Re: [OJB-developers] Re: OJB in C/S mode: TransactionInProgressEx
Brought to you by:
thma
Menu
▾
▴
Re: [OJB-developers] Re: OJB in C/S mode: TransactionInProgressEx
|
From: <ri...@ya...> - 2002-01-31 15:02:43
|
Hello,
As I said before, when 2 users try to use the same part of the =
"manager", one of them gets an Exception, the other process it =
correctly.
Maybe it's because of the fact that I'm using ONE "odmg instance" =
(odmg=3DOJB.getInstance()) in my manager singleton. In every invocation =
of a function in this manager, I use:
Transaction tx =3D odmg.newTransaction()
So, aparently it's not a problem. But I'm not sure...
This is the complete stack trace, when 2 users access the same function:
-------------------------------------------------------------------------=
[ODMG] ERROR: null
org.odmg.QueryException
at ojb.odmg.oql.OQLQueryImpl.execute(OQLQueryImpl.java:251)
at =
charlie.ojb.ProyectoManagerOJB.tareasPorEmpleado(ProyectoManagerOJB.j
ava)
at org.apache.jsp.main$jsp._jspService(main$jsp.java:110)
at =
org.apache.jasper.runtime.HttpJspBase.service(HttpJspBase.java:107)
at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
at =
org.apache.jasper.servlet.JspServlet$JspServletWrapper.service(JspSer
vlet.java:202)
at =
org.apache.jasper.servlet.JspServlet.serviceJspFile(JspServlet.java:3
82)
at =
org.apache.jasper.servlet.JspServlet.service(JspServlet.java:474)
at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
at =
org.apache.catalina.core.ApplicationDispatcher.invoke(ApplicationDisp
atcher.java:679)
at =
org.apache.catalina.core.ApplicationDispatcher.doForward(ApplicationD
ispatcher.java:431)
at =
org.apache.catalina.core.ApplicationDispatcher.forward(ApplicationDis
patcher.java:355)
at =
org.apache.struts.action.ActionServlet.processActionForward(ActionSer
vlet.java:1759)
at =
org.apache.struts.action.ActionServlet.process(ActionServlet.java:159
6)
at =
org.apache.struts.action.ActionServlet.doPost(ActionServlet.java:510)
at javax.servlet.http.HttpServlet.service(HttpServlet.java:760)
at javax.servlet.http.HttpServlet.service(HttpServlet.java:853)
at =
org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(Appl
icationFilterChain.java:247)
at =
org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationF
ilterChain.java:193)
at =
org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperV
alve.java:243)
at =
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline
.java:566)
at =
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.jav
a:472)
at =
org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:943)
at =
org.apache.catalina.core.StandardContextValve.invoke(StandardContextV
alve.java:201)
at =
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline
.java:566)
at =
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.jav
a:472)
at =
org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:943)
at =
org.apache.catalina.core.StandardContext.invoke(StandardContext.java:
2344)
at =
org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.j
ava:164)
at =
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline
.java:566)
at =
org.apache.catalina.valves.ErrorDispatcherValve.invoke(ErrorDispatche
rValve.java:170)
at =
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline
.java:564)
at =
org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.j
ava:170)
at =
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline
.java:564)
at =
org.apache.catalina.valves.AccessLogValve.invoke(AccessLogValve.java:
462)
at =
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline
.java:564)
at =
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.jav
a:472)
at =
org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:943)
at =
org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineVal
ve.java:163)
at =
org.apache.catalina.core.StandardPipeline.invokeNext(StandardPipeline
.java:566)
at =
org.apache.catalina.core.StandardPipeline.invoke(StandardPipeline.jav
a:472)
at =
org.apache.catalina.core.ContainerBase.invoke(ContainerBase.java:943)
at =
org.apache.catalina.connector.http.HttpProcessor.process(HttpProcesso
r.java:1011)
at =
org.apache.catalina.connector.http.HttpProcessor.run(HttpProcessor.ja
va:1106)
at java.lang.Thread.run(Thread.java:484)
-------------------------------------------------------------------------=
---
I'm NOT using OJB in C/S mode. The "TransactionInProgressException =
problem" was related to OJB in C/S mode, this one is ANOTHER problem, as =
you can see in the satck trace.
I hope the patch helps solucionate the C/S problem, but this one seems =
to be another issue, am I right?
Thanks for your help!
Greetings,
Ricardo Arguello
ri...@ya...
----- Original Message -----=20
From: "Mahler Thomas" <tho...@it...>
To: "'Ricardo Arg=FCello'" <ri...@ya...>; "ojb" =
<obj...@li...>
Sent: Thursday, January 31, 2002 8:04 AM
Subject: AW: [OJB-developers] Re: OJB in C/S mode: =
TransactionInProgressEx
Hi again Ricardo,
The bugfix is on the way. I hope to get it posted today!
> -----Urspr=FCngliche Nachricht-----
> Von: Ricardo Arg=FCello [mailto:ri...@ya...]
> Gesendet: Mittwoch, 30. Januar 2002 22:43
> An: ojb
> Betreff: Re: [OJB-developers] Re: OJB in C/S mode:
> TransactionInProgressEx
>=20
>=20
> Hello,
>=20
> Thanks for your reply.
>=20
> I have a "manager class" wich is called from my Struts=20
> actions, and it does al the work, calling the "OJBed=20
> classes", retuning Collections, etc.
>=20
> This manager is a singleton, and when I instanciate it (only=20
> once) I initialize the "odmg" class variable:
>=20
> public class ProyectoManagerOJB {
> private static final String databaseName =3D
> =20
> PersistenceBrokerFactory.getConfiguration().getRepositoryFilename();
>=20
> private Implementation odmg; // maybe this is wrong!=20
This absolutely ok. OJB.getInstance() is a singleton
> public ProyectoManagerOJB() {
> odmg =3D OJB.getInstance();
> Database db =3D odmg.newDatabase();
> try {
> db.open(databaseName, Database.OPEN_READ_WRITE);
> } catch (ODMGException ex) {
> ex.printStackTrace();
> }
> }
> ...
> }
>=20
> Struts actions call functions in this class:
>=20
> ProyectoManager pm =3D ProyectoManager.getInstance(); // it's=20
> an abstract factory, returns ProyectoManagerOJB
> User user =3D pm.getUser("john");
>=20
> The corresponding class in ProyectoManagerOJB is this one:=20
> (User is an interface, UserOJB implements it)
>=20
> public User getUser(String user) {
> try {
> Transaction tx =3D odmg.newTransaction();
> tx.begin();
> OQLQuery query =3D odmg.newOQLQuery();
> query.create(
> "select p from "
> + charlie.ojb.UserOJB.class.getName()
> + " where user =3D \""
> + user
> + "\"");
> DList result =3D (DList) query.execute();
> User user =3D (User) result.get(0);
> tx.commit();
> return user;
> } catch (Exception e) {
> e.printStackTrace();
> return null;
> }
> }
>=20
> As you can see, in getUser() I use the class variable "odmg".=20
That's OK!
> If I have two users login in, I get a TransactionInProgressException.
Please send me a complete stacktrace of this exception. I'm not sure if =
it
is a PersistenceBrokerException or an ODMGException.
> I believe it is because of the fact that odmg is an instance variable.
>=20
> So, could you send me some code examples of your=20
> friends'"odmg helper class" to see how I could implement it=20
> in my architecture?
>=20
I'll will post it to you later this day.
> Thanks in advance!
>=20
> Greetings from Quito, Ecuador
>=20
greetings from Essen, Germany=20
Thomas
> Ricardo Arguello
> ri...@ya...
>=20
>=20
> ----- Original Message -----=20
> From: "Thomas Mahler" <tho...@ho...>
> To: "Ricardo Arguello" <ri...@ya...>; "ojb"=20
> <obj...@li...>
> Sent: Wednesday, January 30, 2002 3:35 PM
> Subject: [OJB-developers] Re: OJB in C/S mode: TransactionInProgressEx
>=20
>=20
> > Hi Ricardo,
> >=20
> > sorry for the trouble OJB is causing for you.
> > We'll try to get a solution for this soon.
> >=20
> > Ricardo Arguello wrote:
> >=20
> > > Hello,
> > >=20
> > > As I said before, I'm getting a=20
> > > TransactionInProgressException when I use OJB in=20
> > > Client/Server mode.
> > >=20
> > > I tried changing the log to DEBUG, bit it imposibe for me=20
> > > to distinguish any of the errors, there is a LOT of DEBUG=20
> > > messages, of the following style:
> > >=20
> >=20
> >=20
> >=20
> > Yup, sometimes it's hard to know which part of the system=20
> might write=20
> > the relevant log infos...
> >=20
> >=20
> > >=20
> > > [ojb.broker.server.RequestProcessor] DEBUG: OJB=20
> > > RequestProcessor 2 waiting...
> > > [ojb.broker.server.RequestProcessor] DEBUG: OJB=20
> > > RequestProcessor 7 working...
> > > [ojb.broker.server.RequestProcessor] DEBUG: dispatching [24,
> > > [ojb.broker.server.RequestProcessor] DEBUG: -1,
> > > [ojb.broker.server.RequestProcessor] DEBUG: class=20
> > > charlie.ojb.EmpleadoOJB,
> > > [ojb.broker.server.RequestProcessor] DEBUG: null]
> > > [ojb.broker.server.BrokerPool] DEBUG: getFreeBrokerId: 1,=20
> > > remaining: [1, 1, 1, 1
> > > , 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 3, 1, 1, 1, 1, 1, 3,=20
> > > 1, 1, 1, 1, 1, 1, 3,
> > > 1, 1, 1, 1, 1, 1, 3, 1, 1, 1, 1, 1, 1, 3, 1, 1, 1, 1, 1, 1,=20
> > > 1, 1, 1, 1, 1, 1, 1,
> > > 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,=20
> > > 1, 1, 1, 1, 1, 1, 1, 1
> > > , 1, 1, 4, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,=20
> > > 1, 1, 1, 5, 1, 1, 1,
> > > 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 6, 1,=20
> > > 1, 1, 1, 1, 1, 1, 1,
> > > 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 7, 1, 1, 1, 8,=20
> > > 1, 1, 1, 9, 1, 1, 9, 1
> > > , 1, 1, 1, 1, 1, 0, 1, 9, 1, 1, 1, 1, 9, 1, 1, 1, 1, 1, 1,=20
> > > 1, 1, 1, 1, 1, 1, 1,
> > > 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1, 1,=20
> > > 1, 1, 1, 1, 1, 1, 1,
> > > 1, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,=20
> > > 1, 1, 1, 1, 1, 1, 1, 1
> > > , 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,=20
> > > 1, 1, 1, 1, 1, 1, 1,
> > > 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
> > >=20
> > >=20
> > > Awful, right?
> >=20
> >=20
> > Yes that's awful to read. But it is also a clear indicator=20
> that there=20
> > must be a bug in the BrokerPool. It should contain exactly=20
> ten brokers=20
> > numbered from 0 - 9.
> > As the list you see contains a lot of "1" entries you are=20
> served with=20
> > the same broker instance for multiple transactions.
> >=20
> > This is definitely a severe bug.
> >=20
> > I will start working on it asap and hope to have a fix by=20
> the weekend.
> >=20
> >=20
> > >=20
> > > I cannot find the errors I metioned in my previous=20
> > > message... But my application stil acts the same: Two=20
> > > persons can't even login at the same time, one of them gets=20
> > > a stack trace...
> > >=20
> >=20
> >=20
> > I assume that it is all related to the above error !
> >=20
> >=20
> > > I was reading the Locking document, and I tried to add a=20
> > > isolation=3D"read-uncommitted" to all my class definitions in=20
> > > the repository.xml file, but tha doesn't help either.
> >=20
> >=20
> > This is definitely a locking problem. OJB uses read-uncommited by=20
> > default already. In the case of locking problems you would=20
> get a ODMG=20
> > LockNotGrantedException!
> >=20
> > > How=20
> > > should I declare my lockings? In the XML file, in my code=20
> > > (you say the ODMG doesnt dictate a way to that).
> > >=20
> >=20
> >=20
> > Locking is declared in the repository only. ODMG specifies=20
> no API for=20
> > setting locklevels.
> >=20
> >=20
> > >=20
> > > Please clarify this to me? Do I need to do C/S to use OJB=20
> > > in a web application?
> >=20
> >=20
> >=20
> > Hmm. I know of people who are using OJB in singlevm mode in=20
> Web apps.
> > As long as the commit phase of ODMG transactions is=20
> relatively short=20
> > there won't be much problems.
> >=20
> > I'm intending to have brokerpool for the singlevm mode too.=20
> then it will=20
> > be totally safe to use the singlevm mode which is much faster !!!
> >=20
> >=20
> > > Wouldn't a "OJB Helper" with a single Database instance=20
> > > (singleton) be enough?
> > >=20
> >=20
> >=20
> > Yes that should be sufficient.
> >=20
> >=20
> > > If that's the case, how could I implement that?=20
> >=20
> >=20
> > Hmm, I'm not quite sure what you want to implement exactly?
> >=20
> >=20
> > > Is OJB thread safe?=20
> >=20
> >=20
> > Yes
> >=20
> > >How can I use OJB with Tomcat + Struts?
> >=20
> >=20
> > A colleague of mine used Struts and the OJB ODMG api sucessfully.
> > He defined special STruts actions that used ODMG=20
> transactions to do the=20
> > persistence work. It's pretty straightforward.
> >=20
> > If you are interested I might post you sample code for this.
> >=20
> >=20
> > > How can I assure that 2 transactions can READ at the same=20
> > > time, without locking?
> > >=20
> >=20
> >=20
> > Currently the OJB ODMG implementation relies on pessimistic locking.
> > There is currently no optimistic locking support. (will be=20
> there in 1.0)
> >=20
> > But if you use "uncommitted-read" there should be no=20
> locking problems.
> > You can do even more do avoid lock conflicts through WRITE=20
> operations.
> > In OJB.properties you'll find an entry LockAssociations set=20
> this to READ.
> >=20
> > The LockAssociations entry defines the behaviour for the OJB
> > implicit locking feature. If set to WRITE (default)=20
> acquiring a write-
> > lock on a given object x implies write locks on all objects=20
> associated
> > to x. If set to READ implicit read-locks are acquired.
> > Acquiring a read-lock on x thus allways results in implicit=20
> read-locks
> > on all associated objects.
> >=20
> > Using uncommited reads in combination with=20
> LockAssociation=3DREAD will=20
> > reduce Locking related problems to a minimum.
> >=20
> > But as I said before: your problem is NOT related to locking.
> >=20
> >=20
> > >=20
> > > Thanks in advance.
> > >=20
> > > Ricardo Arguello (an ex-Castor refugee)
> >=20
> >=20
> > I'll work hard to ensure you won't become an OJB refugee ;-)
> >=20
> >=20
> > cu,
> >=20
> > Thomas
> >=20
> >=20
> >=20
> > _______________________________________________
> > Objectbridge-developers mailing list
> > Obj...@li...
> > https://lists.sourceforge.net/lists/listinfo/objectbridge-developers
> >=20
>=20
>=20
>=20
> _______________________________________________
> Objectbridge-developers mailing list
> Obj...@li...
> https://lists.sourceforge.net/lists/listinfo/objectbridge-developers
>=20
_______________________________________________
Objectbridge-developers mailing list
Obj...@li...
https://lists.sourceforge.net/lists/listinfo/objectbridge-developers
|
