I have now tested to sleep 650 s which exceeds the default transaction timeout of 600 s. This gives me an exception due to transaction timeout. Thus Geronimo transaction timeout works.
I then configure the transaction timeout to 30 s (in the manner I've described previously) and sleep for 60 s in order to force a transaction timeout. I don't get a transaction timeout. To be sure I then checked to see if the configuration was bound and it wasn't! I had just made a fresh start of Karaf (by deleting the bundle cache). I also tried doing a refresh but the configuration was still unbound. I then restarted Karaf (without clearing the bundle cache) and after that the configuration was bound. However, even after my configuration was bound the test I did still did not provoke a transaction timeout. I think there must be something wrong with the way Aries transaction picks up its configuration - perhaps some timing issue. Also, even when the configuration is bound, it still doesn't seem to affect the transaction manager. Perhaps the transaction manager is created to soon and then not recreated when configuration updates are pushed by config admin? Can anyone explain this? Has anyone been able to configure the transaction manager at all? /Bengt 2012/1/13 Bengt Rodehav <[email protected]> > David, > > That test is almost identical to what I did in my code. The difference is > that I don't explicitly access the transaction manager like you do. I use > blueprint to make my method transactional (with the "Requires" transaction > attribute) and I configure the transaction timeout via Config Admin. > > I never doubted that Geronimo's transaction timeout worked but I wonder if > Aries is setting the properties correctly. I will make a test where I sleep > for longer than the default timeout (600 s) to see if that would trigger an > exception. If so, then the problem is how to configure the timeout > correctly. > > /Bengt > > 2012/1/13 David Jencks <[email protected]> > >> >> On Jan 12, 2012, at 11:08 PM, Bengt Rodehav wrote: >> >> David, do you think I should create a JIRA for this? I don't know for >> sure whether I'm doing anything wrong or not but I think it makes sense to >> add an integration test for the transaction timeout anyway. >> >> >> I looked and geronimo has a timeout test like this: >> >> public void testTimeout() throws Exception >> { >> long timeout = tm.getTransactionTimeoutMilliseconds(0L); >> tm.setTransactionTimeout((int)timeout/4000); >> tm.begin(); >> System.out.println("Test to sleep for " + timeout + " millisecs"); >> Thread.sleep(timeout); >> try >> { >> tm.commit(); >> fail("Tx Should get Rollback exception"); >> }catch(RollbackException rex) >> { >> // Caught expected exception >> } >> >> // Now test if the default timeout is active >> tm.begin(); >> System.out.println("Test to sleep for " + (timeout/2) + " >> millisecs"); >> Thread.sleep((timeout/2)); >> tm.commit(); >> // Its a failure if exception occurs. >> } >> >> >> So I tend to think timeout works. Is what's tested for here what you >> observe in your code? >> >> thanks >> david jencks >> >> >> I'm currently just using one resource manager (SQL Server 2005) which I >> guess means that no transaction log is necessary. Thanks for explaining. >> >> /Bengt >> >> 2012/1/13 David Jencks <[email protected]> >> >>> >>> On Jan 12, 2012, at 1:26 PM, Bengt Rodehav wrote: >>> >>> I took a look at the transaction related itests and I can't see any >>> tests regarding the transaction timeout - they seem to be focused on >>> testing the transaction attributes. Could it be that the transaction >>> timeout has never been tested? >>> >>> >>> probably not in aries since the tm implementation is from geronimo. I >>> did think there were some osgi tx timeout tests but I dont have access >>> right now. >>> >>> >>> Regarding the transaction log I'm not sure how this is supposed to work >>> - perhaps someone can enlighten me. Shouldn't every transaction be written >>> to a transaction log file? >>> >>> >>> Every successfully prepared transaction with at least two participating >>> resource managers will have a log record written for prepare and another >>> written for commit. Transaction with only one participant don't need >>> logging. >>> >>> thanks >>> david jencks >>> >>> >>> /Bengt >>> >>> 2012/1/12 Bengt Rodehav <[email protected]> >>> >>>> >>>> David, >>>> >>>> I have verified that the configuration is applied to the transaction >>>> service I just can't see that it has any effect. I guess I can start >>>> debugging but I was hoping to avoid that. Do you know if there are any >>>> integration tests that tests the transaction timeout? >>>> >>>> /Bengt >>>> >>>> >>>> Den torsdagen den 12:e januari 2012 skrev David Jencks< >>>> [email protected]>: >>>> >>>> > I would guess that perhaps your configuration is not actually getting >>>> supplied to the managed service? If you are running in karaf you can use >>>> confiig:list to make sure config admin is aware of your config. You might >>>> also want to debug the aries code and make sure that the configuration is >>>> actually getting supplied to the service. >>>> > david jencks >>>> > On Jan 12, 2012, at 6:36 AM, Bengt Rodehav wrote: >>>> > >>>> > Hello again Alasdair, >>>> > I finally got around to test configuration of Aries Transaction. I >>>> created the following configuration (using >>>> PID org.apache.aries.transaction): >>>> > aries.transaction.timeout=1 >>>> > aries.transaction.howl.logFileDir=${karaf.data}/txlog/ >>>> > I set the logfile directory as well as the transaction timeout (I >>>> peeked at how they do it in Servicemix). I set the timeout to a low value >>>> in order to test that I actually get a transaction timeout. I also perform >>>> a 5s sleep in a transacted method to force a transaction timeout. >>>> > But, I never get a transaction timeout and I never get a transaction >>>> log file in the "logFileDir" directory. I'm not beginning to fear that my >>>> methods are not transacted. How can I verify this? >>>> > Under what circumstances would a transaction log file be created? I >>>> assumed that it would be created when the first transaction was created. >>>> Or, do I also need to add the following line? >>>> > aries.transaction.recoverable=true >>>> > I've tried that but still no logfile was created. >>>> > I also noticed that the default value for >>>> "aries.transaction.howl.maxBuffers" is 0 while the default value for >>>> "aries.transaction.howl.minBuffers" is 4. Is that correct? At first sight I >>>> would think it should be the other way around. >>>> > /Bengt >>>> > >>>> > >>>> > >>>> > >>>> > >>>> > >>>> > 2011/12/29 Bengt Rodehav <[email protected]> >>>> > >>>> > Thanks Alasdair - will look into it. >>>> > /Bengt >>>> > >>>> > 2011/12/28 Alasdair Nottingham <[email protected]> >>>> > >>>> > Hi, >>>> > The pid for configuring transactions is >>>> "org.apache.aries.transaction". >>>> > Alasdair >>>> > >>>> > On 14 December 2011 14:33, Bengt Rodehav <[email protected]> wrote: >>>> > >>>> > It turned out that my error was not due to the transaction timing out >>>> but to violation of unique constraints. Nevertheless, I would like to know >>>> how to control the transaction timeout in Aries transaction. >>>> > /Bengt >>>> > >>>> > 2011/12/14 Bengt Rodehav <[email protected]> >>>> > >>>> > Thanks for your reply David, >>>> > I'll see if I can figure out the pid although this seems like >>>> something that really needs to be documented in Aries. If the default >>>> timeout is 600 seconds then this is probably not the reason of the errors I >>>> see. I need a time out of about 30 s which then is much less than the >>>> default. >>>> > I have been using MySql but I'm in the process of switching to SQL >>>> Server 2005. MySql worked fine but I started having problems committing the >>>> longer transactions with SQL Server 2005 which caused me to suspect a >>>> transaction timeout. Perhaps the timeout is not propagated to SQL Server >>>> like you hinted. >>>> > /Bengt >>>> > >>>> > 2011/12/14 David Jencks <[email protected]> >>>> > >>>> > Transaction is set up as a managed service factory. I haven't >>>> figured out exactly how this results in a tm instance without any visible >>>> configuration. >>>> > If you can figure out what is triggering the creation of a tm and the >>>> pid, the property to set is called aries.transaction.timeout and the >>>> default value is 600 (seconds) or 10 minutes. >>>> >>> >>> >>> >> >> >
