Re: Deadlocks in the Filestore of Silde 2.1
Hi Arne! Your observation really is interesting. Any idea why the request is not terminated even if the client finally gives up? If this really is the case, even my non-blocking "solution" would not do, as the non-blocking lock still remains. A better solution would be a lock that times out after some time. AFAIK there is nothing simliar implemented, yet. You can switch off locking by setting the mode to "none". In that case the underlying store is responsible for all locking. If it is a DB that should be ok, though. HTH Oliver 2007/4/18, Arne v.Irmer <[EMAIL PROTECTED]>: Hi Oliver, how can I switch off locking and what happens in the case of a concurrent file access? Might it be a good idea to replace the locking code of 2.1 with code of 2.2? Yours Arne Oliver Zeigermann schrieb: > I was just having a look at the 2.2 code and it looks that you can > even switch off locking completely. Otherwise the locking code in 2.2 > looks much better than the code in 2.1 > > Oliver > > 2007/4/17, Oliver Zeigermann <[EMAIL PROTECTED]>: >> Hi Arne! >> >> The stack trace tells you that another thread has the global >> read/write lock. So, correctly speaking this is a "live lock". The >> reason most likely is a thread that still runs and has forgotten to >> release this lock. >> >> Having said this, even switching to a deadlock free solution will not >> help, as you have no deadlocks here. >> >> Cheers >> >> Oliver >> >> 2007/4/17, Arne v.Irmer <[EMAIL PROTECTED]>: >> > Hi, >> > we are using slide in our e-learning platform >> > http://ews2.uni-dortmund.de in a production environment. We get into >> > deep trouble with deadlocks in Slide! >> > I checked out Silde 2.2pre1. In the RELEASE-NOTES-2.2 there is the >> sentence: >> > "- New option for a fine grained, deadlock-less locking scheme" >> > My question is: >> > - Should I change to 2.2pre1 for my productive environment? How can I >> > switch back if we discover other problems with this pre-release? >> > - Is there a patch for version 2.1. that fixes these deadlocks? >> > >> > Please help. >> > Thanks in advance >> > Arne v.Irmer >> > >> > Our configuration: >> > Slide 2.1 with fixed hasmap "suspendedLocks" in >> > org.apache.slide.store.ExtendedStore. >> > using a filestore in Tomcat 5.5 >> > >> > Here are some stacktraces from the jconsole in the case of the >> deadlocks: >> > >> > Name: http-129.217.205.43-8080-Processor3 >> > State: TIMED_WAITING on >> > [EMAIL PROTECTED] >> > Total blocked: 273 Total waited: 332 >> > >> > Stack trace: >> > java.lang.Object.wait(Native Method) >> > >> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241) >> >> > >> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165) >> >> > >> org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalWriteLock(AbstractWebdavMethod.java:847) >> >> > >> org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:362) >> >> > org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155) >> > javax.servlet.http.HttpServlet.service(HttpServlet.java:802) >> > >> org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252) >> >> > >> org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) >> >> > org.apache.slide.webdav.filter.LogFilter.doFilter(LogFilter.java:141) >> > >> org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:202) >> >> > >> org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) >> >> > >> org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213) >> >> > >> org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:178) >> >> > >> org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:524) >> >> > >> org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:126) >> >> > >> org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:105) >> >> > >> org.apache.catalina.authenticator.SingleSignOn.invoke(SingleSignOn.java:419) >> >> > >> org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:107) >> >> > >> org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:148) >> >> > >> org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:799) >> >> > >> org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.processConnection(Http11Protocol.java:705) >> >> > >> org.apache.tomcat.util.net.PoolTcpEndpoint.processSocket(PoolTcpEndpoint.java:527) >> >> > >> org.apache.tomcat.util.net.LeaderFollowerWorkerThread.runIt(LeaderFollowerWorkerThread.java:80) >> >> > >> org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:684) >> >> > java.lang.Thread.run(Thread.java:595) >> > >> > Name: http-129.217.205.43-8080-Processor3 >> > State: TIMED_WAITING on >> > [EMAIL PR
Re: Deadlocks in the Filestore of Silde 2.1
Hi Oliver, how can I switch off locking and what happens in the case of a concurrent file access? Might it be a good idea to replace the locking code of 2.1 with code of 2.2? Yours Arne Oliver Zeigermann schrieb: > I was just having a look at the 2.2 code and it looks that you can > even switch off locking completely. Otherwise the locking code in 2.2 > looks much better than the code in 2.1 > > Oliver > > 2007/4/17, Oliver Zeigermann <[EMAIL PROTECTED]>: >> Hi Arne! >> >> The stack trace tells you that another thread has the global >> read/write lock. So, correctly speaking this is a "live lock". The >> reason most likely is a thread that still runs and has forgotten to >> release this lock. >> >> Having said this, even switching to a deadlock free solution will not >> help, as you have no deadlocks here. >> >> Cheers >> >> Oliver >> >> 2007/4/17, Arne v.Irmer <[EMAIL PROTECTED]>: >> > Hi, >> > we are using slide in our e-learning platform >> > http://ews2.uni-dortmund.de in a production environment. We get into >> > deep trouble with deadlocks in Slide! >> > I checked out Silde 2.2pre1. In the RELEASE-NOTES-2.2 there is the >> sentence: >> > "- New option for a fine grained, deadlock-less locking scheme" >> > My question is: >> > - Should I change to 2.2pre1 for my productive environment? How can I >> > switch back if we discover other problems with this pre-release? >> > - Is there a patch for version 2.1. that fixes these deadlocks? >> > >> > Please help. >> > Thanks in advance >> > Arne v.Irmer >> > >> > Our configuration: >> > Slide 2.1 with fixed hasmap "suspendedLocks" in >> > org.apache.slide.store.ExtendedStore. >> > using a filestore in Tomcat 5.5 >> > >> > Here are some stacktraces from the jconsole in the case of the >> deadlocks: >> > >> > Name: http-129.217.205.43-8080-Processor3 >> > State: TIMED_WAITING on >> > [EMAIL PROTECTED] >> > Total blocked: 273 Total waited: 332 >> > >> > Stack trace: >> > java.lang.Object.wait(Native Method) >> > >> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241) >> >> > >> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165) >> >> > >> org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalWriteLock(AbstractWebdavMethod.java:847) >> >> > >> org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:362) >> >> > org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155) >> > javax.servlet.http.HttpServlet.service(HttpServlet.java:802) >> > >> org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252) >> >> > >> org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) >> >> > org.apache.slide.webdav.filter.LogFilter.doFilter(LogFilter.java:141) >> > >> org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:202) >> >> > >> org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) >> >> > >> org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213) >> >> > >> org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:178) >> >> > >> org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:524) >> >> > >> org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:126) >> >> > >> org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:105) >> >> > >> org.apache.catalina.authenticator.SingleSignOn.invoke(SingleSignOn.java:419) >> >> > >> org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:107) >> >> > >> org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:148) >> >> > >> org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:799) >> >> > >> org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.processConnection(Http11Protocol.java:705) >> >> > >> org.apache.tomcat.util.net.PoolTcpEndpoint.processSocket(PoolTcpEndpoint.java:527) >> >> > >> org.apache.tomcat.util.net.LeaderFollowerWorkerThread.runIt(LeaderFollowerWorkerThread.java:80) >> >> > >> org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:684) >> >> > java.lang.Thread.run(Thread.java:595) >> > >> > Name: http-129.217.205.43-8080-Processor3 >> > State: TIMED_WAITING on >> > [EMAIL PROTECTED] >> > Total blocked: 44 Total waited: 80 >> > >> > Stack trace: >> > java.lang.Object.wait(Native Method) >> > >> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241) >> >> > >> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165) >> >> > >> org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalReadLock(AbstractWebdavMethod.java:838) >> >> > >> org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:360) >> >> > org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155) >> > javax.servle
Re: Deadlocks in the Filestore of Silde 2.1
Hi Oliver, thank you for widening my view on that problem. To understand the mechanism of locks I made a small test with slide 2.1. And maybe I found a bug in the locking mechanism that could cause my problem: 1. I downloaded a big file from a test slide with a client that is slowdowned. 2. From another client I do a write access (I tried to delete the file) -> that was locked. (ok!) I saw a process waiting with the same stacktrace in jconsole as I found on my operational system when it stops working. 3. Now the client that wants to delete gives up (timeout etc.) In jconsole one can see, that lock wasn't removed. One process is waiting, even if the reading client gives up, too. 4. Any write access to that file produces a waiting process. If the number of 100 waiting processes is reached, the tomcat goes offline. Maybe this works with a readlock, too. But I don't know how to initiate readlock. ;-) Yours Arne Oliver Zeigermann schrieb: > Hi Arne! > > The stack trace tells you that another thread has the global > read/write lock. So, correctly speaking this is a "live lock". The > reason most likely is a thread that still runs and has forgotten to > release this lock. > > Having said this, even switching to a deadlock free solution will not > help, as you have no deadlocks here. > > Cheers > > Oliver > > 2007/4/17, Arne v.Irmer <[EMAIL PROTECTED]>: >> Hi, >> we are using slide in our e-learning platform >> http://ews2.uni-dortmund.de in a production environment. We get into >> deep trouble with deadlocks in Slide! >> I checked out Silde 2.2pre1. In the RELEASE-NOTES-2.2 there is the >> sentence: >> "- New option for a fine grained, deadlock-less locking scheme" >> My question is: >> - Should I change to 2.2pre1 for my productive environment? How can I >> switch back if we discover other problems with this pre-release? >> - Is there a patch for version 2.1. that fixes these deadlocks? >> >> Please help. >> Thanks in advance >> Arne v.Irmer >> >> Our configuration: >> Slide 2.1 with fixed hasmap "suspendedLocks" in >> org.apache.slide.store.ExtendedStore. >> using a filestore in Tomcat 5.5 >> >> Here are some stacktraces from the jconsole in the case of the >> deadlocks: >> >> Name: http-129.217.205.43-8080-Processor3 >> State: TIMED_WAITING on >> [EMAIL PROTECTED] >> Total blocked: 273 Total waited: 332 >> >> Stack trace: >> java.lang.Object.wait(Native Method) >> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241) >> >> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165) >> >> org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalWriteLock(AbstractWebdavMethod.java:847) >> >> org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:362) >> >> org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155) >> javax.servlet.http.HttpServlet.service(HttpServlet.java:802) >> org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252) >> >> org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) >> >> org.apache.slide.webdav.filter.LogFilter.doFilter(LogFilter.java:141) >> org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:202) >> >> org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) >> >> org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213) >> >> org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:178) >> >> org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:524) >> >> org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:126) >> >> org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:105) >> >> org.apache.catalina.authenticator.SingleSignOn.invoke(SingleSignOn.java:419) >> >> org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:107) >> >> org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:148) >> >> org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:799) >> >> org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.processConnection(Http11Protocol.java:705) >> >> org.apache.tomcat.util.net.PoolTcpEndpoint.processSocket(PoolTcpEndpoint.java:527) >> >> org.apache.tomcat.util.net.LeaderFollowerWorkerThread.runIt(LeaderFollowerWorkerThread.java:80) >> >> org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:684) >> >> java.lang.Thread.run(Thread.java:595) >> >> Name: http-129.217.205.43-8080-Processor3 >> State: TIMED_WAITING on >> [EMAIL PROTECTED] >> Total blocked: 44 Total waited: 80 >> >> Stack trace: >> java.lang.Object.wait(Native Method) >> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241) >> >> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165) >> >> org.apache.slide.webdav.me
Re: Deadlocks in the Filestore of Silde 2.1
Hi Oliver,
thanks for the immediate response.
Does this imply, I should change to version 2.2_pre1 for our production
server? I saw in SVN, you implemented "fail-fast" today. ;-)
Yours
Arne
Oliver Zeigermann schrieb:
> Using the latest source from SVN you can set the "sequential-mode" to
> "fail-fast" now. This checks for read/write conflicts, but does not
> block anything in case of a conflict. It will instead fail immedeatly
> with a conflict error.
>
> HTH
>
> Oliver
>
> 2007/4/18, aslam bari <[EMAIL PROTECTED]>:
>> I was also facing this problem before. After that i did a small
>> change and now it is working fine. What i did is:-
>>
>> 1) In domain.xml change sequential-mode parameter to "full" from
>> "fine-grain" like this:-
>> full
>>
>> 2)If you are using (Lucenen Indexing) then In Domain.xml use
>> asyncron mode for this and also set priority to lowest like 1:- like
>> this:-
>>
>>
>> > classname="org.apache.slide.index.lucene.LuceneContentIndexer">
>> > name="indexpath">${filespath}/../data/store/index_content
>> false
>> 1
>> > name="analyzer">org.apache.lucene.analysis.standard.StandardAnalyzer
>>
>> 10
>>
>>
>> Thanks
>>
>>
>> - Original Message
>> From: Oliver Zeigermann <[EMAIL PROTECTED]>
>> To: Slide Users Mailing List ;
>> [EMAIL PROTECTED]
>> Sent: Wednesday, 18 April, 2007 12:31:42 AM
>> Subject: Re: Deadlocks in the Filestore of Silde 2.1
>>
>>
>> I was just having a look at the 2.2 code and it looks that you can
>> even switch off locking completely. Otherwise the locking code in 2.2
>> looks much better than the code in 2.1
>>
>> Oliver
>>
>> 2007/4/17, Oliver Zeigermann <[EMAIL PROTECTED]>:
>> > Hi Arne!
>> >
>> > The stack trace tells you that another thread has the global
>> > read/write lock. So, correctly speaking this is a "live lock". The
>> > reason most likely is a thread that still runs and has forgotten to
>> > release this lock.
>> >
>> > Having said this, even switching to a deadlock free solution will not
>> > help, as you have no deadlocks here.
>> >
>> > Cheers
>> >
>> > Oliver
>> >
>> > 2007/4/17, Arne v.Irmer <[EMAIL PROTECTED]>:
>> > > Hi,
>> > > we are using slide in our e-learning platform
>> > > http://ews2.uni-dortmund.de in a production environment. We get into
>> > > deep trouble with deadlocks in Slide!
>> > > I checked out Silde 2.2pre1. In the RELEASE-NOTES-2.2 there is
>> the sentence:
>> > > "- New option for a fine grained, deadlock-less locking scheme"
>> > > My question is:
>> > > - Should I change to 2.2pre1 for my productive environment? How
>> can I
>> > > switch back if we discover other problems with this pre-release?
>> > > - Is there a patch for version 2.1. that fixes these deadlocks?
>> > >
>> > > Please help.
>> > > Thanks in advance
>> > > Arne v.Irmer
>> > >
>> > > Our configuration:
>> > > Slide 2.1 with fixed hasmap "suspendedLocks" in
>> > > org.apache.slide.store.ExtendedStore.
>> > > using a filestore in Tomcat 5.5
>> > >
>> > > Here are some stacktraces from the jconsole in the case of the
>> deadlocks:
>> > >
>> > > Name: http-129.217.205.43-8080-Processor3
>> > > State: TIMED_WAITING on
>> > > [EMAIL PROTECTED]
>> > > Total blocked: 273 Total waited: 332
>> > >
>> > > Stack trace:
>> > > java.lang.Object.wait(Native Method)
>> > >
>> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241)
>>
>> > >
>> org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165)
>>
>> > >
>> org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalWriteLock(AbstractWebdavMethod.java:847)
>>
>> > >
>> org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:362)
>>
>> > >
>> org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155)
>> > > javax.servlet.http.HttpServlet.service(HttpServlet.java:802)
>> > >
>> org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.j
Re: Deadlocks in the Filestore of Silde 2.1
Using the latest source from SVN you can set the "sequential-mode" to
"fail-fast" now. This checks for read/write conflicts, but does not
block anything in case of a conflict. It will instead fail immedeatly
with a conflict error.
HTH
Oliver
2007/4/18, aslam bari <[EMAIL PROTECTED]>:
I was also facing this problem before. After that i did a small change and now
it is working fine. What i did is:-
1) In domain.xml change sequential-mode parameter to "full" from "fine-grain"
like this:-
full
2)If you are using (Lucenen Indexing) then In Domain.xml use asyncron mode
for this and also set priority to lowest like 1:- like this:-
${filespath}/../data/store/index_content
false
1
org.apache.lucene.analysis.standard.StandardAnalyzer
10
Thanks
- Original Message
From: Oliver Zeigermann <[EMAIL PROTECTED]>
To: Slide Users Mailing List ; [EMAIL PROTECTED]
Sent: Wednesday, 18 April, 2007 12:31:42 AM
Subject: Re: Deadlocks in the Filestore of Silde 2.1
I was just having a look at the 2.2 code and it looks that you can
even switch off locking completely. Otherwise the locking code in 2.2
looks much better than the code in 2.1
Oliver
2007/4/17, Oliver Zeigermann <[EMAIL PROTECTED]>:
> Hi Arne!
>
> The stack trace tells you that another thread has the global
> read/write lock. So, correctly speaking this is a "live lock". The
> reason most likely is a thread that still runs and has forgotten to
> release this lock.
>
> Having said this, even switching to a deadlock free solution will not
> help, as you have no deadlocks here.
>
> Cheers
>
> Oliver
>
> 2007/4/17, Arne v.Irmer <[EMAIL PROTECTED]>:
> > Hi,
> > we are using slide in our e-learning platform
> > http://ews2.uni-dortmund.de in a production environment. We get into
> > deep trouble with deadlocks in Slide!
> > I checked out Silde 2.2pre1. In the RELEASE-NOTES-2.2 there is the sentence:
> > "- New option for a fine grained, deadlock-less locking scheme"
> > My question is:
> > - Should I change to 2.2pre1 for my productive environment? How can I
> > switch back if we discover other problems with this pre-release?
> > - Is there a patch for version 2.1. that fixes these deadlocks?
> >
> > Please help.
> > Thanks in advance
> > Arne v.Irmer
> >
> > Our configuration:
> > Slide 2.1 with fixed hasmap "suspendedLocks" in
> > org.apache.slide.store.ExtendedStore.
> > using a filestore in Tomcat 5.5
> >
> > Here are some stacktraces from the jconsole in the case of the deadlocks:
> >
> > Name: http-129.217.205.43-8080-Processor3
> > State: TIMED_WAITING on
> > [EMAIL PROTECTED]
> > Total blocked: 273 Total waited: 332
> >
> > Stack trace:
> > java.lang.Object.wait(Native Method)
> >
org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241)
> >
org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165)
> >
org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalWriteLock(AbstractWebdavMethod.java:847)
> >
org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:362)
> > org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155)
> > javax.servlet.http.HttpServlet.service(HttpServlet.java:802)
> >
org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252)
> >
org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173)
> > org.apache.slide.webdav.filter.LogFilter.doFilter(LogFilter.java:141)
> >
org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:202)
> >
org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173)
> >
org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213)
> >
org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:178)
> >
org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:524)
> >
org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:126)
> >
org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:105)
> > org.apache.catalina.authenticator.SingleSignOn.invoke(SingleSignOn.java:419)
> >
org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:107)
> > org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:148)
> > org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:799)
> >
org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.processConnection(Http11Protocol.java:70
Re: Deadlocks in the Filestore of Silde 2.1
I was also facing this problem before. After that i did a small change and now
it is working fine. What i did is:-
1) In domain.xml change sequential-mode parameter to "full" from "fine-grain"
like this:-
full
2)If you are using (Lucenen Indexing) then In Domain.xml use asyncron mode
for this and also set priority to lowest like 1:- like this:-
${filespath}/../data/store/index_content
false
1
org.apache.lucene.analysis.standard.StandardAnalyzer
10
Thanks
- Original Message
From: Oliver Zeigermann <[EMAIL PROTECTED]>
To: Slide Users Mailing List ; [EMAIL PROTECTED]
Sent: Wednesday, 18 April, 2007 12:31:42 AM
Subject: Re: Deadlocks in the Filestore of Silde 2.1
I was just having a look at the 2.2 code and it looks that you can
even switch off locking completely. Otherwise the locking code in 2.2
looks much better than the code in 2.1
Oliver
2007/4/17, Oliver Zeigermann <[EMAIL PROTECTED]>:
> Hi Arne!
>
> The stack trace tells you that another thread has the global
> read/write lock. So, correctly speaking this is a "live lock". The
> reason most likely is a thread that still runs and has forgotten to
> release this lock.
>
> Having said this, even switching to a deadlock free solution will not
> help, as you have no deadlocks here.
>
> Cheers
>
> Oliver
>
> 2007/4/17, Arne v.Irmer <[EMAIL PROTECTED]>:
> > Hi,
> > we are using slide in our e-learning platform
> > http://ews2.uni-dortmund.de in a production environment. We get into
> > deep trouble with deadlocks in Slide!
> > I checked out Silde 2.2pre1. In the RELEASE-NOTES-2.2 there is the sentence:
> > "- New option for a fine grained, deadlock-less locking scheme"
> > My question is:
> > - Should I change to 2.2pre1 for my productive environment? How can I
> > switch back if we discover other problems with this pre-release?
> > - Is there a patch for version 2.1. that fixes these deadlocks?
> >
> > Please help.
> > Thanks in advance
> > Arne v.Irmer
> >
> > Our configuration:
> > Slide 2.1 with fixed hasmap "suspendedLocks" in
> > org.apache.slide.store.ExtendedStore.
> > using a filestore in Tomcat 5.5
> >
> > Here are some stacktraces from the jconsole in the case of the deadlocks:
> >
> > Name: http-129.217.205.43-8080-Processor3
> > State: TIMED_WAITING on
> > [EMAIL PROTECTED]
> > Total blocked: 273 Total waited: 332
> >
> > Stack trace:
> > java.lang.Object.wait(Native Method)
> > org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241)
> > org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165)
> > org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalWriteLock(AbstractWebdavMethod.java:847)
> > org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:362)
> > org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155)
> > javax.servlet.http.HttpServlet.service(HttpServlet.java:802)
> > org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252)
> > org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173)
> > org.apache.slide.webdav.filter.LogFilter.doFilter(LogFilter.java:141)
> > org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:202)
> > org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173)
> > org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213)
> > org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:178)
> > org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:524)
> > org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:126)
> > org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:105)
> > org.apache.catalina.authenticator.SingleSignOn.invoke(SingleSignOn.java:419)
> > org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:107)
> > org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:148)
> > org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:799)
> > org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.processConnection(Http11Protocol.java:705)
> > org.apache.tomcat.util.net.PoolTcpEndpoint.processSocket(PoolTcpEndpoint.java:527)
> > org.apache.tomcat.util.net.LeaderFollowerWorkerThread.runIt(LeaderFollowerWorkerThread.java:80)
> > org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(Th
Re: Deadlocks in the Filestore of Silde 2.1
I was just having a look at the 2.2 code and it looks that you can even switch off locking completely. Otherwise the locking code in 2.2 looks much better than the code in 2.1 Oliver 2007/4/17, Oliver Zeigermann <[EMAIL PROTECTED]>: Hi Arne! The stack trace tells you that another thread has the global read/write lock. So, correctly speaking this is a "live lock". The reason most likely is a thread that still runs and has forgotten to release this lock. Having said this, even switching to a deadlock free solution will not help, as you have no deadlocks here. Cheers Oliver 2007/4/17, Arne v.Irmer <[EMAIL PROTECTED]>: > Hi, > we are using slide in our e-learning platform > http://ews2.uni-dortmund.de in a production environment. We get into > deep trouble with deadlocks in Slide! > I checked out Silde 2.2pre1. In the RELEASE-NOTES-2.2 there is the sentence: > "- New option for a fine grained, deadlock-less locking scheme" > My question is: > - Should I change to 2.2pre1 for my productive environment? How can I > switch back if we discover other problems with this pre-release? > - Is there a patch for version 2.1. that fixes these deadlocks? > > Please help. > Thanks in advance > Arne v.Irmer > > Our configuration: > Slide 2.1 with fixed hasmap "suspendedLocks" in > org.apache.slide.store.ExtendedStore. > using a filestore in Tomcat 5.5 > > Here are some stacktraces from the jconsole in the case of the deadlocks: > > Name: http-129.217.205.43-8080-Processor3 > State: TIMED_WAITING on > [EMAIL PROTECTED] > Total blocked: 273 Total waited: 332 > > Stack trace: > java.lang.Object.wait(Native Method) > org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241) > org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165) > org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalWriteLock(AbstractWebdavMethod.java:847) > org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:362) > org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155) > javax.servlet.http.HttpServlet.service(HttpServlet.java:802) > org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252) > org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) > org.apache.slide.webdav.filter.LogFilter.doFilter(LogFilter.java:141) > org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:202) > org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) > org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213) > org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:178) > org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:524) > org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:126) > org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:105) > org.apache.catalina.authenticator.SingleSignOn.invoke(SingleSignOn.java:419) > org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:107) > org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:148) > org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:799) > org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.processConnection(Http11Protocol.java:705) > org.apache.tomcat.util.net.PoolTcpEndpoint.processSocket(PoolTcpEndpoint.java:527) > org.apache.tomcat.util.net.LeaderFollowerWorkerThread.runIt(LeaderFollowerWorkerThread.java:80) > org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:684) > java.lang.Thread.run(Thread.java:595) > > Name: http-129.217.205.43-8080-Processor3 > State: TIMED_WAITING on > [EMAIL PROTECTED] > Total blocked: 44 Total waited: 80 > > Stack trace: > java.lang.Object.wait(Native Method) > org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241) > org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165) > org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalReadLock(AbstractWebdavMethod.java:838) > org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:360) > org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155) > javax.servlet.http.HttpServlet.service(HttpServlet.java:802) > org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252) > org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) > org.apache.slide.webdav.filter.LogFilter.doFilter(LogFilter.java:141) > org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:202) > org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) > org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213) > org.apa
Re: Deadlocks in the Filestore of Silde 2.1
Hi Arne! The stack trace tells you that another thread has the global read/write lock. So, correctly speaking this is a "live lock". The reason most likely is a thread that still runs and has forgotten to release this lock. Having said this, even switching to a deadlock free solution will not help, as you have no deadlocks here. Cheers Oliver 2007/4/17, Arne v.Irmer <[EMAIL PROTECTED]>: Hi, we are using slide in our e-learning platform http://ews2.uni-dortmund.de in a production environment. We get into deep trouble with deadlocks in Slide! I checked out Silde 2.2pre1. In the RELEASE-NOTES-2.2 there is the sentence: "- New option for a fine grained, deadlock-less locking scheme" My question is: - Should I change to 2.2pre1 for my productive environment? How can I switch back if we discover other problems with this pre-release? - Is there a patch for version 2.1. that fixes these deadlocks? Please help. Thanks in advance Arne v.Irmer Our configuration: Slide 2.1 with fixed hasmap "suspendedLocks" in org.apache.slide.store.ExtendedStore. using a filestore in Tomcat 5.5 Here are some stacktraces from the jconsole in the case of the deadlocks: Name: http-129.217.205.43-8080-Processor3 State: TIMED_WAITING on [EMAIL PROTECTED] Total blocked: 273 Total waited: 332 Stack trace: java.lang.Object.wait(Native Method) org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241) org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165) org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalWriteLock(AbstractWebdavMethod.java:847) org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:362) org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155) javax.servlet.http.HttpServlet.service(HttpServlet.java:802) org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252) org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) org.apache.slide.webdav.filter.LogFilter.doFilter(LogFilter.java:141) org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:202) org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213) org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:178) org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:524) org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:126) org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:105) org.apache.catalina.authenticator.SingleSignOn.invoke(SingleSignOn.java:419) org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:107) org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:148) org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:799) org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.processConnection(Http11Protocol.java:705) org.apache.tomcat.util.net.PoolTcpEndpoint.processSocket(PoolTcpEndpoint.java:527) org.apache.tomcat.util.net.LeaderFollowerWorkerThread.runIt(LeaderFollowerWorkerThread.java:80) org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:684) java.lang.Thread.run(Thread.java:595) Name: http-129.217.205.43-8080-Processor3 State: TIMED_WAITING on [EMAIL PROTECTED] Total blocked: 44 Total waited: 80 Stack trace: java.lang.Object.wait(Native Method) org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241) org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165) org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalReadLock(AbstractWebdavMethod.java:838) org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:360) org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155) javax.servlet.http.HttpServlet.service(HttpServlet.java:802) org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252) org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) org.apache.slide.webdav.filter.LogFilter.doFilter(LogFilter.java:141) org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:202) org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213) org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:178) org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:524) org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:126) org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:105) org.apache.catalina.authenticator.SingleSignOn.invoke(SingleSignOn.java:419) org.apa
Deadlocks in the Filestore of Silde 2.1
Hi, we are using slide in our e-learning platform http://ews2.uni-dortmund.de in a production environment. We get into deep trouble with deadlocks in Slide! I checked out Silde 2.2pre1. In the RELEASE-NOTES-2.2 there is the sentence: "- New option for a fine grained, deadlock-less locking scheme" My question is: - Should I change to 2.2pre1 for my productive environment? How can I switch back if we discover other problems with this pre-release? - Is there a patch for version 2.1. that fixes these deadlocks? Please help. Thanks in advance Arne v.Irmer Our configuration: Slide 2.1 with fixed hasmap "suspendedLocks" in org.apache.slide.store.ExtendedStore. using a filestore in Tomcat 5.5 Here are some stacktraces from the jconsole in the case of the deadlocks: Name: http-129.217.205.43-8080-Processor3 State: TIMED_WAITING on [EMAIL PROTECTED] Total blocked: 273 Total waited: 332 Stack trace: java.lang.Object.wait(Native Method) org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241) org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165) org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalWriteLock(AbstractWebdavMethod.java:847) org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:362) org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155) javax.servlet.http.HttpServlet.service(HttpServlet.java:802) org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252) org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) org.apache.slide.webdav.filter.LogFilter.doFilter(LogFilter.java:141) org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:202) org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213) org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:178) org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:524) org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:126) org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:105) org.apache.catalina.authenticator.SingleSignOn.invoke(SingleSignOn.java:419) org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:107) org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:148) org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:799) org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.processConnection(Http11Protocol.java:705) org.apache.tomcat.util.net.PoolTcpEndpoint.processSocket(PoolTcpEndpoint.java:527) org.apache.tomcat.util.net.LeaderFollowerWorkerThread.runIt(LeaderFollowerWorkerThread.java:80) org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:684) java.lang.Thread.run(Thread.java:595) Name: http-129.217.205.43-8080-Processor3 State: TIMED_WAITING on [EMAIL PROTECTED] Total blocked: 44 Total waited: 80 Stack trace: java.lang.Object.wait(Native Method) org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:241) org.apache.commons.transaction.locking.GenericLock.acquire(GenericLock.java:165) org.apache.slide.webdav.method.AbstractWebdavMethod.assureGlobalReadLock(AbstractWebdavMethod.java:838) org.apache.slide.webdav.method.AbstractWebdavMethod.run(AbstractWebdavMethod.java:360) org.apache.slide.webdav.WebdavServlet.service(WebdavServlet.java:155) javax.servlet.http.HttpServlet.service(HttpServlet.java:802) org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252) org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) org.apache.slide.webdav.filter.LogFilter.doFilter(LogFilter.java:141) org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:202) org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173) org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213) org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:178) org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:524) org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:126) org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:105) org.apache.catalina.authenticator.SingleSignOn.invoke(SingleSignOn.java:419) org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:107) org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:148) org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:799) org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.processConnection(Http11Protocol.java:705) org.apache.tomcat.util.net.PoolTcpEndpoint.processSocket(PoolTcp
