Multiple EOF Stacks Question

2008-02-20 Thread Dov Rosenberg
We have a large scale application that currently uses a single EOF stack. We don¹t use Project Wonder. How would we approach adding multiple EOF stacks to our application? What is the design strategy for determining how many stacks to use? Is it equivalent to setting up a connection pool? Any thoug

Re: Progressively all threads go into deadlock

2008-02-20 Thread Mr. Pierre Frisch
Hi Fabrice, How many stacks have you created for your multithreaded application? Are you using only one? Then you may have a contention as EOF is single threaded. Can you elaborate a little bit more on how your application works? Cheers Pierre -- Pierre Frisch [EMAIL PROTECTED] On Feb

Re: slow app when having a lot of D2W rules ?

2008-02-20 Thread Anjo Krank
I got two apps with around 100-200 page configs and 1000-1500 rules (~150 entities) and two others with ~50 entities, 70 pages and ~500 rules. Netstruxr had (if memory serves) 2k rules, 1.2k pages, 600 entities. Cheers, Anjo Am 20.02.2008 um 19:31 schrieb Dominique Schoenenberger: They ar

Re: Rant: Why I hate Java Client

2008-02-20 Thread Florijan Stamenkovic
So this is some sort of config problem, brought on somehow by the fact that I've moved all the .jar cruft we had hanging out in / Library/Java/Extensions into a framework. It looks like the server is forwarding some sort of EOGeneralAdaptorException to the client which might have had useful

Re: Rant: Why I hate Java Client

2008-02-20 Thread Lachlan Deck
It might be worth overriding the code on server to catch any exceptions and string-ify its stacktrace and throw back to client some exception they both understand. On 21/02/2008, at 7:13 AM, Pierce T. Wetter III wrote: So I get this backtrace (see below). So this is some sort of config pro

Re: Progressively all threads go into deadlock

2008-02-20 Thread Chuck Hill
On Feb 20, 2008, at 5:33 AM, Fabrice Pipart wrote: On Feb 19, 2008, at 8:33 PM, Chuck Hill wrote: We are thinking about a different thing : could it be a memory problem ? We are allocating "only" 256 Mb and are dealing concurrently with thousands of objects in background threads, componen

Rant: Why I hate Java Client

2008-02-20 Thread Pierce T. Wetter III
So I get this backtrace (see below). So this is some sort of config problem, brought on somehow by the fact that I've moved all the .jar cruft we had hanging out in /Library/Java/ Extensions into a framework. It looks like the server is forwarding some sort of EOGeneralAdaptorException to th

Re: slow app when having a lot of D2W rules ?

2008-02-20 Thread Jesse Tayler
if you use significant keys, they may be hitting methods more often than you realize, otherwise the rules are not much more than a cache of a dictionary for each configuration. this can add a bit of latency as apps get warmed up, but D2W rules are pretty light in today's world. having a f

Re: slow app when having a lot of D2W rules ?

2008-02-20 Thread Dominique Schoenenberger
They are hand-written rules and may be between 400-500 page configs. How many rules, page configs do you have in your application ? Dominique On 18 févr. 08, at 21:20, Anjo Krank wrote: Just out of curiosity: are these hand-written rules? If yes, how many page configs do you have? Cheers,

Re: wo features

2008-02-20 Thread Mike Schrag
HasItems: WOConditinal { var1 = [EMAIL PROTECTED]; var2 = [EMAIL PROTECTED]; expr = (var2 == true && var1 > 0); } You can do all of this (well except for the expression operating on other bindings of the same object, but that's sort of weird anyway) with WOOgnl. ms

wo features

2008-02-20 Thread Elim Qiu
Hopefully we can have more built-in (so standard) components and mechanisms so that we can do HasItems: WOConditinal { var1 = [EMAIL PROTECTED]; var2 = [EMAIL PROTECTED]; expr = (var2 == true && var1 > 0); } I don't meant to determine the sytax but like to see kay-value coding can be extended

Cannot rollbackTransaction() while a fetch is in progress (Was: MySQL 5.0 - note)

2008-02-20 Thread Simon McLean
Hi Kieran, In the last couple of days we have been getting the following error cropping up in a couple of places: java.lang.Thread.run(Thread.java:613) Caused by: com.webobjects.jdbcadaptor.JDBCAdaptorException: Cannot rollbackTransaction() while a fetch is in progress at com.webobjects

Re: Progressively all threads go into deadlock

2008-02-20 Thread Fabrice Pipart
On Feb 19, 2008, at 8:33 PM, Chuck Hill wrote: We are thinking about a different thing : could it be a memory problem ? We are allocating "only" 256 Mb and are dealing concurrently with thousands of objects in background threads, components and wolongresponses tasks Could it lead to a

Re: java heap trouble after 5.4.1 upgrade, opensnoop to the rescue

2008-02-20 Thread Anjo Krank
And you just *might* want to quote the file names, unless you want some wisecrack put a folder named "Documents Sites Music Desktop" into your /System/Library/Frameworks/ WebObjects53... Cheers, Anjo Am 20.02.2008 um 11:34 schrieb Johan Henselmans: Just a reminder for some other mindless

java heap trouble after 5.4.1 upgrade, opensnoop to the rescue

2008-02-20 Thread Johan Henselmans
Just a reminder for some other mindless soul... I am one of the many that have WebObjects53 and 54 installed side by side on Leopard, with symbolic links to the respective libraries: lrwxr-xr-x 1 root wheel 70 20 feb 11:11 JavaEOInterfaceSwing.framework -> /System/Library/Frameworks/