In fact I think we need to remove that message.
The honest truth is that I haven't been able to reproduce a real deadlock in
lab conditions, and I wonder if people call "deadlock" the normal "WAITING"
condition, i.e the bean is under load and all is normal.
I say this because of the "well it seems the container recovers" :) well, it
means the container is just handling your calls ;-)
but we never know with these things. for now this message doesn't bring us
developers anything and it seems to scare people that are doing serious
stress tests on JBoss -> gone!
marc
|-----Original Message-----
|From: [EMAIL PROTECTED]
|[mailto:[EMAIL PROTECTED]]On Behalf Of Ole Husgaard
|Sent: Sunday, January 28, 2001 10:01 PM
|To: jBoss
|Subject: Re: [jBoss-User] Possible Entity Bean Deadlock?
|
|
|Hi,
|
|Don't worry about the LOCKING-WAITING (TRANSACTION)
|message. Like Simon said, this is simply a warning
|that the container is waiting to avoid reentry of
|a bean.
|
|Back to original topic:
|Yes, this scenario will result in a deadlock, but
|as soon as one of the transactions time out, the
|other will continue and the deadlock will be over.
|
|
|Best Regards,
|
|Ole Husgaard.
|
|
|"Nortje, Andrew" wrote:
|>
|> I started a thread on Friday regarding the LOCKING-WAITING (TRANSACTION)
|> message that jBoss issues. This may be related to that.
|>
|> > -----Original Message-----
|> > From: Joshua M. White [mailto:[EMAIL PROTECTED]]
|> > Sent: Sunday, January 28, 2001 4:36 PM
|> > To: jBoss
|> > Subject: [jBoss-User] Possible Entity Bean Deadlock?
|> >
|> >
|> > I believe the EJB standard says that if a transaction has
|> > entity bean A
|> > in it, no other transactions may access entity bean A until the first
|> > transaction completes. Is this the default behavior on jboss? If not
|> > default, is it an option. This leads me to the question - If
|> > I have one
|> > transaction that grabs entity bean A while another transaction grabs
|> > entity bean B, and then the first transaction tries to get
|> > entity bean B
|> > while the second transaction tries to get entity bean A, will that
|> > result in deadlock? Or is the appserver smart enough to not even let
|> > this happen in the first place?
|> >
|> >
|> >
|> > --
|> > --------------------------------------------------------------
|> > To subscribe: [EMAIL PROTECTED]
|> > To unsubscribe: [EMAIL PROTECTED]
|> > List Help?: [EMAIL PROTECTED]
|> >
|>
|> --
|> --------------------------------------------------------------
|> To subscribe: [EMAIL PROTECTED]
|> To unsubscribe: [EMAIL PROTECTED]
|> List Help?: [EMAIL PROTECTED]
|
|
|--
|--------------------------------------------------------------
|To subscribe: [EMAIL PROTECTED]
|To unsubscribe: [EMAIL PROTECTED]
|List Help?: [EMAIL PROTECTED]
|
--
--------------------------------------------------------------
To subscribe: [EMAIL PROTECTED]
To unsubscribe: [EMAIL PROTECTED]
List Help?: [EMAIL PROTECTED]
