Lots of free memory, but still swapping
Kaare Digernes
with 1 GB of memory. The swap file is set to 1522 MB (min and max) and
I've run PerfectDisk 6.0 to defragment the disk (including system files,
i.e. swap files, registry, etc).
The problem I'm seeing is that the system regularly 'locks up' when I
minimize or switch between programs I use frequently such as Mozilla
Thunderbird, Mozilla Thunderbird, Palm Desktop and the Adobe Acrobat
Reader to name a few. By 'locks up' I mean that the disk works
constantly for 4-6 seconds and effectively prevents anything else from
happening -- the screen is not refreshing, WinAmp stops playing, the
performance monitor misses samples, etc.
The Windows task manager states that only 300 MB or so of memory is in
use and that 720 MB or so is free. Whenever the system locks up the
performance monitor reports that the 'pages/sec' counter goes up from
zero to between 280 and 480.
Does anyone know the reason I'm seeing this behavior? Given that the
amount of free memory is twice the amount used, I was expecting the
system to not swap at all. I also tried setting the swap file to the
minimum size of 2 MB, but Windows pretty much immediately complained
that the file was too small and increased the size automatically.
Is there any way to prevent the system from swapping? For instance, can
I get rid of the swap file altogether, like I can on Linux? Any advice,
theories, things to try, etc, is greatly appreciated!
Thanks,
-Kaare
Dan G
number of things, like logging. Opening a program would normally produce a
small amount of it.
Do not confuse paging operations with paging file use.
"Kaare Digernes" <kaa...@gmail.com> wrote in message
news:Y%IEc.52058$aJ3....@nwrdny02.gnilink.net...
Tony Hwang
What is your file system? FAT or NTFS?
Tony
Kaare Digernes
It's NTFS and there's only one partition on the drive. The block size is
512 bytes.
-Kaare
Kaare Digernes
this delay. Even if the event was significant, Windows should be able to
postpone the writing to disk (i.e. lazy write) and not force an
immediate write. Furthermore, 4-6 seconds for logging is just excessive.
I wonder if Windows is trying to free up (page to disk) as much working
memory as possible even though there's still plenty to use? It's weird.
Is there anything else I should be watching in the performance monitor
to get a more nuanced view on what's going on?
-Kaare
Tony Hwang
Hi,
Your HD may be sick. How about running some diagnostics?
Tony
Dan G
"format and install" situations. Ruling out hardware issues is your first
step. A reinstall would "rule-in" hardware.
"Tony Hwang" <drag...@shaw.ca> wrote in message
news:%RKEc.916633$Pk3.324120@pd7tw1no...
Gary V. Deutschmann, Sr.
system operating files or placed files that override the normal system
files.
We won't allow anything made by USRobotics (or whatever name they are
hiding under this month) in any of our computers. The problems and
continuous headaches are just not worth touching anything they touch.
TTUL
Gary
Gunrunnerjohn
>It's NTFS and there's only one partition on the drive. The block size is
>512 bytes.
>
>-Kaare
I find it odd the block size is 512, you must have converted this partition from
FAT32. Performance of the disksystem will suck unless you use something like
Partition Magic to make the blocksize the standard 4kb...
Tony
Deutschmann, Sr.) wrote:
When I used dial-up, I found USRobotics to be best. Maybe they've
changed.
>TTUL
>Gary
Jim Berwick
> When I used dial-up, I found USRobotics to be best. Maybe they've
> changed.
I think I rarely have a customer with a USR modem complain about line noise
issues or poor connections.
Kaare Digernes
utility I'm using (HDD Health) says everything's fine, but I have a
spare drive that I can pop in and try. I'll also try bumping up the
block size from 512 bytes to 4 KB (as gunrunnerjohn pointed out). Stay
tuned. :-)
Thanks, guys!
-Kaare
Kaare Digernes
but I'll definitely try experimenting with a block size of 4 KB or
higher. Thanks for pointing this out!
-Kaare
Kaare Digernes
I agree. I swear by US Robotics Courier modems, although it's been quite
a few years since the last time I had to use a modem for anything.
Thankfully. :-) I know the Sportster modems were different, though they
were priced way cheaper (budget model).
-Kaare
Skywise
Every modem I've ever had, right up to my current 4 year old Sporster
56k fax/modem has been a USR and I've never ever ever ever had a
problem. Wait, my very first modem wasn't a USR and it crapped out on
me. I think it was a Zoom modem internal 2400, IIRC.
Brian
--
Sed quis custodiet ipsos Custodes?
Gary V. Deutschmann, Sr.
Maybe you were lucky and got one of their consumer modems that worked.
We went through 7 of them and NONE of them worked, but at least
something different was wrong with each one.
By the way, the return shipping for replacement was on our backs, NOT
USRobitics. So we had over $185.00 tied up in return shipping charges
alone and NEVER got a modem that worked at all.
My wife bought a Palm Pilot (3-Com which is also a USRobitics Alias),
it changed files in the computers which rendered some needed
accounting functions inoperable and none of our form printers worked
until we cleaned the hard drive and reinstalled the OS and basically
started over from scratch.
I do realize that MOST ISP's use USRobotics commercial grade modems
and have no trouble with them. But that has no bearing on what they
manufacture for the consumer market.
Of all the modems I've used over the years, everything from the
cheapest no name discount import to the top of the line of the
consumer units, ONLY USRobotics consumer equipment has given us MAJOR
headaches and COST US MONEY.
Reputable companies will send a call tag or pay for return shipping
for a product that has manufacturer defects!
USRobotics (3-Com) is NOT one of these reputable companies!
TTUL
Gary
Tony verbositized:
Tony
Deutschmann, Sr.) wrote:
The best modem I ever had was a USR, and it was one of the
lower-priced ones they used to call "Sportster". I've had several
others before (beginning about 1990) and the USR ones have always been
good. The other brands were slower andfailed more often (I've had one
TI that was good too).
Gary V. Deutschmann, Sr.
I'm glad to hear SOMEONE got something from them that worked and were
NOT BURNED.
I am surprised to hear you had so many other modems fail though!
I usually have 4 to 6 computers at home (business from home) and my
brother has over 25 computers in his offices. NONE of them are USR
and NONE of them have ever failed or caused problems and ALL are low
end discount priced modems.
Of course we upgraded from 1200 baud to 14k to 56k in steps as the
technology became available to do so. But none of the modems were
replaced because of failure or erratic behavior.
In the case of USR, we NEVER got a single one that worked, even from
the beginning. There was always a feature of the modem that did not
work at all.
A lot of people buy a modem plug it in and say it works, without ever
testing the features of the modem. If it's an answer/fax/modem and
the answer part don't work, the unit is defective, even if the modem
part itself functions.
You say your USR modem works, have you tried all the functions it is
supposed to do to see for sure that it works?
TTUL
Gary
Kaare Digernes
bytes to 4 KB using Partition Magic and the problem went away. Now there
is no delay at all when minimizing any windows and no signs of swapping.
Thanks to everybody pitching in offering advice, especially
gunrunnerjohn! Thanks, guys, it's much appreciated! :-)
-Kaare
Kylesb
message news:40e6c11b....@news.galilei.com...
This continues the off-topic comments, but compelled to reply I am.
Over the years, I've used a good variety of modems, and it seems each
has it's own little "gotchas". Only 1 or 2 has worked as promised,
and Boca and Supra are now out of the business.
Now that I'm upgrading to all PCI machines, I have to discard my old
tried and true reliable ISA voice/faxmodems. I tried a couple of new
"cheapie" softmodems, both have their unique problems. Getting a
voice/faxmodem that works across the board in all categories of my
expectations (or their promises) is proving to be very difficult. I
recently tried an Agere/Lucent that worked nicely, once I overcame the
line impedance mismatch problem that caused a "no dial tone" error (I
just used a dial delay, then the dial tone was detected).
Unfortunately, the "silence deletion" feature of the driver is
non-functional for voicemail operation. Tried a newer Intel chip
based modem, good voice operation, but the callerid won't work. So
guess what I'm saying is I know where Gary's coming from on his
comments. As to internet connections with modems, all the new one's
seem to work comparably, even the cheapie models. The $12 agere modem
gets 50k-52k connects reliably, but I want a voice/faxmodem that works
nicely for voice and fax operation, since I now have broadband, I
don't even care about the fastest connects, lol.
--
Best regards,
Kyle
Tony
I've always found the best modems to be external ones (RS232). One
advantage is a physical way to break the connection. Also, modems are
more susceptible to surge damage (through the phoneline), an external
modem helps to protect your computer.
Note that I have had 3 instances of a computer component fail
permanently during a thunderstorm. All were modems.
If your computer has only USB, there's serial adapters.
> I tried a couple of new
>"cheapie" softmodems, both have their unique problems.
Yes, those HSP modems do have problems.
w_tom
better. But I know better. Same is true of external modems offering
better protection or that destructive surges enter on phone lines.
The last is myth because protectors are installed by the phone
company, for free, on phone lines. But AC electric - the most common
source of transistor damage - makes a direct connection to the modem.
A connection that even bypasses computer power supply. This having
been demonstrated repeatedly by repairing lightning damaged modems.
This modem was lightning damaged and has now worked for years without
failure.
This modem discussion being irrelevant to the original post but was
discussed extensively elsewhere:
"RJ-11 line protection?" on 31 Dec 2003 in pdx.computing at
http://tinyurl.com/2hl53
"strange problem after power surge/thunderstorm" in comp.dcom.modems
on 31 Mar 2003 at
http://tinyurl.com/2gumt
Tony wrote in message news:<jpqie0ponsaa83c64...@4ax.com>...
Tony
On 5 Jul 2004 16:40:31 -0700, w_t...@usa.net (w_tom) wrote:
>I also find that when I wax a car, then the engine appears to run
>better. But I know better. Same is true of external modems offering
>better protection or that destructive surges enter on phone lines.
>The last is myth because protectors are installed by the phone
>company, for free, on phone lines. But AC electric - the most common
>source of transistor damage - makes a direct connection to the modem.
>A connection that even bypasses computer power supply. This having
>been demonstrated repeatedly by repairing lightning damaged modems.
What component did you replace, and how do you think it failed?
>This modem was lightning damaged and has now worked for years without
>failure.
>
You will have AC electric connected to your computer, whatever type
modem you have (or with no modem at all), so it is not relevant to
which (internal or external modem) leads to more surge risk. If you
have an internal moden, you have an ADDITIONAL path where a surge can
enter.
There are lightning protectors installed on phone lines, which help
stop major surges (such as those that can start fires). It does not
block small surges which can damage electronic equipmemt. No matter
how well protected that phone line is, it still adds some surge risk.
> This modem discussion being irrelevant to the original post but was
>discussed extensively elsewhere:
>"RJ-11 line protection?" on 31 Dec 2003 in pdx.computing at
> http://tinyurl.com/2hl53
> "strange problem after power surge/thunderstorm" in comp.dcom.modems
>on 31 Mar 2003 at
> http://tinyurl.com/2gumt
>
There is some important stuff in those (such as the need for a ground
connection for surge protection).
Gunrunnerjohn
Did you ever wonder why you are cautioned not to use a wired phone during a
thunderstorm?
Hint: It's not because the lightning can reach your ear through the electrical
lines!
In point of fact, by far the most common point of failure of an analog modem is
the line interface, not the power supply.
Oh, BTW, my car does run better after I wax it. :-)
On 5 Jul 2004 16:40:31 -0700, w_t...@usa.net (w_tom) wrote:
w_tom
found in adjacent (and grossly overpriced) plug-in protectors.
Internal protection so resilient as to also make small surges
irrelevant.
> It does not block small surges which can damage electronic equipmemt
Again, effective surge protectors do not block surges. Effective
protectors only connect all wires together during the transient. That
is what protectors do - shunt, divert, connect. Protectors only
block, stop, filter, or absorb surges where myths are promoted.
Effective protectors shunt (divert, connect) the destructive transient
less than 10 feet to earth ground - which plug-in protectors cannot
do. Same reason why direct lightning strikes did not damage in the
1930s. And yet, 70+ years later, we still have people posting that a
protector *blocks* surges. Only scam protector manufacturers would
make that claim.
Have been fixing lightning damaged equipment maybe for longer than
some posters have even existed. On modems, a most typically damaged
part is the driver transistor for the off hook relay. This results in
the "No Dialtone Detected" error message. This is also discussed in
those previously cited posts. No more reason to discuss what was
answered in detail in that discussion - including no difference
between internal and external modems.
IOW damage is most often in the modems DAA section. Makes no
difference whether modem is internal or external. The incoming surge
is typically incoming on AC mains and outgoing (via that DAA section)
on phone lines. Furthermore, I have examples where (by following the
surge path and damaged components) the adjacent (plug-in) protector
completed (formed) the surge destructive circuit. That is correct.
The adjacent protector made it easier for a surge to damage multiple
powered off computers. Why? Shunt mode protectors do not stop,
block, or absorb surges despite the promoted myths.
Adjacent protector is ineffective for too many reasons. But the
most damning reason is that IF those $0.10 parts inside the $10 or $50
protector were effective, then those same $0.10 parts would already be
inside computer and modem. Once they were installed. Find those same
parts on some Apple II computers. Today, manufacturers no longer
install those parts because they were not effective: too close to
transistors and too far from earth ground means ineffective
protection. So please, someone say why those same parts will suddenly
be effective inside the adjacent power strip protector? Why would
those parts not be effective inside the appliance and yet be effective
only 5 feet down the power cord? Just part of the myths that claim a
protector will block, stop, or absorb the transient.
External modems and internal modems suffer damage for the same
reason. Most typical reason is a direct connection from AC mains to
modem ICs either via the motherboard bus or via that RS-232 cable -
the incoming path. Outgoing to earth ground via phone line. After a
surge is carried on everything in that circuit, only then does one
component in that circuit fail - often in the modem's DAA section.
Modems and computers are so resilient because they already have
internal protection. Modem galvanic isolation can be rated as high as
1500 volts. But that internal protection *assumes* destructive
transients will be earthed before entering the building. If not - if
the human has failed to properly earth all incoming utilities to the
same point - then internal protection can be overwhelemed. That
internal protection is effective - but only if every incoming utility
is either hardwired to earth or earthed via a 'whole house' protector.
Notice the word that defines protection - earthing.
Adjacent protector can even be damaged by surges too small to
overwhelm appliance internal protection. And yet that too is
perverted into another myth - "the surge protector sacrificed itself
to save my computer". In reality the protector was ineffective for
just another reason - grossly undersized as well as overpriced.
Plug-in protectors are promoted by myths such as 'my protector
blocks small surges'. The manufacturer does not even make that claim!
A surge protector is only as effective as its earth ground. Any
protector that does not discuss or provide earthing - ineffective and
yet still promoted by myths. Myths such as a protector will stop,
block, or absorb surges.
Tony wrote in message news:<bh3ke0p0pqjle33vc...@4ax.com>...
Tony
>All appliances have internal protection that is superior to anything
>found in adjacent (and grossly overpriced) plug-in protectors.
>Internal protection so resilient as to also make small surges
>irrelevant.
>
>> It does not block small surges which can damage electronic equipmemt
>
> Again, effective surge protectors do not block surges. Effective
>protectors only connect all wires together during the transient. That
>is what protectors do - shunt, divert, connect. Protectors only
>block, stop, filter, or absorb surges where myths are promoted.
>Effective protectors shunt (divert, connect) the destructive transient
>less than 10 feet to earth ground -
Which has the effect of blocking it from the connected equipment. The
same was arresting people who try to bring bombs into public buildings
has the effect of blocking bombs from those buildings.
and, yes, you do need a ground connection to the surge protector.
>which plug-in protectors cannot
>do. Same reason why direct lightning strikes did not damage in the
>1930s. And yet, 70+ years later, we still have people posting that a
>protector *blocks* surges. Only scam protector manufacturers would
>make that claim.
>
> Have been fixing lightning damaged equipment maybe for longer than
>some posters have even existed. On modems, a most typically damaged
>part is the driver transistor for the off hook relay. This results in
>the "No Dialtone Detected" error message. This is also discussed in
>those previously cited posts. No more reason to discuss what was
>answered in detail in that discussion - including no difference
>between internal and external modems.
>
> IOW damage is most often in the modems DAA section. Makes no
>difference whether modem is internal or external.
I wasn't saying anything about THAT difference, but was considering
the possability of the surge spreading to the rest of the computer
system (VERY close for an internal modem).
> The incoming surge
>is typically incoming on AC mains and outgoing (via that DAA section)
>on phone lines. Furthermore, I have examples where (by following the
>surge path and damaged components) the adjacent (plug-in) protector
>completed (formed) the surge destructive circuit. That is correct.
>The adjacent protector made it easier for a surge to damage multiple
>powered off computers. Why? Shunt mode protectors do not stop,
>block, or absorb surges despite the promoted myths.
>
The surge is diverted to ground, so there's some other place it's NOT
GOING (where it's blocked). Think about that.
> Adjacent protector is ineffective for too many reasons. But the
>most damning reason is that IF those $0.10 parts inside the $10 or $50
>protector were effective, then those same $0.10 parts would already be
>inside computer and modem. Once they were installed. Find those same
>parts on some Apple II computers. Today, manufacturers no longer
>install those parts because they were not effective: too close to
>transistors and too far from earth ground means ineffective
>protection. So please, someone say why those same parts will suddenly
>be effective inside the adjacent power strip protector? Why would
>those parts not be effective inside the appliance and yet be effective
>only 5 feet down the power cord? Just part of the myths that claim a
>protector will block, stop, or absorb the transient.
>
> External modems and internal modems suffer damage for the same
>reason. Most typical reason is a direct connection from AC mains to
>modem ICs either via the motherboard bus or via that RS-232 cable -
>the incoming path.
What? Power usually goes through a transformer, where it's
electrically isolated. No direct path.
> Outgoing to earth ground via phone line. After a
>surge is carried on everything in that circuit, only then does one
>component in that circuit fail - often in the modem's DAA section.
>
The modems I've had damaged, it usually gets the microcontroller.
> Modems and computers are so resilient because they already have
>internal protection. Modem galvanic isolation can be rated as high as
>1500 volts. But that internal protection *assumes* destructive
>transients will be earthed before entering the building. If not - if
>the human has failed to properly earth all incoming utilities to the
>same point - then internal protection can be overwhelemed. That
>internal protection is effective - but only if every incoming utility
>is either hardwired to earth or earthed via a 'whole house' protector.
> Notice the word that defines protection - earthing.
>
> Adjacent protector can even be damaged by surges too small to
>overwhelm appliance internal protection. And yet that too is
>perverted into another myth - "the surge protector sacrificed itself
>to save my computer". In reality the protector was ineffective for
>just another reason - grossly undersized as well as overpriced.
>
> Plug-in protectors are promoted by myths such as 'my protector
>blocks small surges'. The manufacturer does not even make that claim!
> A surge protector is only as effective as its earth ground.
There's also transients between hot and neutral. No ground involved.
Ronnie
wrote:
>I'm amused at the mis-information you managed to post.
>
>Did you ever wonder why you are cautioned not to use a wired phone during a
>thunderstorm?
>
>Hint: It's not because the lightning can reach your ear through the electrical
>lines!
>
>In point of fact, by far the most common point of failure of an analog modem is
>the line interface, not the power supply.
>
>Oh, BTW, my car does run better after I wax it. :-)
>
Mine does too. :-)
Gary V. Deutschmann, Sr.
I have a rather large (like 300 lbs) Post-Glover line isolation
transformer and all lines into the room go through heavily grounded
gas bottles similar to (Transi-Traps).
During a storm, when most folks are unplugging, we are powering up
everthing we have to handle what emergencies might arise.
Even after a direct lightning hit that entered through an abandoned
wire bundle, took out a couple of file cabinets and arced to one of
the power supplies, we suffered no damage to any of the equipment or
computers.
We had a few monitors that changed color until they were degaussed,
but other than that, all was fine.
It's great that you posted the information you did for all to see!
TTUL
Gary
w_tom
remove their headsets so during thunderstorms. That commerical radio
and TV engineers abandon the equipment during thunderstorms. Then he
accused me of misinformation?
Gunrunnerjohn is first encouraged to reread what was posted. No
where is there any suggestion that the power supply was a point of
failure OR that the power supply is even in the destructive path from
AC mains, destructively through modem, and outgoing to earth ground
via phone line. Yes I did not specifically cite the wire in that
circuit of damage. But if Gunrunnerjohn would first learn how his
computer is wired, then the modem destructive circuit would be
obvious. Hint: one prong on the AC electric plug makes a direct
connection to some IC pins on the modem - internal or external.
Gunrunnerjohn is encouraged to first use his digital multimeter to
find this circuit. That is if he even has basic electrical equipment
such as the multimeter. He better if he knows so much more.
The only place that humans are encouraged not to use the phone
during T-storms is where technically naive humans don't even
understand importance of earthing. This is where those same humans
disconnect essential ground wire saying, "My phone works just fine
without all this crap." So yes, because so many humans don't even
comprehend what I have posted, then it is simply better for the
telephone company to ask everyone to hang up during T-storms.
In the meantime I am always using my computers during T-storms -
especially to follow the progress of such storms. I need not worry
about such dangers because, well, the reasons were posted previously.
Same reasons that Gunrunnerjohn still does not comprehend. Even
though the concepts were well proven over 60 years ago, there are
still today many who deny those concepts.
In the meantime, those 911 operators and commercial broadcasters
need not abandon headsets during thunderstorms. Their equipment is
properly earthed.
Gunrunnerjohn <nob...@world.com> wrote in message news:<u69le0pg5i1p2tmt6...@4ax.com>...