My pictures went blank (out of memory?)

I was gettinginto a project and was told by an ae message that I was running out of memory. Now most of my imported imeges are white. thank goodness the properties and animations I set are still there (I can see the frames). How do I get the images back without getting more memory. Can I proxy them all? I tried using half and third resolution but that didnt work. I am guessing the answer will be to cut up th eproject in pieces but any suggestions that will help me avoid that would be great!

dansch37 wrote:
it was a blessing in disguise because I cleaned up the document very well to troubleshoot.
Hehe, yes, sometimes we all have to be reminded "to bring out the garbage" Glad it worked out and was in fact so simple to fix.
Mylenium

Similar Messages

Out of memory? i have deleted programs and recordings and no changes are recognized in memory

I cannot take pictures because the "out of memory" comes up on the built in app. I can however take pics with hiptsamatic. What is going on with this memory. I deleted several apps with 100 megs and the memory numbers do not change or adjust downward. I ahve restarted the device and still no changes. Any ideas ?

What is your final product? And do these 28 sequence need to be all in the same timeline?
Try again with a new project and load the in one at at time. Give Premiere a chance to index etc if neccessary before loading up the next one.

HT201317 dear, i have a question please. if i want to keep a folder of pictures in my icloud account; then delete this folder off my phone would it be deleted from icloud account too?? cause im running out of memory on my iphone

dear, i have a question please. if i want to keep a folder of pictures in my icloud account; then delete this folder off my phone would it be deleted from icloud account too?? cause im running out of memory on my iphone
thanks

You can't use icloud to store photos. There's a service called photo stream which is used to sync photos from one device to another, but this can't be used to store photos permanently.
Always sync photos from a device to your computer, Then you can delete them from the device.
Read this... import photos to your computer...
http://support.apple.com/kb/HT4083

Out of Memory message -- even with a blank sequence!

Not sure what happened, but all of a sudden I can't get any timelines to open. Even if I just start up the program without any open sequences, when I click on the empty "Sequence 1" I get the dreaded Error: Out of Memory message. I've tried the usual fixes, but nothing seems to help. I recently ran Cache Out (to delete various caches) and Monolingual (to get rid of non-English language support files), I wonder if something essential got trashed?

Got 10GB or so on both my external hard drive (which I'm not even using now), as well as the internal one. My media is on the external, but to simplify I disengaged it and am just trying to load ANYTHING. Just working on my internal drive, which I have permission to use, of course. I've got 1.25GB of RAM, no issue there. Activity Monitor was a good idea, in fact it occurred to me a few minutes ago and I checked that, didn't see any obvious problems.
I trashed the preferences numerous times, deleted some old files, was going to dump the render files -- but I realized it's irrelevant, since it won't even open a blank file. Here's further evidence of the weirdness -- when the blank project opens up, it gives me the Browser window, and I can get the Viewer to open (but never the Timeline). However, if I try to open up anything in the Viewer, it immediately disappears! I'm losing my mind...

Mapviewer out-of-memory A3-size pictures

I'm getting a Java.lang.out-of-memory error when I try to generate a A3 size picture using 300 dpi. A4 size pictures with 300 dpi are no problem. This is with Mapviewer version 1012. I'm not getting this error on another PC with the Mapviewer version before 1012.
I tried to enlarge the cache size to 128 Mb but this doesn't help.

More info:
The PC where the error occurs has Windows XP, 1 Gb memory and has an Oracle 9.2 database running. The other PC without the error has Windows 2000, 256 Mb memory but no Oracle database running.
I use OC4J standalone on both PC's.

Somehow my pictures in the iPhoto library are blanked out

I have just been updated my Mac and the iphoto to 11 v. 9.2. I have switched to iCloud and is now using photostream. I had a lot of pics I started streaming and during I was editing the pics. from our vaction. Suddenly I'm told that the iPhoto library has an error and needs to be updated. I let the MAC do so, and then suddenly all my pictures in my iPhoto are blanked out. At this time all my 1000 pics are still not all streamed, but iPhoto it is still streaming though I can't see any pictures in the photolibrary??? I can only see empty squares where the pics were supposed to be. I then check my iPhone and I see that pictures at still streamed from iPhoto. Then I search for my pics in "Finder" and it seems they are still there however I would like to get them back in the same order as before?? Also I made a half made photobook, but the pictures in the book are also blanked out? Can anyone help?

Option 1
Back Up and try rebuild the library: hold down the command and option (or alt) keys while launching iPhoto. Use the resulting dialogue to rebuild. Choose to Rebuild iPhoto Library Database from automatic backup.
If that fails:
Option 2
Download iPhoto Library Manager and use its rebuild function. This will create a new library based on data in the albumdata.xml file. Not everything will be brought over - no slideshows, books or calendars, for instance - but it should get all your albums and keywords back.
Because this process creates an entirely new library and leaves your old one untouched, it is non-destructive, and if you're not happy with the results you can simply return to your old one. .
Regards
TD

How to blank out part of a picture in LR 3?

Update: I found that the clone tool can do what I want. If there is a better way though, I would be interested.
I have a picture I want to post in a public on-line forum, but there are some cars in the scene and the license plate numbers are clearly visibile. I don't know whose cars they are, but I doubt they want their license plate numbers posted in a public forum. So I would like to blank them out before posting the picture on-line.
I used the LR 3 adjustment brush with the exposure set to zero and brushed over the plate numbers. And indeed it did it did turn them black. But just to be sure they detail couldn't be recovered, I then increased the exposure for the entire picture and found that the numbers because visibile again as the overall exposure increased.
Is there a way in LR 3 to blank out these numbers in a way that no one could use a photo editing program to recover the details in the blanked out area? If it matters, the picture is a .jpg, not RAW.

Thanks. I just tried your suggestion. However, after exporting I used another editing program on the exported photo, increased the overall brightness, and could read the numbers. At normal brightness they appear blacked out, like they did at the time of exporting them from LR.
I had the brush density and flow both at 100, so I am not sure whether there are other settings I need to change to make it work as intended. It would be easier to use the brush than to use cloning. But then this is not something I do often (this is the first time, actually), so any method that works is OK. But it would be good to learn how to use the brush to do this, just for my own LR learning purposes.
Thanks again for your suggestion.

HT1267 I plugged my iPhone 4 into my computer with the USB charger cable, and the screen went blank and came up with a picture of the usb cable with an arrow pointing to iTunes....Can anyone help get this fixed????

I plugged my iPhone 4 into my computer with the USB charger cable, and the screen went blank and came up with a picture of the usb cable with an arrow pointing to iTunes....Can anyone help get this fixed????

Connect your iPhone to iTunes and restore or follow the prompts.

My iphone went blank and when I plugged it into outlet the screen shows the itunes logo and a picture of what looks like the charger. I tried to reset it and it does not respond. Help!

My iphone went blank and when I plugged it into outlet the screen shows the itunes logo and a picture of what looks like the charger. I tried to reset it and it does not respond. Can anyone help?

That screen means it has gone into recovery mode. You need to connect to a computer running iTunes by following the instructions in http://support.apple.com/kb/HT1808 for the procedure of recovery mode connection.

Picture goes blank when new grad filter or brush used.

When I attempt to use a new grad filter or adjustment brush in LR3.6 the picture goes blank i.e. totally black. The history does not update and I can do no other edits until I Ctrl-Z or go to the grid pane or move to the next image. I am running on Windows 7 64 bit and I have re-installed, twice, (repaired) using the latest downloaded LR3.6 off the adobe site. I have done several integity checks which are successful and the problem occurs no matter what catalog I open, small or big. However, my LR 4 beta works perfectly!? This problem came out of the blue and the only software that I have installed on my machine recently was some Windows updates and Firetower which checks changes to my Startup processes.
Alan

I have experienced this on my old machine running Win XP with 4 GB RAM except that my picture went totally white, not black. I think it happens when LR demands more memory than what is currently available.
To alleviate the problem you could try the following:
- Increase your LR Cache size to at least 25 GB (by default it is set to 1 GB). Set Cache size in >Edit >Preferences >File handling tab.
- Increase your page file, and - if possible - set your page file (a) on a different physical drive than where your OS resides, (b) on a different physical drive where your LR catalog resides.
See also here for optimizing LR performance: http://kb2.adobe.com/cps/400/kb400808.html
WW

Out of Memory ODBC - 100

Dear Experts,
I have created a Query Print Layout to generate a picture report. When i run the report through manager id the report gets generated, on the other hand when i run the Query Print Layout from user/ id it throws the following error Out of Memory ODBC - 100. I tried checking the user giving super user rights but it did not generate report. The report screen is blank. Could you be kind enough to let me know any possible solution for the same.
Saravanan
Edited by: Sanbrahma39 on Jan 18, 2010 5:35 AM
Edited by: Sanbrahma39 on Jan 18, 2010 5:38 AM

Dear Rahul,
I worked out your suggestion like dropping the SP from the SQL/FMS and try running the report. Again i face the same issue, when i am generating the report with manager user id the report is perfectly getting generated, same time when i try to generate the report with the user the issue arises.
Dear Gordon,
I agree that the pictures that is there are around 1000+ items in the report. But as stated earlier if the issue occurs across the users i could accept, but here for manager id is generating only at the user level it is not happening.

Re: EnvMgr out of memory

Thank you for the light you threw on this,
In fact, we are already using a custom file-transfer solution for
deploying the
client partitions, so it is not what makes our EnvMgr running out of
memory.
It seems to be happening at random while the application is running
normally,
but our max -fm flag is still rather low (x=32000). At first we are going
to try to
increase it.
Vincent Figari
On Mon, 01 Jun 1998 15:58:09 -0700 John Jamison <[email protected]>
writes:
This is a multi-part message in MIME format.
--------------E8C26410ECFD267FF808352A
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Kamran and Vincent,
The key is actually what node the application distribution is on,
however if it
is also the env mgr this distinction is irrelevant. I mention this to
you since
later I reference this as part of a theoretical solution.
The basic problem here is that the process of serving ISCRIPT with an
actual
application upload consumes resources in the partition which hosts the
environment
manager. Resources are consumed both for (a) cycling through
installed
apps
and checking versions, and (b) causing the file uploads of newly
deployed apps
to those clients. Turns out that (b) is more of a load on the
"distribution
server" (e.g. env mgr in your case). In order to copy the partition
files (which can be substantial), the env mgr needs to allocate
buffers,
read the application partition
files into those buffers, pass those buffers back to ISCRIPT (client)
which then
writes them out. Lets say your client partition is 5MB and 400 people
hit the
env mgr at the same time for a deployment. Thats 400 * 5MB of dynamic
memory
required. Granted they won't all happen at the same time, but you
get
the picture.
What is really required in this case to handle the scalability
requirement is a queueing mechanism whereby the env mgr only processes
N
number of ISCRIPT requests
at a time, and holds off on the rest of them until they are done.
Mostly for this reason we have implemented a WAN based application
deployment mechanism which uses FTP instead of envmgr. In this way
the
ftp server may or may not fail (it usually doesn't), but if it were to
fail it protects the forte environment from crashing. Another
potential
key is to set up ISCRIPT to limit checking to only one application
rather than all applications in the environment. This might take some
investigation.
Another technique we have employed is to do a form of application
version checking in a custom "launcher" type application. It does
version checking and when it detects a variance, invokes the
aforementioned FTP process to download the new application partition
files.
One thing to try is to run a separate node manager on your server node
with the huge memory assignment. Put your application distributions
on
that nodemgr instead of the env mgr. Set up this nodemgr to be
started
in the AM and shut down during some other time period. Perhaps this
will separate the "load" put on by the 600 simultaneous ISCRIPT
requests
to a node mgr that can fail and not cause havok. This is all theory
by
the way. So your env mgr runs with the usual env mgr assigned, but a
separate nodemgr is started on the same box with a different
FORTE_ROOT
setting. This means it will have its own appdist/... tree. Copy
your
application distributions from the envmgr $FORTE_ROOT/appdist/... tree
to this new tree, and use this virtual "node" for subsequent
distributions.
Then monitor the processes (both env mgr and this new nodemgr) during
the mass ISCRIPT. If my theory holds, then the nodemgr will get busy
but the env mgr won't, and therefore
your environment will continue to stay up. At least half your
problems
go away, anyway.
Kamran Amin wrote:
Vincent,
We are having the same problem at Oxford and using the same version of
Forte. We were having this problem when we started to add more people
to our environment. We talked it over with Tech support and they had
recognized this problem as a bug in Forte 2.0. The problem was that a
lot of people had the "same node name" in their control panel. We have
1200 deployed clients and about 600 concurrent users. The users are
setup to run iscript every morning when they turn their 95 machine on to
down load the lates version of the application. This caused the node
manager to go crazy and finally seg-v. When we went to each and every
client(this was when we had about 300 client installed) the node manage
stop crashing. Tech support said this has been fixed in 3.0. We still
have this problem because some client still have the same node name but
we make sure the -fm flag is high(about x = 300000). With the high
memory setting the environment have stop comming down every morning and
only come down when we install a new verson of the application. I think
they have fixed this problem in a later version if 2.0H?? but can not
confirm this. I have been talking with my sales rep to find out if this
is true.
Some Stats:
Number of client deployed: 1200 to 1300
Number of concurrent users: about 600
Number of application in the environment: two
Number of environments: 8(four main and four for failover)
Number of Box: two IBM RS600 with 1 gig of memory and 4 CPU each.
Memory setting on the Environment: -fm (x=300000)
Hope this helps
ka
From: [email protected][SMTP:[email protected]]
Sent: Monday, June 01, 1998 9:59 AM
To: [email protected]
Subject: EnvMgr out of memory
Hello all,
Has anyone experienced their EnvMgr dying due to
an out of memory ?
If yes, do you have any clue about what is eating the
most the EnvMgr memory :
- Number of registered partitions ? (server and client)
- Number of simultaneous queries to the Name Server ?
- Number of applications having been installed/uninstalled since
the
EnvMgr was started ?
Also, what x= did you have to put for the EnvMgr memory flags ?
Although we increased the Memory settings, we are still running
into this problem once in a while in our production environment
(Forte 2.0H1).
Thanks to all,
Vincent Figari
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive
<URL:http://pinehurst.sageit.com/listarchive/>
>
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:http://pinehurst.sageit.com/listarchive/>
John Jamison [email protected]
Vice President and Chief Technology Officer
Sage IT Partners, Inc.
Voice: 415 392-7243 x 306
Fax: 415 391-3899
Internet Enabled Business Change
http://www.sageit.com
--------------E8C26410ECFD267FF808352A
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for John Jamison
Content-Disposition: attachment; filename="vcard.vcf"
begin: vcard
fn: John Jamison
n: Jamison;John
org: Sage IT Partners
adr: 353 Sacramento Street;;Suite 1360;San
Francisco;CA;94111;USA
email;internet: [email protected]
title: Vice President and Chief Technology Officer
tel;work: 415 392 7243 x 306
tel;fax: 415 391 3899
x-mozilla-cpt: ;0
x-mozilla-html: FALSE
version: 2.1
end: vcard
--------------E8C26410ECFD267FF808352A--
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:http://pinehurst.sageit.com/listarchive/>

Thank you for the light you threw on this,
In fact, we are already using a custom file-transfer solution for
deploying the
client partitions, so it is not what makes our EnvMgr running out of
memory.
It seems to be happening at random while the application is running
normally,
but our max -fm flag is still rather low (x=32000). At first we are going
to try to
increase it.
Vincent Figari
On Mon, 01 Jun 1998 15:58:09 -0700 John Jamison <[email protected]>
writes:
This is a multi-part message in MIME format.
--------------E8C26410ECFD267FF808352A
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Kamran and Vincent,
The key is actually what node the application distribution is on,
however if it
is also the env mgr this distinction is irrelevant. I mention this to
you since
later I reference this as part of a theoretical solution.
The basic problem here is that the process of serving ISCRIPT with an
actual
application upload consumes resources in the partition which hosts the
environment
manager. Resources are consumed both for (a) cycling through
installed
apps
and checking versions, and (b) causing the file uploads of newly
deployed apps
to those clients. Turns out that (b) is more of a load on the
"distribution
server" (e.g. env mgr in your case). In order to copy the partition
files (which can be substantial), the env mgr needs to allocate
buffers,
read the application partition
files into those buffers, pass those buffers back to ISCRIPT (client)
which then
writes them out. Lets say your client partition is 5MB and 400 people
hit the
env mgr at the same time for a deployment. Thats 400 * 5MB of dynamic
memory
required. Granted they won't all happen at the same time, but you
get
the picture.
What is really required in this case to handle the scalability
requirement is a queueing mechanism whereby the env mgr only processes
N
number of ISCRIPT requests
at a time, and holds off on the rest of them until they are done.
Mostly for this reason we have implemented a WAN based application
deployment mechanism which uses FTP instead of envmgr. In this way
the
ftp server may or may not fail (it usually doesn't), but if it were to
fail it protects the forte environment from crashing. Another
potential
key is to set up ISCRIPT to limit checking to only one application
rather than all applications in the environment. This might take some
investigation.
Another technique we have employed is to do a form of application
version checking in a custom "launcher" type application. It does
version checking and when it detects a variance, invokes the
aforementioned FTP process to download the new application partition
files.
One thing to try is to run a separate node manager on your server node
with the huge memory assignment. Put your application distributions
on
that nodemgr instead of the env mgr. Set up this nodemgr to be
started
in the AM and shut down during some other time period. Perhaps this
will separate the "load" put on by the 600 simultaneous ISCRIPT
requests
to a node mgr that can fail and not cause havok. This is all theory
by
the way. So your env mgr runs with the usual env mgr assigned, but a
separate nodemgr is started on the same box with a different
FORTE_ROOT
setting. This means it will have its own appdist/... tree. Copy
your
application distributions from the envmgr $FORTE_ROOT/appdist/... tree
to this new tree, and use this virtual "node" for subsequent
distributions.
Then monitor the processes (both env mgr and this new nodemgr) during
the mass ISCRIPT. If my theory holds, then the nodemgr will get busy
but the env mgr won't, and therefore
your environment will continue to stay up. At least half your
problems
go away, anyway.
Kamran Amin wrote:
Vincent,
We are having the same problem at Oxford and using the same version of
Forte. We were having this problem when we started to add more people
to our environment. We talked it over with Tech support and they had
recognized this problem as a bug in Forte 2.0. The problem was that a
lot of people had the "same node name" in their control panel. We have
1200 deployed clients and about 600 concurrent users. The users are
setup to run iscript every morning when they turn their 95 machine on to
down load the lates version of the application. This caused the node
manager to go crazy and finally seg-v. When we went to each and every
client(this was when we had about 300 client installed) the node manage
stop crashing. Tech support said this has been fixed in 3.0. We still
have this problem because some client still have the same node name but
we make sure the -fm flag is high(about x = 300000). With the high
memory setting the environment have stop comming down every morning and
only come down when we install a new verson of the application. I think
they have fixed this problem in a later version if 2.0H?? but can not
confirm this. I have been talking with my sales rep to find out if this
is true.
Some Stats:
Number of client deployed: 1200 to 1300
Number of concurrent users: about 600
Number of application in the environment: two
Number of environments: 8(four main and four for failover)
Number of Box: two IBM RS600 with 1 gig of memory and 4 CPU each.
Memory setting on the Environment: -fm (x=300000)
Hope this helps
ka
From: [email protected][SMTP:[email protected]]
Sent: Monday, June 01, 1998 9:59 AM
To: [email protected]
Subject: EnvMgr out of memory
Hello all,
Has anyone experienced their EnvMgr dying due to
an out of memory ?
If yes, do you have any clue about what is eating the
most the EnvMgr memory :
- Number of registered partitions ? (server and client)
- Number of simultaneous queries to the Name Server ?
- Number of applications having been installed/uninstalled since
the
EnvMgr was started ?
Also, what x= did you have to put for the EnvMgr memory flags ?
Although we increased the Memory settings, we are still running
into this problem once in a while in our production environment
(Forte 2.0H1).
Thanks to all,
Vincent Figari
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive
<URL:http://pinehurst.sageit.com/listarchive/>
>
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:http://pinehurst.sageit.com/listarchive/>
John Jamison [email protected]
Vice President and Chief Technology Officer
Sage IT Partners, Inc.
Voice: 415 392-7243 x 306
Fax: 415 391-3899
Internet Enabled Business Change
http://www.sageit.com
--------------E8C26410ECFD267FF808352A
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for John Jamison
Content-Disposition: attachment; filename="vcard.vcf"
begin: vcard
fn: John Jamison
n: Jamison;John
org: Sage IT Partners
adr: 353 Sacramento Street;;Suite 1360;San
Francisco;CA;94111;USA
email;internet: [email protected]
title: Vice President and Chief Technology Officer
tel;work: 415 392 7243 x 306
tel;fax: 415 391 3899
x-mozilla-cpt: ;0
x-mozilla-html: FALSE
version: 2.1
end: vcard
--------------E8C26410ECFD267FF808352A--
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:http://pinehurst.sageit.com/listarchive/>

What the **** is OUT OF MEMORY !!!

We started a session everything is going well and just when i insert a plug-in on the master
the freaking Logic paused and gave me "WARNING" out of memory
the it says the plug-in could not be inserted or deleted, thats it CRASH
we lost a good session
and now the good feeling we had when we started is not there anymore
i got people sitting looking at me like i wasted my money on logic and mac
G5 Dual 2.5
4GB of rams
plus a Node
is it enough to run 7 instruments and 4 space design and 2 Audio tracks ?
GOD
Any one knows why?

hi stash, yes i've had the same problem - just another bug, i suppose.
here's another one - "memory too tight for drawing". generally occurs in matrix of score editors.
man, i would - and i've been working in sydney as a musician/studio owner since 1965 -
NEVER TRUST a critical session to a buggy piece of software like logic is now.
whenever i do a vocal session i boot to os9 and use protools(with apogee in/outs). i simply do a rough mix of the track and port 'er in. logic is too unreliable - especially when people pay you money.
motu software is starting to LOOOK GOOOOD!!!!!
most of the gear we're buying is pro-sumer, not pro-fessional. it has a known shelf life. i still have really excellent stuff from my analog studio days which just KEEPS ON WORKIN'. my tannoys are 29 years young. my power amp has NEVER NEEDED A SERVICE in 20years.
as a comparison, since i went "digital" (DAWs), in 2000, i've had to pay out for a new logic board for the mac, been thru three cd/dvd burners, two monitors
2 external drives,one scsi card and two drives, 2 apple mice, never mind the dud blank media, etc. etc.
i know a lot of tech people in sydney and i take their advice seriously, and their opinions generally reinforce my own - digital gear ain't RUGGED enough.
statisically, i've have more failures in the last 6 years with digital, than i had in the 20 years before that with an analog studio.
i have many friends and clients and they are having similar problems, (excluding these forums) - maybe not as many but then they don't use the studio 24/7 like i do.
my suggestion to everybody who buys new gear: trash it! work it hard from day one - if it doesn't fall over, it's cool. if it does malfunction, take it back to where you bought it from and get your dollars back! if shop owners complain
to the manufacturers and start sending gear back, these people WILL take it seriously. nobody likes a drop in profits/market share.
G4 450DP Mac OS X (10.4.4) digidesign, motu, apogee

EnvMgr out of memory

Hello all,
Has anyone experienced their EnvMgr dying due to
an out of memory ?
If yes, do you have any clue about what is eating the
most the EnvMgr memory :
- Number of registered partitions ? (server and client)
- Number of simultaneous queries to the Name Server ?
- Number of applications having been installed/uninstalled since the
EnvMgr was started ?
Also, what x= did you have to put for the EnvMgr memory flags ?
Although we increased the Memory settings, we are still running
into this problem once in a while in our production environment
(Forte 2.0H1).
Thanks to all,
Vincent Figari
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:http://pinehurst.sageit.com/listarchive/>

John,
That sound like a good idea. It will take some setting up but we will
give it a try. As you stated that 600 clients will require about 600 *
client partition size of memory. That is why we have broken down the
clients by regions and put them on 4 environments. Only a maximum of
300 clients will be on one environment. But al these environment are on
one box therefor the box will be very busy handling all the requests. A
better solution would be to distribute the application on a node
environment on another box with a lot of memory. I will keep you posted
on how it goes.
thanks
ka
From: John Jamison[SMTP:[email protected]]
Sent: Monday, June 01, 1998 6:58 PM
To: Kamran Amin
Cc: '[email protected]'; '[email protected]'
Subject: Re: EnvMgr out of memory
<<File: vcard.vcf>>
Kamran and Vincent,
The key is actually what node the application distribution is on,
however if it
is also the env mgr this distinction is irrelevant. I mention this to
you since
later I reference this as part of a theoretical solution.
The basic problem here is that the process of serving ISCRIPT with an
actual
application upload consumes resources in the partition which hosts the
environment
manager. Resources are consumed both for (a) cycling through installed
apps
and checking versions, and (b) causing the file uploads of newly
deployed apps
to those clients. Turns out that (b) is more of a load on the
"distribution
server" (e.g. env mgr in your case). In order to copy the partition
files (which can be substantial), the env mgr needs to allocate buffers,
read the application partition
files into those buffers, pass those buffers back to ISCRIPT (client)
which then
writes them out. Lets say your client partition is 5MB and 400 people
hit the
env mgr at the same time for a deployment. Thats 400 * 5MB of dynamic
memory
required. Granted they won't all happen at the same time, but you get
the picture.
What is really required in this case to handle the scalability
requirement is a queueing mechanism whereby the env mgr only processes N
number of ISCRIPT requests
at a time, and holds off on the rest of them until they are done.
Mostly for this reason we have implemented a WAN based application
deployment mechanism which uses FTP instead of envmgr. In this way the
ftp server may or may not fail (it usually doesn't), but if it were to
fail it protects the forte environment from crashing. Another potential
key is to set up ISCRIPT to limit checking to only one application
rather than all applications in the environment. This might take some
investigation.
Another technique we have employed is to do a form of application
version checking in a custom "launcher" type application. It does
version checking and when it detects a variance, invokes the
aforementioned FTP process to download the new application partition
files.
One thing to try is to run a separate node manager on your server node
with the huge memory assignment. Put your application distributions on
that nodemgr instead of the env mgr. Set up this nodemgr to be started
in the AM and shut down during some other time period. Perhaps this
will separate the "load" put on by the 600 simultaneous ISCRIPT requests
to a node mgr that can fail and not cause havok. This is all theory by
the way. So your env mgr runs with the usual env mgr assigned, but a
separate nodemgr is started on the same box with a different FORTE_ROOT
setting. This means it will have its own appdist/... tree. Copy your
application distributions from the envmgr $FORTE_ROOT/appdist/... tree
to this new tree, and use this virtual "node" for subsequent
distributions.
Then monitor the processes (both env mgr and this new nodemgr) during
the mass ISCRIPT. If my theory holds, then the nodemgr will get busy
but the env mgr won't, and therefore
your environment will continue to stay up. At least half your problems
go away, anyway.
Kamran Amin wrote:
Vincent,
We are having the same problem at Oxford and using the same version
of
Forte. We were having this problem when we started to add more people
to our environment. We talked it over with Tech support and they had
recognized this problem as a bug in Forte 2.0. The problem was that a
lot of people had the "same node name" in their control panel. We have
1200 deployed clients and about 600 concurrent users. The users are
setup to run iscript every morning when they turn their 95 machine on to
down load the lates version of the application. This caused the node
manager to go crazy and finally seg-v. When we went to each and every
client(this was when we had about 300 client installed) the node manage
stop crashing. Tech support said this has been fixed in 3.0. We still
have this problem because some client still have the same node name but
we make sure the -fm flag is high(about x = 300000). With the high
memory setting the environment have stop comming down every morning and
only come down when we install a new verson of the application. I think
they have fixed this problem in a later version if 2.0H?? but can not
confirm this. I have been talking with my sales rep to find out if this
is true.
Some Stats:
Number of client deployed: 1200 to 1300
Number of concurrent users: about 600
Number of application in the environment: two
Number of environments: 8(four main and four for failover)
Number of Box: two IBM RS600 with 1 gig of memory and 4 CPU each.
Memory setting on the Environment: -fm (x=300000)
Hope this helps
ka
From: [email protected][SMTP:[email protected]]
Sent: Monday, June 01, 1998 9:59 AM
To: [email protected]
Subject: EnvMgr out of memory
Hello all,
Has anyone experienced their EnvMgr dying due to
an out of memory ?
If yes, do you have any clue about what is eating the
most the EnvMgr memory :
- Number of registered partitions ? (server and client)
- Number of simultaneous queries to the Name Server ?
- Number of applications having been installed/uninstalled since the
EnvMgr was started ?
Also, what x= did you have to put for the EnvMgr memory flags ?
Although we increased the Memory settings, we are still running
into this problem once in a while in our production environment
(Forte 2.0H1).
Thanks to all,
Vincent Figari
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:http://pinehurst.sageit.com/listarchive/>
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:http://pinehurst.sageit.com/listarchive/>--
John Jamison [email protected]
Vice President and Chief Technology Officer
Sage IT Partners, Inc.
Voice: 415 392-7243 x 306
Fax: 415 391-3899
Internet Enabled Business Change
http://www.sageit.com
To unsubscribe, email '[email protected]' with
'unsubscribe forte-users' as the body of the message.
Searchable thread archive <URL:http://pinehurst.sageit.com/listarchive/>

Printer "Out of Memory". Can not print spreadsheets

Up to the last few months I had no problem printing everything including spreadsheets. Now I get the following when I bring up a Microsoft spreadsheet and try to print it "Out of Memory" "Save or close any open files or applications" I can print all other files including pictures, Word test, emails, etc.
Bult I have no open files or applications.
Printer: HP Photosmart C4780, Windows XP, Compaq/HP computer.
No changes to my software recently except adding RosettaStone language CDs.
Suggetions?
Thanks,
Clay

I'm not printing spreadsheets -I'm trying to print Quark or AdobePDF files on a Business Inkjet 2800dtn (large tabloid size) - I've had trouble w/this printer before but not this "OUT OF MEMORY" problem.
I can print Notepad files...but it will not print other files that it printed fine before. I even got one PDF out of this machine yesterday before it went bonkers. It keeps saying "file won't print" & then "Printer error- turn off power then on" = I must have done it 100 times. No luck at all.
I KNOW there is a way to use "computer memory" to increase the capacity of this machine without buying additional printer memory, but I'm not sure how to do it. Does anyone know? Will anyone tell? Obviously, mine was working BEFORE. Another message came up in the Configuration page that I WAS able to print, saying it had a "Firmware error" - HP hasn't updated this firmware as far as I can see, since 2005. They haven't updated ANYTHING on this POS machine, to be blunt about it.
(inHP's favor, I love,love, love my 2340duplex Laserjet - it's a wonderous miracle for me).
Anyone know how to connect the computer memory to the printer - or anything about Firmware Updates (or should I reinstall the software?)

My pictures went blank (out of memory?)

Similar Messages

Maybe you are looking for