keep failing to solve 'insufficient resources' problem in windows 7

[n13l5]

This problem is getting pretty old, and I've gone through a lot of searches and potential solutions already. Please read 'solutions tried' before answering...

Details about my systems.

1) The older computer: Samsung HDD, 6GB RAM, Core 2 Duo, running on Asus P5K mainboard. Four times as much software installed as on the new one,. Easily reaches one or two months of uptime without rebooting windows. Never runs out of resources.

2) The new computer: OCZ Vertex 4, i7-3770 CPU, GTX 670, 8GB CL8 RAM, Asus P8Z77i-Deluxe (not overclocked). This system has only a small subset of the software installed that resides on the old system.

Windows 7 Professional 64bit SP1 with updates and drivers current on both machines.

The Problem: After running the new machine for 30 minutes to an hour, it will fail even small tasks like copying a file with an "Error 1450: insufficient resources exist to complete..."

Causes pretty much ruled out so far:
- Installed Software other than drivers, since none of my software produces any such problem on my old machine
- Registry size - this Registry is half the size of the one on the old computer
- Virus - Kaspersky has a pretty good reputation and should find pretty much everything when run from outside windows. Windows Security Essentials, Windows Defender and Sphinx Software firewall are always active.
- Available hard drive space - 256GB windows drive is less than half full.
- RAM, Swapfile and CPU utilization: all 3 show well below 50% usage at any given time.
- Temperatures - the system runs perfectly cool, using Asus Fan Control, everything stays around 85-95F.

Solutions tried:
- 3 different Microsoft KB's regarding max handles (set to max)
- 1 Microsoft KB regarding max registry size (set to max)
- Tracking Handles with ProcessExplorer.exe
- Trackign Handles with handles.exe
- scouring system logs for errors (found nothing that gave me any clue)
- Uninstalling or disabling autorun for all non-essential software.
- Uninstalling nonessential mainboard drivers, Asus Utilities etc
- Tried different Nvidia drivers
- used Registry cleaner
- used System File Checker (SFC /scannow)
- ran Windows Security Essentials full system scan
- used Kaspersky Antivirus from its own boot disk with newest definitions
- Sysinternals Rootkit Revealer (I'm not really pro at evaluating the results on that)
- CC Cleaner
- Malwarebytes
- tested RAM with memtest
- manually underclocked RAM
- asked all my programmer friends...

They just told me the same Microsoft KB's that I had already found when googling my error.

Possibly interesting fact:
Using ProcessExplorer.exe or Handles.exe, once the problem becomes acute, will both fail with the message "Insufficient system resources to get handle information."

So, the very trouble shooting tools I try to use quit working.

I just tested it again: Since cold boot on this sytem, I have run nothing except internet explorer, opened this forum and have been typing this for ~15 minutes. I have 6 tabs open from my search. That's all! Still, as soon as I start up Process Explorer, it pops the error message. If I start it as the first thing, it will track handles ok until the system runs out of resources. But I don't really know what I'm looking for, I don't see any one process have way more handles than any of the other processes. Nothing looks conspicuously out of order.

If I ignore the fact that some programs start refusing to run (can't load DirectX after it runs out of resources, can't copy etc) and keep running the browser, maybe use word and excel (which oddly always start fine), it'll continue chugging along for some time still. It does not BSOD, it'll just hard freeze, so I have to hit reset.

Any of my 'bigger' Software including games will run fine, as long as I start it right after a reboot - even for hours on end. But once I close and try to start it again, no cigar, until reboot.

.
. .
Could this be a hardware problem of some sort? Or could one of Asus' or Intel's drivers be bad and leak resources like no tomorrow? Is there a better way for me to trouble shoot?

.
.
Any advice would be much appreciated, hope this is the right section of the forum...

 

[bassfisher6522]

Sounds like to me it's a corrupted file or registry. How long has the new system been up and running? If it were me, since you've done some extensive trouble shooting all ready, I would just re-do a clean install of the OS.

 

[n13l5]

It did it from when it was a brand new install..

Now its a month old, still the same... doing my hardest to avoid having to do a new install... that robs me of a week of my time to get everything back in place... I need to work

What good is an OS that can't be repaired but only be trashed.
SFC should have found any corrupt files.
I guess the registry is so hard to fix, programs like registry cleaner and its ilk can't be relied upon?


looking at the resource manager, I'm finding that my SSD is being continuously thrashed by writing wfpdiag.etl at 10kb/s...

Since 10 minutes, even more ceaseless thrashing is going on with MSStmp.log (250kb/s to 1000+kb/s and Windows.edb (~250kb/s) even while I'm just here writing this, and nothing much should be happening on the system.

 

[Night Hawk]

The most likely cause is a bad install of Windows. I ran into that back when adding a WD Green Power model 500gb drive in for the extra drive for RC testing on 7. Nothing would run right on that install where I finally decided to leave that one in for storage only until building the present case.

Any and all Windows installs whether XP,, Vista, or 7 beta to RC back then would run normally on the WD Caviar SE 500gb drives while that one particular drive removed from an external enclosure was problematic from the start. Since you are running an SSD the board's bios setting as far as the drive mode could also be other thing to be looking at. AHCI or ide?

 

[n13l5]

Thanks for the info... Your question reminds me of a couple of things that might be relevant to the issue:

The BIOS is set to AHCI, the SSD is hooked up to the first SATA-3 port on the mainboard. The drive has been NTFS formatted with a single partition prior to installing windows.

The UEFI BIOS is set to legacy mode, due to the fact that I had to manually select the drive to boot in UEFI mode every time, while it boots without intervention in legacy mode.

The computer has also been unable to hibernate or enter sleep mode from the day of installation.


Just remembered that I also have an unknown device in device manager, listed with hardware ID ACPI\INT\33A0. Windows can't find a driver for it and neither can Google, it seems. Some posts seem to think its a virtual Intel driver of some sort...

 

[Night Hawk]

You may be right about the need for an Intel driver. That seems to surround an Intel Z77 chipset if the search on that is correct. The search results ended up on a vendor's product/information page for an Asrock board which could be part of the problem while it is usually the drive mode or need for Sata/Raid type controller drivers that can stall the works.

When first getting into this build the board came set default for AHCI and Windows ran like crap at first until switching modes to the Native ide option and suddenly all ran well. Recently during the summer when looking into the AHCI guide here and switching the modes all drivers were already on and all went smooth as silk with both OS drive mounted as SCSI devices.

It sounds like the soluttion for you may be similar in fashion where you run things in one mode until everything is on and then switch over to see a normal running OS. That would be switching to ide and watching to see if performance suddenly presents itself and then going to the Intel support site to look things up there as well as at Asus.

 

[n13l5]

Ok, I changed the SATA settings in the BIOS from AHCI to IDE mode.

Rebooted, got BSOD right after the boot menu (couldn't read, cause the system immediately rebooted).
After trying it several times without change, I went back into the BIOS and set SATA back to AHCI mode.

System booted up again, I went to Device Manager to follow a suggestion I found to look for "PCI VEN & DEV numbers" for the Unknown Device, only to find that now, there was no unknown device anymore!

Still, the problem remains...


PS: I wrote this up twice, cause when the forum's token expires, it fails to post, and after clicking "back" as advised in the error message, you find that the forum has discarded all your text in the process.

 

[OldMX]

You screwed your system with those registry cleaners, get a fresh install to sort those problems, if possible run a couple of memtest loops to discard memory errors, and load bios defaults.

 

[n13l5]

I did run memtest, because some people suggested this issue could have to do with RAM problems. However, it came up with no problems. I've also tried under clocking the RAM, with no apparent change.

Further, this issue was certainly not caused by a registry cleaner.

I used a registry cleaner only after about every other strategy of fixing this failed. After its use, the issue didn't go away, but it didn't get any worse either.

This mainboard also came with VIRTU drivers/software. I removed that early on and didn't put it back, even though that removal didn't alleviate the problem either.

 

[Night Hawk]

Upon looking up the error message you are seeing I ran across one thread you may be interested in loooking over at the MS TechNet site. This is regards to the same error seen when someone was trying to install the Enterprise edition where the advice was to set a registry value to unlimited. Windows Modules Installer - Error 1450 Insufficient System Resources

The error message itself can also be seen with several other versions like XP or Exchange Server 2007 as another example and not pointing directly at anything wrong with the new build so far. "Insufficient system resources exist to complete the requested service" error message when you try to extend database files in an Exchange Server 2007 environment

 

[n13l5]

Thanks, Nighthawk, I had seen both of those and tried the solutions to no effect.

Seeing as things weren't getting better, I changed my plans from waiting with Windows 8 for at least a year, to just buying an upgrade, hoping it might fix the problem.

After installing Windows 8, I'm not only dealing with a sketchy, half baked interface, but also with the fact that the same problem is still there.

This made me think of hardware once more, and so I ran Windows 8's Boot-time Memtest instead of the 3rd party test I had used before. There's 3 flavors of that: Basic, Standard and Extensive. It absolved the Standard set of tests several times without trouble. Then I thought: lets be triple sure, and set it to repeat on the Extensive test set. Guess what: on the first run, it got to 21% and then the computer froze solid. I was not sitting impatiently in front of it, I actually went into the city to take care of several errands and ate at a restaurant, so it hadn't moved from 21% in 3 hours, and the memtest interface elements were unresponsive. This is before windows boots...

Here's hoping this gives some of you an idea what might be wrong with my RAM compatibility or settings or if it might plainly be bad RAM. Its currently not overclocked, its actually running only at 1333 MHz, rather than the 1600 MHz its rated for.

What I'm using is Crucial (Micron) "Ballistix" 2x 4GB DDR3-1600 CL8 1.5V UDIMM W/XMP. (I specifically picked those for the CL8 latency rating)

Should I be getting an RMA for these, or should I test some other BIOS settings...?

 

[Night Hawk]

When seeing the DDR3 1600 running at 1333 it reminds me all too well about first getting into 1.9v memory on a board that takes 1.5v! The Gigabyte memory supported listed indicated the 1.9v OCZ Gold while the user manual once the board was in told a different story!

Besides the higher voltage backclocking the OCZ down the memory clock in the bios had to be raised up for the present 1.5v 1600 memory. But even with the first set of OCZ dimms there wasn't any problems as far as the clean 7 install on the then new case at the time except the need to change from AHCI to Native ide due to Windows booting way too slow for the norm!

I can't see where you would be seeing any errors from the slight backclocking there either. Timigs not set right or not compatible still wouldn't explain much either while the wrong type of memory for the board would! If that model board as seen previously at times with Asus does not take performance memory there's half the problem. I haven't run into that one for several years now but it is one known issue that can be a cause for concern especially with memtest hanging on you.

 

[n13l5]

Thanks again, I will contact Asus Support in regards to that board and the RAM I chose, letting them know that it finishes basic and standard tests without trouble, but hangs on extended...

 

[n13l5]

It turned out to be much quicker to contact Micron than Asus.

Micron Support checked their RAM modules against my Asus MB and said 100% compatible, should require no special BIOS settings.

They told me to use Memtest86 from a bootable USB stick to test the memory.

I did that, and it ran for many hours without generating a single error...

 

[Night Hawk]

I suspect then that the W8 flavor of memtest is simply hanging on you while the "3rd party" memtest is the actual original version you download and see written to a flash drive. Apparently the included form on the Windows dvd or in the download is a bit buggy.

The good news is for you however if Micron confirmed the memory is suitable. The memory clock is set low however for you to be seeing 1333 which is a fast trip into the bios to bring the clock up a notch to see the 1600mhz speed there.

That has nothing to do with the insufficient resources error or compatibility but simply seeing the speed the memory is designed to run at. The model board you have supports DDR3 2400(oced) ruling out any need to worry about simply raising that up one notch.

The only other possible concern would be the note at the Asus product page there about the "Hyper Dimm support is subject to the physical characteristics of individual cpus". I just have to wonder if the cpu and memory combination is where the problem is while the board would obviously support each on it's own without any problems seen.

It wouldn't be much any fault to find with any hardware. But with certain Intel model cpus some have run into problems with a select few in particular. That is usually a problem between cpu and chipset however while it was other thing that could explain the errors you have been seeing unless the board itself has a hidden fault or problem with the bios itself.

First however you will need to look at the things you now know you can rule out like bad ram for a start. Hardwares and drives in good working order, etc.

You can try running the System File Checker with the sfc /scannow command just in case of a corrupted system file being at fault. It's going to turn out to be one of two things however either being an OS or hardware problem despite not being able to pinpoint one hardware in particular or an OS problem other then that error.

 

[n13l5]

omg I lost a full page of text again, just cause my "token has expired"

I understand that you don't allow a post if my token is expired, but do you really have to clear my text?

In other forums, if there is an error with posting, when I click the back button, my text is always still there in the editing field!

 

[Brink]

Hello 13l5,

When posting something that takes a while to do, I find it best to have two tabs open with the site, refresh the other tab before posting to make sure you are no longer timed out, then post in the other tab.