• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: Your session has timed out.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Your session has timed out.


  • Subject: Re: Your session has timed out.
  • From: Chuck Hill <email@hidden>
  • Date: Wed, 04 Jun 2003 14:36:12 -0700

I've never used session.defaultEditingContext() in the thread processing
the long request.  Seemed like a bad idea.  Hence the next request does not
block.  Locking the EC doesn't (last time I checked) lock the DB Context.
Multiple treads can lock multiple ECs at the same time.  They just can't
lock the same one.


Chuck


At 04:07 PM 04/06/2003 -0500, Jonathan Rochkind wrote:
>At 01:08 PM 6/4/2003 -0700, Chuck Hill wrote:
>>Yes, that is a problem.  Not so much for the update on the LR page (which
>>can just check a flag set by the thread with the EOF lock)
>
>PS: How can the status page check this flag?  Ordinarily, it will do so
>with a request to the server. Ordinarily, a session-based request to the
>server automatically takes out a lock on
>session.defaultEditingContext().  If it can't acquire this lock becuase of
>work going on in the long computation....  then the request going to block
>until the long computation is done.  If you want to give the user a
>progress bar, that's not going to work.  If you just want to give the user
>a non-changing 'please wait' message until the computation is complete, it
>may work---unless the browser (or the web server!) times out the request
>before the computation is complete!
>
>Am I missing something?
>
>--Jonathan
>
>
>>but for other
>>requests which will block on the DB context lock.  I think this would only
>>be a problem when a call to saveChanges() took a long time.  That would
>>lock the database context and all further DB access would block until the
>>save completed.  It should be OK for reporting where the app would read a
>>bit of data (lock and unlock the DB context), process it, read a bit more
>>etc.  Or am I missing something?
>>
>>Chuck
>>
>>
>>At 02:39 PM 04/06/2003 -0500, Jonathan Rochkind wrote:
>> >At 11:41 AM 6/4/2003 -0700, Chuck Hill wrote:
>> >
>> >>3. Use the WOLongResponsePage to generate the report.  While this is a
>> >>somewhat elegant and appropriate solution it also implicitly makes
your app
>> >>dispatch requests concurrently.  This makes it very important that
your app
>> >>is threadsafe and all the EOF objects are getting locked properly.  You
>> >>will be wanting to test this!
>> >
>> >It's very tricky to get WOLongResponsePage to work when the
>> >long-running-computation involves EOF.  With the newly better documented
>> >EOF locking APIs, it should be possible somehow, but I haven't had time to
>> >figure it out. You can't just do what seems obvious----because if the long
>> >running computation has the right locks out on the EOF stack, then the
>> >'update' page's request to see if the computation was done will end up
>> >blocking until the computation IS done. Which usually isn't the behavior
>> >you want, although if you don't need to show a 'progress bar', it may be
>> >sufficient. But it can be tricky to get it working right.  I wouldn't
>> >reccomend messing around with WOLongResponse for EOF operations unless you
>> >have a reasonable idea of what's going on, and have some time to figure it
>> >out.
>> >
>> >--Jonathan
>> >
>> >
>> >>4. Move the reporting to another reporting only application and increase
>> >>the Connect and Receive timeouts.  This will prevent the session
>> >>restoration errors while not blocking users of the main application.
>> >>
>> >>
>> >>Chuck
>> >>
>> >>--
>> >>
>> >>Chuck Hill                                 email@hidden
>> >>Global Village Consulting Inc.             http://www.global-village.net
>> >>_______________________________________________
>> >>webobjects-dev mailing list | email@hidden
>> >>Help/Unsubscribe/Archives:
>> >>http://www.lists.apple.com/mailman/listinfo/webobjects-dev
>> >>Do not post admin requests to the list. They will be ignored.
>> >
>> >
>>
>>--
>>
>>Chuck Hill                                 email@hidden
>>Global Village Consulting Inc.             http://www.global-village.net
>>_______________________________________________
>>webobjects-dev mailing list | email@hidden
>>Help/Unsubscribe/Archives:
>>http://www.lists.apple.com/mailman/listinfo/webobjects-dev
>>Do not post admin requests to the list. They will be ignored.
>_______________________________________________
>webobjects-dev mailing list | email@hidden
>Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/webobjects-dev
>Do not post admin requests to the list. They will be ignored.
>

--

Chuck Hill                                 email@hidden
Global Village Consulting Inc.             http://www.global-village.net
_______________________________________________
webobjects-dev mailing list | email@hidden
Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/webobjects-dev
Do not post admin requests to the list. They will be ignored.

  • Follow-Ups:
    • Re: Your session has timed out.
      • From: Nick Pilch <email@hidden>
References: 
 >Re: Your session has timed out. (From: Chuck Hill <email@hidden>)
 >Re: Your session has timed out. (From: Chuck Hill <email@hidden>)
 >Your session has timed out. (From: Jevon Hills <email@hidden>)

  • Prev by Date: JDBC woes
  • Next by Date: Re: JDBC woes
  • Previous by thread: Re: Your session has timed out.
  • Next by thread: Re: Your session has timed out.
  • Index(es):
    • Date
    • Thread