HI Damon, Thanks for the response. This happens only for a couple of documents and not all.I will definetely reach for Marklogic support but meanwhile I was trying to reach out to the group to see if anyone could help me with addressing the problem of fixing these two documents(either by deleting them or changing the cpf properties) I want to avoid restarting production sever to fix couple of documents , also when I am not sure if restarting would help in this scenario.
For the long term fix for this issue,I will certainly follow the advice below and check the performance. Also the insertion here is not done in bulk, its only one document at a given time and the document is very small. I did run the query in cq , profiled it and everything looks fine. Thanks, Sujith. On Fri, Jun 15, 2012 at 7:41 AM, Damon Feldman <[email protected]>wrote: > Sujith, > > It sounds like you should contact support if you think the CPF state is > not advancing properly. > > But in addition to possible CPF issues it sounds like you have a > performance problem resulting in time exceeded. The first things I check if > a module execution or query is too slow is what the server is doing while > the insert is happening. The "top" command (Linux) will tell you if the job > is CPU bound, and on which server if you have a cluster; particularly look > for kswapd processes which indicate swapping/thrashing. iostat can also > help. > > Also run the query (or portions of the query) in cq (queryConsole in > MarkLogic 5) and check output of xdmp:query-meters and run in "profile" > mode. > > The Query Performance and Tuning Guide may also help: > http://docs.marklogic.com/5.0doc/docapp.xqy#display.xqy?fname=http%3a//pubs/5.0doc/xml/performance/title.xml > > Yours, > Damon > ------------------------------ > *From:* [email protected] [ > [email protected]] On Behalf Of Sujith [ > [email protected]] > *Sent:* Thursday, June 14, 2012 11:11 PM > *To:* MarkLogic Developer Discussion > *Subject:* [MarkLogic Dev General] Timeout Error in CPF - XDMP:EXTIME > > I have a situation where Timeout exception(XDMP:EXTIME) occurs in CPF > and document neither enters the error state, nor advances to the next state > and the status of the cpf remains in "active". Reinsertion of the document > fails with a 500 internal server error. > > I tried to change the state and status of the cpf properties but was > unable to do so, neither I am able to delete the document. Any action that > I try to perform on this document gives me back XDMP:EXTIME.can anyone > comment on why would this happen. > > Also will restarting the Marklogic server change the state to initial? I > believe CPF will restart from where it left off on restarting the > server.This happens only in Production . I am not able to replicate the > same issue in Dev or QA/Test. > > Can anyone comment on why would this happen and is their is a way I could > Modify the CPF Status or delete the document. Any help on this is > appreciated. > > I did go through this post and tried few options, but of no help. > http://community.marklogic.com/pipermail/general/2010-March/005011.html > > Here are my server Details > > ML Version : 4.2.7 > OS : RHES 5.6 > > -- > Thanks & Regards > SujithMaram > [email protected] > (469) 995-5907 > > > > _______________________________________________ > General mailing list > [email protected] > http://community.marklogic.com/mailman/listinfo/general > > -- Thanks & Regards SujithMaram [email protected] (469) 995-5907
_______________________________________________ General mailing list [email protected] http://community.marklogic.com/mailman/listinfo/general
