Why does public caching break in 5.0?

I have a gadget that prints out the time. That's how I can test caching.
I have used the response.expires as well, with a setting of 0 for testing, however, in practice it wont work b/c I want it to cache. The problem is not that caching doesn't work entirely. I want it to cache for 10 minutes, and it does that just fine. However, my problem is that in 5.0 the cache is not shared by all users. So, every user gets his/her own copy of the gadget, cached for 10 minutes. I want the cache to be shared by the users. (I have set the cache control to PUBLIC (which is how you tell the portal to share the cache for all users, at least in 4.5), and everything works in 4.5. However, in 5.0 I can not find any way to get the cache to be shared by all users.
Thanks very much for your help!

When the portal caches Portlet content, there are 2 options. The Portlet cache can be "private", which means that every user has his/her own portal cache for that Portlet, even if 2 users have the same preferences and settings. Or the user cache can be "public", which means that the cache spans all users, so that if 2 users have the exact same preferences they will use the same cached content (so the portal only stores one copy of it). Obviously, users with different settings will use only the caches that pertain to their settings.
For Plumtree Portlets, the way that you specify which type of cache your Portlet will use is through the HTTP "Cache-control" header. If the Portlet sets this header to "private", then the portal uses "private" caches, and if it's set to "public" then the portal uses "public" caching. I think the default that Plumtree uses / used in 4.5 is "public", but I'm not positive, particularly about 5.0. In some GDK's, there were some shortcuts in the GDK to set the cache settings, but in the .NET GDK, there are no caching functions, as .NET has an HttpCachePolicy helper class that developers are supposed to use instead.
The issue is with .NET Portlets. If you have basic authentication on your Portlet server, a .NET .aspx page will ALWAYS specify "private" caching (ie, set the cache-control header to "private"). Even if you set the cache-control to "public" in your Portlet, the .aspx page will set the cache control to private. ( Interestingly, both headers are sent - the header string looks like "cache-control: public; cache-control: private". Unfortunately, Plumtree uses private in this case. ) ASP files on IIS can send a "public" cache-control header and thus use public caching in Plumtree when the directory has basic authentication turned on, but it is doesn't seem to be possible with .NET gadgets. As a result, you can not have .NET Portlets use public caching and be located on a secure web server. There is nothing in the HTTP specification that precludes using public caching with basic authentication.
Plumtree is looking into this with Microsoft.

Similar Messages

Why does Apple keep breaking the podcast app?

Why does Apple keep breaking the podcast app?

Here are my issues, maybe you can help... And yes "breaking" and "nightmare" are accurate descriptions of this update...
1. Syncing is now impossible without WiFi or a Cellular connection. I've tried your guide in the other thread, but it didn't help.
When I sync podcasts with iTunes, they DO NOT SHOW UP in the app unless I connect my iPhone 5S to WiFi or Cellular. As soon as I connect it to something, new episodes that I manually synced finally show up and can be played.
2. Arranging the podcasts in any order is now pointless. Any sort of wireless connection will reset my order to a randomized one. It's not even "podcasts with the newest episodes on top", it's just random.
3. Apple doesn't seem to allow any other podcast app to use the manual sync in iTunes. Am I wrong about that? I looked at countless paid apps and they all have their own cloud or system to only deliver episodes wirelessly...
4. And that constant prompt about Cellular data access could drive anyone insane!
I have never been this frustrated with an Apple product in my life...

Why does the eraser break calligraphic strokes?

Hi,
When creating strokes with the calligraphic brush (using an Intuos5), I noticed that the eraser tool breakes the stroke thickness when apllied to the calligraphic stroke.
The entire stroke defaults to an undesired,uniform thickness.
See video here:
http://bit.ly/10AN9WV
So why does this happen?
Is there some way to prevent the eraser from destroying the stroke thickness?
Input appreciated.
Kind Regards / Alex

http://blogs.adobe.com/adobeillustrator/2009/04/mysteries_of_the_eraser_tool_r.html

[Unknown] Bash: Why does this script break my environment?

I'm trying to build a script to do a couple makepkg tasks and it looks like I'm having trouble with a function breaking the environment. The function does several commands and then breaks makepkg. After I run the script, makepkg runs as if no options are entered:
makepkg -g
==> Check: Installer is the correct size.
==> Extracting archive...
==> Copying files...
cp: cannot overwrite non-directory `/var/abs/local/personal/worldofgoo/pkg/usr/share/games/worldofgoo/WorldOfGoo' with directory `WorldOfGoo'
==> ERROR: An unknown error has occurred. Exiting...
Instead of the expected:
makepkg -g
==> Retrieving Sources...
-> Found worldofgoo.desktop in build dir
==> Generating checksums for source files...
md5sums=('e49849a66aa50065d4d548653a33cc23')
Here's the script it just started happening when I added the function. I think it has to do with the curly brackets that awk uses.
#!/bin/bash
# mp - makepkg package building tasks
# Add md5sums following source array in PKGBUILD
md5add () {
# Delete previous md5sum entries
sed -i '/^md5sums/,/).*$/d' PKGBUILD
# Add md5sums to end if PKGBUILD
makepkg -g >> PKGBUILD
# Move md5sums to follow source array
awk 'BEGIN {
checkAt = 0
filesAt = 0
scanning = 0
/md5sums=\(/ {
checkAt = NR
scanning = 1
/source=\(/ {
filesAt = NR
scanning = 1
/)$/ {
if (scanning) {
if (checkAt > filesAt) {
checkEnd = NR
} else {
filesEnd = NR
scanning = 0
lines[NR] = $0
END {
for (i = 1; i <= NR; ++i) {
if (checkAt <= i && i <= checkEnd) {
continue
print lines[i]
if (i == filesEnd) {
for (j = checkAt; j <= checkEnd; ++j) {
print lines[j]
}' PKGBUILD > /tmp/PKGBUILD.tmp
if [ -f /tmp/PKGBUILD.tmp ]; then
mv /tmp/PKGBUILD.tmp PKGBUILD
fi
# Remove trailing blank lines
while [ "$(tail -n 1 PKGBUILD)" == "" ]; do
sed -i '$d' PKGBUILD
done
# Options
case $1 in
p ) if [ ! -f ./PKGBUILD ]; then
echo "No PKGBUILD in this directory ($(pwd))"
exit; else
makepkg -sf
fi
s ) if [ ! -f ./PKGBUILD ]; then
echo "No PKGBUILD in this directory ($(pwd))"
exit 1; else
echo "adding md5sums"
md5add
makepkg -f --source
fi
* ) # Display usage if full argument isn't given
echo " ${0##*/} <option> - makepkg building tasks:"
echo " p - build package (also installs dependencies)"
echo " s - build source-only tarball (adds md5sums, tars for submission)"
exit
esac
I have to restart to reset the environment (logout doesn't work) so I'l like to be able to fix this . Any ideas?
Last edited by Gen2ly (2009-10-30 15:19:17)

Just reinstalled and not having it come up again. Putting this in wonderland.
Last edited by Gen2ly (2009-10-30 15:18:44)

Why does my paragraph break before and after a subscript?

RH 9.0.2.271, Word 2010, Windows 7, 64-bit PC
This is an intermittent problem occurring in the Word output. The WebHelp is fine. In the following image you can see that the subscripts worked in the early part of the paragraph, but then split.
Here is the html:
where qtot
is the total rate of heat transfer (which you supply), kc is the laminar thermal conductivity
of the primary fluid in the cell, ρc
is the density of the primary fluid in the cell, CPc
is the specific heat capacity of the primary fluid in the cell, Vc is the volume of the cell,
<img src="HX_EQ2a.jpg" alt="" style="border: none;" border="0" /> is the
magnitude of velocity in the cell, Tref is the reference temperature
for the heat exchanger, and Tc is the temperature in the cell.
The summation in the denominator of equation 1 is carried out over all
the cells in the heat exchanger. Thus, it can be seen that the sum of
the weights for all the cells always add up to 1, so that the distribution
scheme ensures that the sum of heat transferred to all the cells in

Peter's not available at the moment, but perhaps Rick can do that.
I've just run your HTML through the Word output with RoboHelp 11/Word 2013 and it worked perfectly there. Now that doesn't help, but perhaps you can try the following: instead of adding tags in the tags, apply the classed directly to the element. (Don't forget to update your CSS as well.) Perhaps this will let RoboHelp correctly create the output.
|f that doesn't help, check the Word document. Which style is assigned to the separate elements in the Word documents? Perhaps we can find something in there.
Kind regards,
Willam

Why does the menu break every time I adjust it?

The menu drop downs seem to lose functionality as I change it. This only gets worse each time I publish/preview it.

Problem solved: I have been using Walnut for Firefox ver. 2.0.28 theme. For some reason this has been causing the problem. Started FF in Safe Mode, and the slider problem disappeared. Then proceeded to re-enable the Walnut theme: the problem reappeared. Disabled Walnut theme and reenabled all my other add-ons (about 6 of them). Still no problem. So, clearly there is a problem with the use of Walnut theme in FF ver. 32, at least on my Windows 7 installation.
Thanks for all your suggestions and help. Good job, all!

Why does this formula break VLOOKUP?

I have the table below. The formula in B3 is causes the VLOOKUPS in C3 to be off. If I just use the value 0.20 in B3, I get "sixty" in C3.
The B3 formula is:
=ABS(A2-A4)
and C3 is:
=VLOOKUP(B3,D1:E5, 2)
What needs to change for this to work correctly?

Functionally, there's no difference between the original VLOOKUP formula, where the 4th argument is omitted, and the one suggested by Wayne, where the 4th argument is given the value 1 (or TRUE).
As described in the excerpt from the Formulas and Functions User Guide shown in Wayne's post:
Wayne Contello wrote:
close match (TRUE, 1, or omitted): If there’s no exact match, select the column with the largest top-row value that is less than the search value.
Reproducing the table shown on my own machine produces the value (fifty-five) with either version of the formula.
I suspect that the result in C3, like any other spreadsheet cal\culation, is subject to a rounding error caused when the values in A2 and A4 are translated to binary for the calculation, then reinterpreted as a decimal value for the display. If C3 is set to show two decimal places, the actual value could be anything in the range 0.195 - 0.2099999999999. VLOOKUP uses the actual value, not the displayed value, so any actual value in the range 0.195-0.1999999999999999999 will display in C3 as 0.20, but will return the result for 0.10, 'the largest search-for value that is less than the search value.'
Try revising your C3 formula to =ROUND(ABS(A2-A4),2)
On my test table, this produced the expected result (sixty).
Regards,
Barry

Why does my cache go up to 600+ kb when I've only visited onr site? This is happening w/ the latest auto update. No wonder I can't connect to anything!

I just noticed this, so it's an addendum to my last question/problem with the "latest edition". Thanks!

Sounds like your wi-fi signal strength has degraded. Try rebooting your wi-fi router, or get it serviced.
Do you have similar issues with any other wi-fi routers?

Y does the cd break up when its downloaded

why does the cd break up when its downloaded

It may not be convenient long term but if you can test with ethernet and see if the problem goes away, then you'll have identified wifi as the issue.
If problem persists on ethernet it's something else.
AC

Why won't InDesign break hyphenated words at the hyphen?

Why does InDesign not break hyphenated words at the hyphen?
For example: "empty-handed".
InDesign will break either "empty" or "handed" on either side of the hyphen. But not right at the hyphen, even if "empty" and "handed" are each set to No Break.
Anyone know why?
The hypen is typed in normally using the hyphen key on the keyboard.
If I type a space after the hyphen, InDesign will break there, of course: "empty- handed"
InDesign will break at an en-space: "empty–handed".
But I prefer not to have to type a space or insert an en-space or other characters because I also need to export text to ePUB and want only the actual text.
Correct break (if needed) at end of a line should be:
. . . empty-
handed . . .

There is no reason why InDesign should avoid breaking hyphenated words at the hyphen. That's the preferred break location.
This must be a bug.
Evidence:
When I turn Hyphenation OFF for a paragraph that contains a hyphenated word (such as my example, "empty-handed"), the hyphenated word is correctly broken at the hyphen (if needed at the end of a line). But when I turn hyphenation back ON, the word will not break at the hyphen.
See the two screenshots below.
1) Hyphenation is ON. "Empty-handed" is unbrokenat start of Line 3, yet line 2 is very loose, with room to fit "Empty-":
2) Hyphenation is OFF. "Empty-handed" is broken at the end of line two, at the hyphen (resulting in better word spacing).
By the way, the other settings in this Hyphenation dialog don't affect this issue. I tried changing them all, with no different result relating to the break at the hyphen.

Why does iTunes use an iPod Photo Cache at all?

Hi,
When iTunes is used to synchronize photos to a device, it creates an "iPod Photo Cache" folder. Why does this folder even exist?
For users that have multiple devices with multiple resolutions (e.g., iPad 3rd Generation, iPad Mini, iPhone 4S, iPhone 3G), the iPod Photo Cache is grow to approximately four times the original size of the photo. If I have 15 GBs of photos and a 250 GB hard drive, then the combined usage of 75 GBs (15 GB photos + 60 GB cache) is extremely unreasonble.
Device resolutions change often enough (these days) such that having multiple devices is probably more common than having more than one of a single product. iTunes should be designed to do the real-time conversion (which it already does), but not save it to the hard drive.
Also allowing the ability to change the location of the "iPod Photo Cache" folder would help many users, but not myself since I do not have an external drive.
Any help is appreciated.
Thanks.

Thank you for your response. I should have clarified. When trying to delete from the library, it gives me the option to delete them from the computer or just from the library. I choose not to delete them from the computer, just the library. As a result, I would not expect the deletions to free up room on the hard drive, but I don't see how it actually increases the space used up on the hard drive. If I delete a song from the library, but not the computer, does iTunes save the song again someplace else (such that there are now two copies of the song somewhere on the system)? That seems to be the only explanation for why the free space on my hard drive would decrease after deleting a song from the library.
Windows XP Pro

Why does Acrobat 11 open cached or "ghost" versions of my PDFs?

Here's my work flow...
Save a PDF from an Illustrator doc/open PDF in Acrobat to sanitize/email PDF
-get revision request-
Revise same Illustrator file/save as a PDF/open PDF in Acrobat and it views as the original PDF!
When I "quick look" on the desktop the PDF displays the revised comp. So my question is why does Acrobat 11 keep the cached image of the PDFs and how can I prevent this? Quitting Acrobat will fix the issue once so if I go to revise the same file for a 3rd, 4th time it'll show the cached version every time.

If you choose 'Preserve Illustrator editing' when exporting from AI, it actually embeds the entire AI file inside the PDF as a hidden attachment. Opening the PDF in Illustrator it will ignore the 'real' PDF content and open the AI object instead, so you never see the changes from Acrobat.

Why does Encore copy the file into Cache?

It seems that Encore will copy the m2v file into ProjectDir/Cache\Media Cache Files\mlf.cache.MACC
The file is rather large, nearly 4 GB (2gb for each ac3 file).
I used to run the previous version and I don't think it makes copy in Cache.
Why does Encore do that?
Any way to make Encore just use the m2v file directly without copying it into Cache?
Thanks!

Same prob for me. Very annoying becuz the program itself is very useful, probably the best dvd menu making system ever but I was tryin to set up an audio dvd wit my own design, prepared sound files like converted them all to ac3 format but every time it is the same, despite deleting cache, the program recreates and doubles these unknown file types of large size taking up irrationally large space. A normal dvd size of 4.7 gbytes to set up would in this case require 50 gbytes or more as I have calculated. More than ten times larger than one dvd. Ridiculous.
If there's no need for transcode by file format ac3, then why should they time after time be stored in cache mostly taking up double space ? It should be working as cd or dvd write where you're offered two options, either creating an image file on hard disk or on the fly which doesn't take any space of the disk and is more direct.
So what should I do now ? Look for another program or surrender this one and let it eat up more than 50 gigas ?
Awful situation is I don't have time to look after programs.
But I ain't got this amount of disk space either.
My project must be ready in a few days but no hope for that like this.
Help me, it's urgent. Thank you.

Why does Firefox create so many temp files in cache? Once it filled up entire free HD space (46GB)?

Every now & then, Firefox will have created zillions of directories and files in Cache within the directory for a Firefox profile. The directories are called 0 (zero) through F (using the hex numbers as names, i.e.g), and within each of these will be more directories (with two digit hex names). The last time this happened (1 week ago), these directories ate up ALL of my free hard drive space (about 46 GB worth).
Why does this happen? Can I stop it from happening? If so, how?
THanks.

Yes it should be easy to reset.
It may be set from
* '''Options -> | Advanced | -> |Network| -> Cached Web Content''' [] Override automatic cache management
* see [[#w_cached-web-content]]'''_cached-web-content'''
Presumably your automatic setting has been overridden and the box needs unchecking.
I know Firefox does split the cache up like that IIRC it allows the cache to be faster and solves fragmentation issues.
46GB That seems a surprising amount and can not be right if it eats up all hard disk space. I also know that the file places.sqlite is a variable size, but that is based on the system in use.
The ordinary HTTP cache is also set presumably intelligently, but can be over-ridden by the user
This has some background info
* https://developer.mozilla.org/en-US/docs/HTTP_Caching_FAQ
*http://kb.mozillazine.org/Browser.cache.disk.capacity
My own settings as viewed in about:support are
browser.cache.disk.capacity 358400
browser.cache.disk.smart_size_cached_value 266240
browser.cache.disk.smart_size.first_run false
My cache use is limited to about 350MB

Why does pages enter a paragraph after a page break?

Why does pages enter a paragraph after a page break? This comes into play when I am using bullets and numbers.

Because Page Breaks and Section Breaks are returns. Use soft returns [Line breaks] if that is what you want.
Peter

Why does public caching break in 5.0?

Similar Messages

Maybe you are looking for