.15.4~1/RELEASE_I386 i386
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: NVIDIA GeForce 9400 OpenGL Engine

OpenGL Version: 2.0 NVIDIA-1.5.48
RestrainedLove API: RLV v2.7.0 / RLVa v1.4.3a
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 264
Bandwidth : 6000
LOD factor : 4
Built with GCC version 40201

Have been using devFS32 since early this morning! Visited sims that have always prompted the bad memory, etc. So far so good today!

The build that Whirly gave me (4.0.1.24267) seems to have fixed this bug. Tested with clean & dirty cache, HTTP textures on & off - none of the 4 combinations induced a crash.

On a purely subjective note, the UI also seems more responsive - don't know if it's related to this or not.

The build supplied by Whirly worked well until a few minutes ago when I got another Fatal Error... not the same one... this one says:

ERROR: llimage.cpp(173) : LLImageBase::allocateData:LLImageBase:allocateData called with bad dimensions: 0x0x1

Hi Whirly,
Tested the nightly build you gave me, and still crashed within 40 minutes on my land on Hellishly, both with texture cache cleared and with it uncleared (made no difference either way). I have uploaded the zipfile you wanted.

System specs:
Firestorm 4.0.1 (24267) Dec 14 2011 03:03:14 (Firestorm-private-devFS32)
Release Notes

You are at 113,443.0, 313,524.0, 22.9 in Blessed Land located at sim8491.agni.lindenlab.com (216.82.39.50:13011)
Second Life Server 11.12.03.246118
Error fetching server release notes URL.

CPU: Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz (2400.05 MHz)
Memory: 3583 MB
OS Version: Microsoft Windows XP Service Pack 3 (Build 2600)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 9800 GT/PCI/SSE2

Windows Graphics Driver Version: 6.14.0012.7533
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 1500
LOD factor : 2
Built with MSVC version 1600
Packets Lost: 142/8,768 (1.6%)

Also tested this build with 2 of my alts. I was UNABLE to crash the viewer at all with either alt, no matter how often I tp'd to laggy, crowded, crash-fest places.

So again it looks like Hellishly is the only one of my avatars to have this issue.

Just to update...I've been inworld for 5 hours and have not experienced a crash on the dev build, and I've TP'd hither and yon. Well, maybe not hither..but definitely yon.

@Hellishly
Does Hellishly have a much bigger inventory then the 2 alts?

Whirly, yes she does. Well around double their size. I'm in the process of loading up one alt with inventory to test this.

I've been getting this on login for the past 2 days but only with one avi, my alt logged in fine. And the one giving me the error only has 6K in inventory while the alt not crashing has way more so I doubt that being the issue. Manually clearing cache seems to help. Funny though since the supprt in Phoenix group told me not to clear my cache daily while this issue exists when I don't.

Whirly, I took my alt's inventory up to 65,000, and tp'd her around to some of the laggiest shopping places I know of and still no error, or crash.
Even after 4+ hours of tping and flying fast around my island (flying fast around my island will usually crash Hellishly on any viewer) I still cannot create the error on my alt. Inventory is now at 70,000.

Thanks for all the info!
When you log Hellishly in, how much RAM roughly is Firestorm taking up in Windows task manager -> Processes compared to the alt? Does Hell have more nested folders within folders? (As far as I know, this is more expensive for memory (but dont quote me on that)
I can only look at whats happening here from a Support standpoint & try to come up with things for people to try to workaround this bug, while the devs are looking at the underlying reasons.
So far, if Im understanding this correctly, there seems to be at least 2 things going on here. People who are being hit with this error after being online a while, or when in certain regions or busy/texture rich regions are maxing out on RAM & get an out of memory crash (this is whats happening in my case & only on Whirly (96k inventory), my alts have never had this happen (small inventories)).
This doesnt explain why some people are getting this crash constantly right at login though. That appears to be due to something else.

@Siggy
Normally cache clearing daily just for the sake of it is a bad thing. However, if this is working for you to avoid this crash at the moment, please continue to do so. Let us know if manually clearing cache prelogin makes the error stop for you again. Another thing to try is only manually deleting your texturecache folder prelogin & see if this works as well as deleting the full cache. Depending on the results, we can try to narrow it down further as to which part of the cache deleting is "fixing" this.

So far these settings changes are helping some people. Please try changing one at a time, run for a day or so with each setting changed & let us know if any are helping:

From the login screen:

• Viewer -> Preferences -> Graphics -> Hardware settings -> Untick enable streamed VBO's then untick enable open GL vertex buffer objects. (This will likely lower your overall FPS, particularly on Nvidia cards, so be aware of that)

• Viewer -> Preferences -> Graphics -> Rendering -> Untick HTTP Textures
  Then go to Viewer -> Prefs -> Network & cache -> Cache location -> open -> close login screen & delete the folder named texturecache -> relog & see if theres any improvement.
  If this does help, please try the following:
  Reenable HTTP Textures & set the Max number of concurrent HTTP GET requests to 1.
  Delete your texture cache folder again & please let us know the results of this test

• Viewer -> Preferences -> Graphics -> Hardware settings -> Viewer texture memory buffer -> Lower this to 256 MB

4whirly..so i downloaded the dev. version yesterday after u imd me..as soon as it uploaded and asked for it to start,i logged in,and got the buffer errors,took 3 times to get online,it seems that this happens before the bridge starts to create,cuz if the bridge starts it doesnt happen,i dont know if thats important or not..it didnt happen anymore yesterday and i logged on successfully this morning,so i guess see how it goes,i provided my info in case ya needed it .my inventory is at 22k,but i do have a alt that ive experience the same problem with and her inventory is at like 4k
Firestorm 4.0.1 (24267) Dec 14 2011 03:03:14 (Firestorm-private-devFS32)
Release Notes

You are at 264,660.0, 281,500.0, 24.8 in Avalon Island located at sim8937.agni.lindenlab.com (216.82.41.113:13005)
Second Life RC Magnum 11.12.12.246583
Release Notes

CPU: Intel(R) Core(TM) i5 CPU M 480 @ 2.67GHz (2660.37 MHz)
Memory: 3894 MB
OS Version: Microsoft Windows 7 64-bit Service Pack 1 (Build 7601)
Graphics Card Vendor: ATI Technologies Inc.
Graphics Card: ATI Mobility Radeon HD 6370

Windows Graphics Driver Version: 8.14.0001.6143
OpenGL Version: 4.0.10188 Compatibility Profile Context
RestrainedLove API: RLV v2.7.0 / RLVa v1.4.3a
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox 2.1.3010.6270

Settings mode: Phoenix
Viewer Skin: metaharper (appalachian_sky)
Font Used: Deja Vu (96)
Draw distance : 1000
Bandwidth : 1700
LOD factor : 2
Built with MSVC version 1600
Packets Lost: 1/7,211 (0.0%)

Hey Whirly - Still no crashes and I have been moving around to places that had caused problems before. I think my scenario must be more like yours where after awhile, I just run out of RAM. My avatar has about 9K for inventory with folders nested in other folders alot.

I have HTTP Textures on, HTTP Get requests = 8

Firestorm 4.0.1 (24267) Dec 14 2011 03:03:14 (Firestorm-private-devFS32)
Release Notes

You are at 237,008.0, 241,370.0, 3,599.9 in Revelations located at sim2521.agni.lindenlab.com (216.82.19.235:12035)
Second Life Server 11.12.03.246118
Release Notes

CPU: Intel(R) Core(TM)2 Duo CPU P8400 @ 2.26GHz (2260.99 MHz)
Memory: 3037 MB
OS Version: Microsoft Windows 7 32-bit Service Pack 1 (Build 7601)
Graphics Card Vendor: ATI Technologies Inc.
Graphics Card: ATI Mobility Radeon HD 3400 Series

Windows Graphics Driver Version: 8.762-100803a-103697C-ATI
OpenGL Version: 3.3.10151 Compatibility Profile Context
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Hybrid
Viewer Skin: metaharper (appalachian_sky)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 3000
LOD factor : 4
Built with MSVC version 1600
Packets Lost: 33/7,728 (0.4%)

Was getting this whenever I'd get to a certain nearly-maxed point with my ram running 32 bit Win7. Considering updating to 64 bit to enable use of more ram as I know this is a ram error. Usually would happen after I had been on for a while as I have a pretty good amount of ram.

Today, I tp'd into a fairly high lag / highly textured area (Earthstones) and got this error after being there for some time. Relog and I get the error before I can fully sign in. Tried a relog, same thing. Pulled up resource monitor and find I am no where near maxing out ram according to it. Don't know what to think now. Finally logged back in elsewhere with no troubles.

Whirly, I logged Helli in with a completely clean cache. TP'd her around the grid some, then flew around and around my sim while watching Mem Usage. While I couldn't get the "bad memory" error this time, my RAM usage kept going up and up until it hit just over 1.7GB, and the viewer silently crashed to desktop (this is common on Helli in every viewer I have tried). Her inventory is just over 95,000 at the moment, and most of that is just nested once (folder within a folder) under fairly broad categories.

Midnite (my alt), seems to be able to stay on with no issue. Currently her inventory is 75,000. Her inventory structure is pretty much the same as Helli's. Memory usage on Midnite doesn't go over 870 MBs.

Day two using Firestorm-private-devFS32. Still working very well. No crashes and/or errors.

I crashed with the dev version(24267) after running several hours.I had set my video cache buffer to 256MB I also had upped my cache size to almost 2 GB and was running medium graphics, disabled VBO and no HTTP textures. I recall the last displayed memory usage to be at around 1,520 MB

I hope login crashes will get fix because i can't live without firestorm.
Here's the crashes error: ERROR: llimage/llimage.cpp(244):LLimageBase::getData: Bad memory allocation for the image buffer.
My pc is:
processor Intel Core Quad CPU Q9650 @ 3.Ghz
Memory: 6.00GB
System type 64-bit OS

I get a memopry error every time I log in after logging out. The only way to log in succesfully is to clear the cache every time. While this is not critical. it is annoying. Especially since I have a new computer. here are the specs:

OS Name Microsoft Windows 7 Professional
Version 6.1.7601 Service Pack 1 Build 7601
Other OS Description Not Available
OS Manufacturer Microsoft Corporation
System Name GEMINI7SYS
System Manufacturer Gigabyte Technology Co., Ltd.
System Model GA-970A-UD3
System Type x64-based PC
Processor AMD FX(tm)-8150 Eight-Core Processor, 3600 Mhz, 8 Core(s), 8 Logical Processor(s)
BIOS Version/Date Award Software International, Inc. F3, 8/31/2011
SMBIOS Version 2.4
Windows Directory C:\Windows
System Directory C:\Windows\system32
Boot Device \Device\HarddiskVolume1
Locale United States
Hardware Abstraction Layer Version = "6.1.7601.17514"
User Name Gemini7sys\Anthony
Time Zone Pacific Standard Time
Installed Physical Memory (RAM) 8.00 GB
Total Physical Memory 8.00 GB
Available Physical Memory 5.21 GB
Total Virtual Memory 16.0 GB
Available Virtual Memory 10.7 GB
Page File Space 8.00 GB
Page File C:\pagefile.sys

I have two AMD 6850 graphic Cards set up with Crossfire each with 1gigabyte of memory.

Here is my Viewer info:

Firestorm 3.2.2 (24336) Nov 27 2011 17:05:37 (Firestorm-Release)
Release Notes

You are at 263,387.0, 253,408.0, 126.5 in Keener located at sim9839.agni.lindenlab.com (216.82.45.173:13002)
Second Life Server 11.12.03.246118
Error fetching server release notes URL.

CPU: AMD FX(tm)-8150 Eight-Core Processor (3624.33 MHz)
Memory: 8190 MB
OS Version: Microsoft Windows 7 64-bit Service Pack 1 (Build 7601)
Graphics Card Vendor: ATI Technologies Inc.
Graphics Card: AMD Radeon HD 6800 Series

Windows Graphics Driver Version: 8.17.0010.1103
OpenGL Version: 4.1.11251 Compatibility Profile Context
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: metaharper_sidetabs (default)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 1000
LOD factor : 4
Built with MSVC version 1600
Packets Lost: 10,687/670,303 (1.6%)

Not sure why I have to clear cache3 every time. If I do not clear Cache, SL Crashes and closes almost instantly upon showing me my avie with the error msg regarding memory. which I can not believe, I have double the memory of my old computer. Oh yes, SL runs SMOOTHLY with no problems at all if I clear cache before logging in. However if I log out and then try to log in later without clearing cache it wont work until I clear cache. I have done a clean install twice. I think it is a viewer error when trying to access cache upon logging.

@Adria

When you persistently get this error on login, does it make any difference if you login to a different region?

Can you test something please.
Instead of deleting your full cache when this happens, can you just delete the texturecache folder & see if this is enough to stop the error
From the login screen, Viewer -> Prefs -> Network & cache -> Cache location -> open
Only delete the folder named texturecache. Can you then login without the error message?
If not. next time try seleting everything in that folder except texturecache folder & please let us know the results.
Thanks!

Updating. On Helli today I was able to stay logged in a fair while. Granted I was just mostly in inventory unpacking stuff, trying stuff on, etc. Everything ran smooth as silk until I went to take a high-res photo, then boom crashed with the error.

Edit: Looks like taking a photo with or without hi-res is a death knell for Helli in this viewer.
100% reproducible. Log in, pose, open photo interface, check save to hard disk, hit refresh, viewer crashes.

2nd Edit as I forgot to include RAM usage...RAM usage hit roughly 1.6GB when this error occurs on Helli.

Helli, if you disable VBO, do you still crash taking a photo?

Whirly, disabling VBO fixed the photo issue. Thanks

Updating on Midnite. Was able to TP around, visit a very busy Photo Studio area (several photo studios for public use, very popular), and pose, taking high res photos for almost an hour (VBO enabled). TP some more to places like Truth hair, and still never got the error.
Midnite's inventory is at 75,000, her RAM usage did not got over 1GB.

I tried yesterday the version Whirly sent me.
I must tell I wasn't online that much, just a couple of hours. The first thing I did was visiting the sim where I experienced the error the first time. I can't guarantee the conditions were exactly the same, but I didn't crash. Then I visited an event venue (rather empty actually), setting graphics on high, LOD on 4 and took some snapshots, everything went fine. Last, I took some pics in studio, still no crash. As a whole it looks better, although I haven't tried exploring very busy sims, where I had experienced the error too.
Winter's inventory is 109k, rather nested.

I had firestorm working just fine until it crashed and I wasn't able to use it again. I've been getting that message every time I try logging in. I uninstalled it. Reinstalled it. Cleared cache. Turned off HTTP textures and it still doesn't work.

Trying to provide documentation of this issue.

Two pieces of information for the jira.

ok here it goes again.

Model Name: iMac
Model Identifier: iMac12,2
Processor Name: Intel Core i7
Processor Speed: 3.4 GHz
Number of Processors: 1
Total Number of Cores: 4
L2 Cache (per Core): 256 KB
L3 Cache: 8 MB
Memory: 8 GB
Chipset Model: AMD Radeon HD 6970M
Type: GPU
Bus: PCIe
PCIe Lane Width: x16
VRAM (Total): 2048 MB

It only seems to happen when I log in. And often can log in immediately afterward. Strangely enough, this hasn't happened until just recently for me. Only been happening for a couple of days for me.

------------------------------------------------------------

Firestorm 3.2.2 (24336) Nov 27 2011 17:05:37 (Firestorm-Release)
Release Notes

You are at 192,145.0, 357,030.0, 27.9 in Hathian located at sim4477.agni.lindenlab.com (216.82.22.63:13002)
Second Life Server 11.12.03.246118
Release Notes

CPU: Intel(R) Pentium(R) D CPU 3.20GHz (3192.08 MHz)
Memory: 1023 MB
OS Version: Microsoft Windows XP Service Pack 3 (Build 2600)
Graphics Card Vendor: ATI Technologies Inc.
Graphics Card: ATI Radeon HD 4600 Series (It's an ATI Radeon HD 4670 1GB)

Windows Graphics Driver Version: 8.911-111025a-128238C-ATI
OpenGL Version: 3.3.11251 Compatibility Profile Context
RestrainedLove API: RLV v2.7.0 / RLVa v1.4.3a
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox

Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 1000
LOD factor : 2
Built with MSVC version 1600
Packets Lost: 8/4,903 (0.2%)

Tadaaaa, I crashed with a brand new (to me) fatal error! I was taking pics, switching skies, graphics set to high. (using the program Whirly gave me)

ERROR: newview/llappviewer.cpp(1619) : LLAppViewer::mainLoop: Bad memory allocation in LLAppViewer:::mainLoop()!

Don't laugh at my poor old pc

Firestorm 4.0.1 (24267) Dec 14 2011 03:03:14 (Firestorm-private-devFS32)
Release Notes

CPU: Intel(R) Core(TM)2 CPU 4300 @ 1.80GHz (1805.51 MHz)
Memory: 3008 MB
OS Version: Microsoft Windows XP Service Pack 3 (Build 2600)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 9600 GT/PCI/SSE2

Windows Graphics Driver Version: 6.14.0011.9045
OpenGL Version: 3.1.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 1000
LOD factor : 4
Built with MSVC version 1600

Just pitching in...
ERROR: llimage/llimage.cpp(244) : getData: Bad memory allocation for the image buffer!

Firestorm 3.2.2 (24336) Nov 27 2011 15:24:57 (Firestorm-Release)
Release Notes

CPU: AMD Athlon(tm) 64 X2 Dual Core Processor 6000+ (1000 MHz)
Memory: 4023 MB
OS Version: Linux 2.6.35-31-generic-pae #63-Ubuntu SMP Mon Nov 28 20:48:50 UTC 2011 i686
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 9800 GT/PCI/SSE2/3DNOW!

OpenGL Version: 3.3.0 NVIDIA 260.19.06
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/1.0.0d zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: OpenAL, version 1.1 ALSOFT 1.11.753 / OpenAL Community / OpenAL Soft: PulseAudio Software
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 200
Bandwidth : 1500
LOD factor : 2
Built with GCC version 40405

I did a complete wipe of everything SL related off my computer (including registry) and re-installed the version Whirly gave us only. Will see how this goes, and see if it helps fix the issue for Helli. I figure there must be something specific to her, and it could have been a corrupt setting somewhere from all the different viewers she has been on over the years.

I am having the same issue and it is really anonying

@Whirley Fizzie

I tried doing as you asked. The error I am getting is 'ERROR: llimage/llimage.cpp(244): LLImageBase:getData: Bad memory allocation for the image buffer!'

1st: I tried logging out of a different region then I normally do and logging into last location there. I still would crash.

2nd: I tried just deleting the texturcache folder per your instructions. I still crashed.

3rd: I tried just deleting the items inside the texturcache folder, but left the folder. I still crashed Note: The items would re-appear almost immediately inside the texture cache folder after deleting except for the file 'texture.cache'

4th: Just to test again, I deleted only the items in texturecache, once more the items almost immediately re-apppeared except for texture.cache file and I got the error message again.

5th: I thought maybe I should not delete items a second time when the re-appear, oddly enough though they would not re-appear any longer on this attempt and I was able to log in.

6th: To see if this solved the issue I tried logging in again immediately after logging out. I crashed.

7th: Once more deleted everything in folder, the items re-appeared without the texture.cache file. I did not delete the itemws a second time though. I was able to log in. It looks like it may be the texture.cacher file causing it. I will attempt a log out, try logging in. If the error occurs again, I will check for that file and just delete that one and see if I can log in.

8th: Sure enough I crashed on the re-log attempt after logging out. Will try logging in 9th time with only deleting texture.cache file.

9th: Still got the error message when just deleting the texture.cache file. So must be something else. Must be something that is added or changed after login. As I said I did not delete a second time after they re-appeared immediately (less texture.cache) and I was able to login without error. So something in there gets changed after login and its not the texture.cache file as I was unable to log in after solely deleting that.

10th: Deleting only texture.entries does not work also. Have to delete all items in textr=urecache folder at least once before having a succesful log in. Though it does not always work deleting only once, sometimes you need to delete the re-appears. Clearing Cache every time though seems to work and so far deleting twice works also.

So far have not seen this error since deleting all SL content off my HD, and only installing the version from Whirly. Have had several silent crash to desktops though. And one "ERROR: llrender/llvertexbuffer.cpp(1197) : LLVertexBuffer::mapVertexBuffer: memory allocation for vertex data failed" error (which was new to me).

Is there an updated test viewer we can test on at all?

updating. Have now had the above error twice (cpp 1197). RAM hits 1.7 GB at that point.
Not yet had the cpp (244) error again.

I get this error on 3.2.1 when i got to a crowded area, usually around 20+ avs there. Usually ram hits 1.5gb or so then I get the fatal error. Could it be the cache size and what is a good size for it? I got that slider all the way right.

Before starting Firestorm 3.2.2.24336, I deleted all cache files and verified that my 32-bit Windows XP system had 2.7GB of available memory (4GB physical memory). Before my large inventory loads, my viewer gets a "LLImageBase::getData: bad memory allocation for the image buffer" error. This crash happens when the amount of memory used by Firestorm while loading inventory reaches about 1.7GB. The periodic memory usage information in the crash log contains "INFO: llcommon/llmemory.cpp(167) : LLMemory::logMemoryInfo: Current max usable memory(KB): 1677721". I repeated this failure several times as I disabled various graphics settings.

How did Firestorm determine the "current max usable memory" of about 1.7GB?
Why didn't Firestorm use the full 2.7GB of memory that was actually available?
How can I get Firestorm to use more of my available memory?
Would a large 64-bit memory system also be limited to only 1.7GB of memory for Firestorm?

Twice I've been able to predict that this error was about to happen because textures went funny. Once an attachment I was wearing turned transparent a few seconds before the crash; another time, a prim near me turned black before the crash.

Windows XP Pro v.2002 SP2 32 bit
Intel Core 2 Quad Q6600 @ 2.40GHz
RAM 3.50 GB

Graphics: NVIDIA GeForce 9600 GSO
Driver version 6.14.12.8026 date 8/3/2011

This seems to be happening more frequently during last week or two. Happens every 3 to 5 minutes when I'm near 20+ avatars. Otherwise, it happens about every 20 to 45 minutes.

Reducing the Viewer Texture Memory Buffer setting seems to have helped somewhat.

ERROR:llimage/llimage.cpp(244) : LLImageBase::getData: Bad memory allocation for the image buffer!

Firestorm 3.2.2 (24336) Nov 27 2011 17:05:37 (Firestorm-Release)
Release Notes

You are at 183,710.0, 281,682.0, 26.9 in Raad Ny Ashlinagh located at sim4305.agni.lindenlab.com (216.82.21.51:13000)
Second Life Server 11.12.03.246118
Release Notes

CPU: AMD Athlon(tm) 64 X2 Dual Core Processor 4400+ (2310.48 MHz)
Memory: 3326 MB
OS Version: Microsoft Windows Vista 32-bit Service Pack 2 (Build 6002)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 210/PCI/SSE2/3DNOW!

Windows Graphics Driver Version: 8.17.0012.8562
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox 2.1.3010.6270

Settings mode: V3
Viewer Skin: starlight (original_teal)
Font Used: Deja Vu (96)
Draw distance : 96
Bandwidth : 10000
LOD factor : 4
Built with MSVC version 1600
Packets Lost: 346/43,687 (0.8%)

I didn't crash at login but after running Firestorm-Release 3.2.1 for more than an hour yesterday. I did what was recommended, did a clean reinstall and unchecked http get textures. Textures loaded a lot faster in the beginning. Today I noticed that about 1-2 hours of being online textures took ages to load. I'm sorry I didn't have crash logs open when it happened again. Bad memory location for the image buffer, and then crash to desktop. Here's the computer infos:

Firestorm 3.2.1 (24179) Nov 27 2011 15:25:09 (Firestorm-Release)
Release Notes

CPU: Intel(R) Core(TM) i7 CPU 860 @ 2.80GHz (2800.23 MHz)
Memory: 3319 MB
OS Version: Microsoft Windows 7 32-bit Service Pack 1 (Build 7601)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 9600 GT/PCI/SSE2

Windows Graphics Driver Version: 8.17.0012.7533
OpenGL Version: 3.3.0
RestrainedLove API: RLV v2.7.0 / RLVa v1.4.3a
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 1000
LOD factor : 4
Built with MSVC version 1600

I crashed again, same settings as above, but this time the fatal error was worded a bit different:

LLImage Base: allocateData called with bad dimensions 0x0x1

This morning I already crashed twice with this error, the second time I only managed to be online for 18 minutes. Crash log attached.

I keep on crashing to desktop with this error(s), and this critical bug is still unressolved and open after almost 4 weeks? FS has become useless for me, as I am experiencing crashes to desktop several times an hour unless I am all alone in an empty region. Somebody fix this asap.

Same thing here.. First time I get this message today.. Crashing in the mid of the loging with the same message llimage/llimage.cpp(244) : LLImageBase::getData: Bad memory allocation for the image buffer!

I've been getting this error on Linux (x86_64) since upgrading from Fedora 14 to Fedora 16. Firestorm worked fine before that.

In this particular instance, I was closing an IM tab, but have seen it when just moving the camera as well.

2011-12-27T13:58:46Z WARNING: llimage/llimage.cpp(198) : allocateData: allocate image data: 4194304
2011-12-27T13:58:46Z WARNING: llimage/llimage.cpp(198) : allocateData: allocate image data: 4194304
2011-12-27T13:58:46Z WARNING: llimage/llimage.cpp(198) : allocateData: allocate image data: 4194304
2011-12-27T13:58:48Z WARNING: llimage/llimage.cpp(198) : allocateData: allocate image data: 4194304
2011-12-27T13:58:48Z WARNING: llimage/llimage.cpp(198) : allocateData: allocate image data: 4194304
2011-12-27T13:58:48Z llimage/llimage.cpp(244) : error
2011-12-27T13:58:48Z ERROR: llimage/llimage.cpp(244) : getData: Bad memory allocation for the image buffer!
LogLock::LogLock: failed to get mutex for log
LogLock::LogLock: failed to get mutex for log

I've been crashing a lot unless i am at my own home. the error i'm getting is error:llimage/llimage.cpp(244):LLImageBase::getData:Bad memory allocation for image buffer
my system info is:
CPU: Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHz (3000.04 MHz)
Memory: 3328 MB
OS Version: Microsoft Windows XP Service Pack 3 (Build 2600)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 9800 GTX+/PCI/SSE2

Windows Graphics Driver Version: 6.14.0012.8558
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox 2.1.3010.6270

Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 1500
LOD factor : 4
Built with MSVC version 1600
Packets Lost: 46/58,867 (0.1%)

thank you for any help you might be able to give me with this

I asked in group chat yesterday and got a website with tips what to do if you crash often. I chose the ones that felt like making sense: I cleared cache manually, then unchecked the box "Enable open GL Vertex Buffer Objects" in Hardware Settings/Graphics tab. I also set cache size to 1088 (it has been bigger before because my friend had suggested it, but I myself never made the cache size bigger than default). Since then I've been online twice for several hours without any crash yet, which is a major improvement compared to crashing twice every 20 minutes yesterday. Hope this helps.

Hey there, I've been getting this same error intermittently since I upgraded to the release. Duration online and population of sim have varied, though it happens more frequently in highly populated sims.

System info:
Firestorm 3.2.2 (24336) Nov 27 2011 17:05:37 (Firestorm-Release)

Second Life Server 11.12.03.246118
Error fetching server release notes URL.

CPU: AMD Athlon(tm) 64 X2 Dual Core Processor 3800+ (2010.3 MHz)
Memory: 2048 MB
OS Version: Microsoft Windows XP Service Pack 3 (Build 2600)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 8500 GT/PCI/SSE2/3DNOW!

Windows Graphics Driver Version: 6.14.0012.6658
OpenGL Version: 3.3.0
RestrainedLove API: RLV v2.7.0 / RLVa v1.4.3a
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox 2.1.3010.6270

Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 96
Bandwidth : 1000
LOD factor : 2
Built with MSVC version 1600
Packets Lost: 19/14,467 (0.1%)

I'm testing version 3.2.1.24179 to see if using it instead of 3.2.2.24336 improves stability.

i have this error but it seems at least for me to have nothing to do with camming around or crowded places at all ill be in an almost empty room and simply typing and it will happen even without a restart ill try running in terminal mode from now on and see if i cant catch more info from that

Switching to version 3.2.1 did not help with the crashing while in crowded areas; and I still received the same errors.

However; reducing graphics setting to the standard "Low" settings seems to have not only stopped the crashing, but my FPS has doubled in all areas that I usually visit, in both mostly empty and crowded areas.

Touch wood what ever has been done to either LL or firestorm has stemmed the flow of ctd for me... My draw distance is set at 192m which is fine for me...

Firestorm 3.2.2 (24336) Nov 27 2011 17:05:37 (Firestorm-Release)
Release Notes

You are at 193,354.0, 260,006.0, 68.5 in Slaver Bay located at sim5964.agni.lindenlab.com (216.82.12.18:13003)
Second Life Server 11.12.03.246118
Release Notes

CPU: AMD Athlon(tm) 64 X2 Dual Core Processor 5200+ (2712.57 MHz)
Memory: 3328 MB
OS Version: Microsoft Windows XP Service Pack 3 (Build 2600)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 8500 GT/PCI/SSE2/3DNOW!

Windows Graphics Driver Version: 6.14.0012.7533
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox 2.1.3010.6270

Settings mode: V3
Viewer Skin: starlight (original_orange)
Font Used: Deja Vu (96)
Draw distance : 192
Bandwidth : 1000
LOD factor : 2
Built with MSVC version 1600
Packets Lost: 1,016/64,495 (1.6%)

so on dec 15th i started using the new private dev fs32..i had problems the first time i signed on..actually twice that nite but i had no problems whatsoever since, until last nite and its happened 3 times..not sure what changed:/...i jst tried to log in now and got it again.

This actually appears to be the same cause as https://jira.secondlife.com/browse/SH-1650 and https://jira.secondlife.com/browse/SH-2793. Latest Shining version also crashes in cases with lots of avatars and textures.

I experienced this problem too. After I did this:

Debug -> Show debug settings -> UseHTTPInventory -> Set to FALSE

I could login and stay logged in without any problems.

I'm getting this error regularly on Max OSX Snow Leopard.

Here are my Computer configs:
Firestorm 3.2.2 (24336) Nov 27 2011 16:12:10 (Firestorm-Release)
Release Notes

You are at 241533.0, 256100.0, 23.9 in Zyrra located at sim6010.agni.lindenlab.com (216.82.12.64:13001)
Second Life Server 11.12.12.246583
Error fetching server release notes URL.

CPU: Intel(R) Core(TM) i7 CPU 860 @ 2.80GHz (2800 MHz)
Memory: 8192 MB
OS Version: Mac OS X 10.6.8 Darwin 10.8.0 Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 i386
Graphics Card Vendor: ATI Technologies Inc.
Graphics Card: ATI Radeon HD 4850 OpenGL Engine

OpenGL Version: 2.1 ATI-1.6.36
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Not Connected
Settings mode: Phoenix
Viewer Skin: starlight (silver_blue)
Font Used: Deja Vu (96)
Draw distance : 100
Bandwidth : 300
LOD factor : 4
Built with GCC version 40201
Packets Lost: 21/21865 (0.1%)

I got my first CTD from this (not sure whether to cheer or growl
Logs, including sysinfo, attached.

Shortly before the crash, maybe five minutes, I did move a LOT of objects in inventory. I selected enough to freeze the viewer for a couple seconds. CPU usage went up on 3 cores but RAM use didn't go over 8MB (I have 12).

VRAM was roughly 800/1024MB in use.

In the 5 minutes after moving inventory, I was in IM, then moved my camera.

ok now the private dev isnt working,,i cant get on firestorm at all:/

https://jira.secondlife.com/browse/SH-2494
Maybe related?

Just to add further to this, I continued to experience the error and CTD's until 2 weeks ago. I upgraded my motherboard and went from 4 to 8gig of ram. I also reformatted my hard drive and reinstalled all regularly used software and reinstalled the priv dev software advised by whirly. Since then I have only had random CTD's and no Fatal Crass Errors. My updated info is
Firestorm 4.0.1 (24267) Dec 14 2011 03:03:14 (Firestorm-private-devFS32)
Release Notes

You are at 115,412.0, 358,465.0, 1,499.0 in Mad World located at sim5365.agni.lindenlab.com (216.82.20.55:13002)
Second Life Server 11.12.12.246583
Error fetching server release notes URL.

CPU: AMD Phenom(tm) II X4 840 Processor (3221.6 MHz)
Memory: 8190 MB
OS Version: Microsoft Windows 7 64-bit Service Pack 1 (Build 7601)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 9600 GT/PCI/SSE2

Windows Graphics Driver Version: 8.17.0012.6099
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox 2.1.3010.6270

Settings mode: Phoenix
Viewer Skin: metaharper (appalachian_sky)
Font Used: Deja Vu (96)
Draw distance : 512
Bandwidth : 2100
LOD factor : 4
Built with MSVC version 1600
Packets Lost: 12/3,852 (0.3%)

having same issue did HTTPinventory fix and that helped

Firestorm 3.2.1 (24179) Nov 27 2011 15:25:09 (Firestorm-Release)
Release Notes

You are at 246,964.0, 275,061.0, 1,629.7 in Sylfie located at sim2842.agni.lindenlab.com (216.82.0.116:13002)
Second Life Server 11.12.12.246583
Release Notes

CPU: Intel(R) Core(TM) i7 CPU 860 @ 2.80GHz (2808.74 MHz)
Memory: 6136 MB
OS Version: Microsoft Windows 7 64-bit Service Pack 1 (Build 7601)
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: GeForce 8800 GTX/PCI/SSE2

Windows Graphics Driver Version: 8.17.0012.8562
OpenGL Version: 3.3.0
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox 2.1.3010.6270

Settings mode: Hybrid
Viewer Skin: metaharper (appalachian_sky)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 1000
LOD factor : 2
Built with MSVC version 1600
Packets Lost: 26/41,047 (0.1%)

Having multiple crashes daily. I tried going back to Beta-Mesh, and kept crashing on that, too (different message given). I finally uninstalled everything Firestorm from my computer and deleted remaining Firestorm folders from Application Data folders - then did a clean reinstall of Firestorm v3.2.2.24336 earlier this evening. On logging in the first time, before any play, I made these changes in Preferences in hopes to prevent a crash:

-reduced graphics setting to "Low" and unchecked all graphics extras, everything from transparent water to OpenGL.
-unchecked Use HTTP Textures
-increased cache to 1024

After a couple hours of play on a fairly crowded/laggy sim, I got the (244) crash.

Per Whirly's post made on Dec 5, I created a folder called "Fiona Info 244 crash" with some files that I hope will be useful, and will try to attach/upload it. Included in the folder are files copied from texturecache immediately after the crash as directed by Whirly:

texture.cache
texture.entries
directorylist.txt (with some personal info edited out, but SL/Firestorm info included)
Firestorm.log
plus a text file of my Environment

Getting near the end of my rope here. Let me know if I can be of further help - and please do let me know if you have any advice for me. Thank you.

PS. Earlier today, I also tried the DirectX update fix/link mentioned by some above, and was notified that I already have the latest version.

Just as a side note, possibly related, I discovered that my machine was running pretty tight on virtual memory - - w/ explorer.exe hogging an ungodly amount at various times, depending upon what I was doing prior. Paging file size seemed smaller than it should be, so I'm playing w/paging file size (and ending/restarting the explorer.exe process if it gets too huge) and will make a note here if that seems to help.

January 24, 2012 update. HAVE NOT CRASHED since lowering Viewer texture memory buffer to 256 MB as recommended by Whirly above in mid-December. Wish I'd seen that earlier, thanks Whirly:

"Viewer -> Preferences -> Graphics -> Hardware settings -> Viewer texture memory buffer -> Lower this to 256 MB"

I was crashing a few times a day before that, with different error messages, including the (244) error. Lowering and unticking all my graphics settings, including HTTP textures, had NOT helped. But since making the change above I have not crashed.

This error occurred first for me after my first cache clean after a totally clean reinstall. The error occurred only at login and not during any sessions. Have just logged in successfully with both HTTP texture settings disabled,and rezzed immediately. Logout is smooth. Previously freezing at logout. Now logged in successfully again with UseHTTPInventory set to TRUE.

And one more for the growing list of "bad <insert graphic thing here>" errors. I got a "bad Vertex buffer" error tonight while on a Linden sim. I was crossing sim boundaries between a cluster of these sims and each time I'd cross a specific boundary, I'd crash. I couldn't even log to the last place without crashing with this fatal error. Screenshot of error and log to follow.

System specs:
Firestorm 3.2.2 (24336) Nov 27 2011 16:12:10 (Firestorm-Release)
Release Notes

You are at 261913.0, 244515.0, 34.2 in Magnum Sandbox 3 located at sim9028.agni.lindenlab.com (216.82.41.204:13001)
Second Life RC Magnum 12.01.24.248357
Release Notes

CPU: Intel(R) Core(TM) i5 CPU M 540 @ 2.53GHz (2530 MHz)
Memory: 4096 MB
OS Version: Mac OS X 10.6.8 Darwin 10.8.0 Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 i386
Graphics Card Vendor: NVIDIA Corporation
Graphics Card: NVIDIA GeForce GT 330M OpenGL Engine

OpenGL Version: 2.1 NVIDIA-1.6.36
RestrainedLove API: (disabled)
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox 2.1.3010.6270

Settings mode: Hybrid
Viewer Skin: metaharper_sidetabs (default)
Font Used: Deja Vu (96)
Draw distance : 128
Bandwidth : 1000
LOD factor : 2
Built with GCC version 40201
Packets Lost: 1/2534 (0.0%)

Not sure if that might help everyone, but I had that vertex buffer error, you've mentioned lately, nearly every login.
And someone told me to install the phoenix appfix and the latest driver for my graphics card.
For me that worked. Never had a vertex buffer crash since installing these two things.

Have had this issue for quiet some time, returned to using Phoenix instead. Decided to return to Firestorm viewer. Did a clean reinstall of Firestorm. Worked for a couple of days, now I am unable to stay logged in. As soon as I land I get the following error:

ERROR: llimage/llimage.cpp(244):LLImageBase::getData: Bad memory allocation for the image buffer!

Firestorm 3.2.2 (24336) Nov 27 2011 17:05:37 (Firestorm-Release)
Release Notes

You are at 210,593.0, 331,996.0, 3,008.2 in Widus located at sim8651.agni.lindenlab.com (216.82.39.210:13002)
Second Life Server 12.01.06.247303
Release Notes

CPU: Intel(R) Core(TM) i3 CPU 530 @ 2.93GHz (2926.2 MHz)
Memory: 6072 MB
OS Version: Microsoft Windows 7 64-bit Service Pack 1 (Build 7601)
Graphics Card Vendor: ATI Technologies Inc.
Graphics Card: ATI Radeon HD 5450

Windows Graphics Driver Version: 8.17.0010.1065
OpenGL Version: 4.1.10524 Compatibility Profile Context
RestrainedLove API: RLV v2.7.0 / RLVa v1.4.3a
libcurl Version: libcurl/7.21.1 OpenSSL/0.9.8q zlib/1.2.5 c-ares/1.7.1
J2C Decoder Version: KDU
Audio Driver Version: FMOD version 3.750000
Qt Webkit Version: 4.7.1 (version number hard-coded)
Voice Server Version: Vivox 2.1.3010.6270

Settings mode: Phoenix
Viewer Skin: firestorm (grey)
Font Used: Deja Vu (96)
Draw distance : 96
Bandwidth : 4100
LOD factor : 2
Built with MSVC version 1600
Packets Lost: 112/9,132 (1.2%)

Believed to be substantially improved in 3.3.1

I have been getting this Fatal Error when I am in a crowded sim and I cam around. I've tried lowering my viewer texture memory buffer to 256 as suggested. Didn't help. I tried Changing my http texture setting to False. Oh my god, my inventory wouldn't load (I have around 34,000 items)! Crashed with that off anyway. This was a clean install. I crashed three times today! I was at dances each time so lots of people around. Something's gotta work! What else can I try until 3.3.0 gets released? Phoenix 1160??

Again:
Not sure if that might help everyone, but I had that vertex buffer error, you've mentioned lately, nearly every login.

And someone told me to install the

phoenix appfix

and the latest driver for my graphics card.

For me that worked. Never had a vertex buffer crash since installing these two things.

Is anyone still seeing this on 3.3.0.24880?

Howdy, Whirly. I installed 3.3.0 right after the Phoenix Hour yesterday. After that, I had 4 crash to desktops on my own sim with no one else around. No error messages, no warnings, no nothing. Just suddenly...KA-WHAM! Off it went. Yes I did a clean install, and yes my video card drivers are completely up to date.

I had a crash to desktop happen today. The logs/info are in "FS3_3_0 crash to desktop.zip. I had been running in the sim that originally caused the problem but today I was there for around 75 minutes before the crash happened (previously I would crash in under 10 minutes). As it was, my memory crept slowly up to around 1,580mb and then stayed around that for quite a while before going to 1620mb and crashing shortly after. Congratulations to the Firestorm and LL developers who made this release so much more stable.

@Marcus
If you keep getting these crashes, please file a seperate JIRA & attach your crash logs (instructions on where to find your logs on this page: http://wiki.phoenixviewer.com/doku.php?id=file_a_jira

@Richard
Thanks for the update & glad to hear things have improved. There is still some way to go to fix the memory bloating but it is much improved with the LL fixes already. They are still working on this problem.
As far as Im aware, the constant crashes right at login with this error message should not hapen at all now. But it will still happen when you normally max out on RAM, which on a 32 bit system will be around 1.6-1.7 Gb as youve observed.

This one is back after the merge up to the 3.1.1 V3 codebase.

In my opinion Nicky's VBO fixes were working much much better then the LL ones now in the post merge 4.1.1 builds.
Reopening & setting for 4.1.1 so this doesnt get forgotten about before next release.

Current implementation from LL V3.3.1 is still not the more comprehensive fix from the viewermaster repo.

Here's a location where I can reliable get this crash on post merge builds within about 10 mins of walking & caming around: http://maps.secondlife.com/secondlife/Ozza/13/43/79

Last lines of log file are always the same:

2012-05-05T09:43:46Z WARNING: llimage/llimage.cpp(216) : LLImageBase::allocateData: failed to allocate image data, size: 3145728 width: 1024 height: 1024
2012-05-05T09:43:46Z WARNING: llimage/llimage.cpp(216) : LLImageBase::allocateData: failed to allocate image data, size: 3145728 width: 1024 height: 1024
2012-05-05T09:43:46Z WARNING: llimage/llimage.cpp(216) : LLImageBase::allocateData: failed to allocate image data, size: 3145728 width: 1024 height: 1024
2012-05-05T09:43:47Z WARNING: llimage/llimage.cpp(216) : LLImageBase::allocateData: failed to allocate image data, size: 3145728 width: 1024 height: 1024
2012-05-05T09:43:47Z WARNING: llimage/llimage.cpp(216) : LLImageBase::allocateData: failed to allocate image data, size: 3145728 width: 1024 height: 1024
2012-05-05T09:43:47Z WARNING: llimage/llimage.cpp(263) : LLImageBase::getData: Bad memory allocation for the image buffer!
2012-05-05T09:43:48Z INFO: llcommon/llapp.cpp(956) : windows_post_minidump_callback: generated minidump: C:\Users\Kirstie\AppData\Roaming\Firestorm\logs\17fd48fe-7fb5-4e32-8ed8-dbb3312e8dca.dmp
2012-05-05T09:43:48Z INFO: llcommon/llapp.cpp(961) : windows_post_minidump_callback: Entering Windows Exception Handler...
2012-05-05T09:43:48Z INFO: newview/llappviewer.cpp(3351) : LLAppViewer::handleViewerCrash: Handle viewer crash entry.
2012-05-05T09:43:48Z INFO: newview/llappviewer.cpp(3353) : LLAppViewer::handleViewerCrash: Last render pool type: 9
2012-05-05T09:43:48Z INFO: llcommon/llmemory.cpp(165) : LLMemory::logMemoryInfo: Current allocated physical memory(KB): 1414380
2012-05-05T09:43:48Z INFO: llcommon/llmemory.cpp(166) : LLMemory::logMemoryInfo: Current allocated page size (KB): 1600664
2012-05-05T09:43:48Z INFO: llcommon/llmemory.cpp(167) : LLMemory::logMemoryInfo: Current availabe physical memory(KB): 263341
2012-05-05T09:43:48Z INFO: llcommon/llmemory.cpp(168) : LLMemory::logMemoryInfo: Current max usable memory(KB): 1677721
2012-05-05T09:43:48Z INFO: llcommon/llmemory.cpp(170) : LLMemory::logMemoryInfo: --- private pool information -- 
2012-05-05T09:43:48Z INFO: llcommon/llmemory.cpp(171) : LLMemory::logMemoryInfo: Total reserved (KB): 0
2012-05-05T09:43:48Z INFO: llcommon/llmemory.cpp(172) : LLMemory::logMemoryInfo: Total allocated (KB): 0
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1384) : LLError::LLCallStacks::print:  ************* PRINT OUT LL CALL STACKS ************* 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3519 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3489 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3435 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3344 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3339 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3328 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3519 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3489 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3435 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3344 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3339 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3328 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3519 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3489 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3435 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3344 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3339 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1388) : LLError::LLCallStacks::print: LLPipeline::postSort line: 3328 
2012-05-05T09:43:48Z INFO: llcommon/llerror.cpp(1390) : LLError::LLCallStacks::print:  *************** END OF LL CALL STACKS *************** 
2012-05-05T09:43:48Z INFO: newview/llappviewer.cpp(3478) : LLAppViewer::handleViewerCrash: Creating crash marker file C:\Users\Kirstie\AppData\Roaming\Firestorm\logs\Firestorm-Fizzlefire_4_1_1_27424.error_marker
2012-05-05T09:43:48Z INFO: newview/llappviewer.cpp(3484) : LLAppViewer::handleViewerCrash: Created crash marker file C:\Users\Kirstie\AppData\Roaming\Firestorm\logs\Firestorm-Fizzlefire_4_1_1_27424.error_marker
2012-05-05T09:43:48Z INFO: newview/llappviewer.cpp(3499) : LLAppViewer::handleViewerCrash: Handle viewer crash generating stats log.
2012-05-05T09:43:48Z INFO: newview/llappviewer.cpp(3220) : LLAppViewer::writeDebugInfo: Opening debug file C:\Users\Kirstie\AppData\Roaming\Firestorm\logs\debug_info.log

Callstack:

>	Firestorm-Fizzlefire.exe!transfer_bytes(unsigned char * dest=0x00000000, kdu_line_buf & src={...}, int gap=3, int precision=8)  Line 891	C++
 	Firestorm-Fizzlefire.exe!LLKDUDecodeState::processTileDecode(float decode_time=0.097525455, int limit_time=1)  Line 1043 + 0x18 bytes	C++
 	Firestorm-Fizzlefire.exe!LLImageJ2CKDU::decodeImpl(LLImageJ2C & base={...}, LLImageRaw & raw_image={...}, float decode_time=0.10000000, int first_channel=0, int max_channel_count=4)  Line 495 + 0x34 bytes	C++
 	Firestorm-Fizzlefire.exe!LLImageJ2C::decodeChannels(LLImageRaw * raw_imagep=0x680dfca0, float decode_time=0.10000000, int first_channel=0, int max_channel_count=4)  Line 189 + 0x29 bytes	C++
 	Firestorm-Fizzlefire.exe!LLImageJ2C::decode(LLImageRaw * raw_imagep=0x680dfca0, float decode_time=0.10000000)  Line 165	C++
 	Firestorm-Fizzlefire.exe!LLImageDecodeThread::ImageRequest::processRequest()  Line 146 + 0x1c bytes	C++
 	llcommon.dll!LLQueuedThread::processNextRequest()  Line 448	C++
 	llcommon.dll!LLQueuedThread::run()  Line 511	C++
 	llcommon.dll!LLThread::staticRun(apr_thread_t * apr_threadp=0x0793a058, void * datap=0x077d6000)  Line 91	C++
 	libapr-1.dll!76d84e22() 	
 	[Frames below may be incorrect and/or missing, no symbols loaded for libapr-1.dll]	
 	msvcr100.dll!_callthreadstartex()  Line 314 + 0x6 bytes	C
 	msvcr100.dll!_threadstartex(void * ptd=0x00727a88)  Line 292 + 0x5 bytes	C
 	kernel32.dll!@BaseThreadInitThunk@12()  + 0x12 bytes	
 	ntdll.dll!___RtlUserThreadStart@8()  + 0x27 bytes	
 	ntdll.dll!__RtlUserThreadStart@8()  + 0x1b bytes

Getting QA to test for this issue on the next smoketest.