Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hello
We have a problem with a job that failed because of java.lang.OutOfMemoryError: GC overhead limit exceeded.
the job is reading 13 million records from oracle to thashoutput . I would like to know what cause this error .
I know that there is xmx and xms parameter but I dont realy know how should I config those parameter.
is there any legality for changing those parameters.
Thanks
@Boof1977 wrote:
what is the relation between xmx and xms?
You ask me - what relations between minimum value and maximum value? 😉
second - is bigger 🙂
Sorry for humour, but - really what are You try to ask?
https://stackoverflow.com/questions/14763079/what-are-the-xms-and-xmx-parameters-when-starting-jvms
https://docs.oracle.com/cd/E15523_01/web.1111/e13814/jvm_tuning.htm#PERFM150
does the length of the field at the "edit schema" represent the number of bytes?
do not know - may be, not sure. in any case - minimum size already calculated on previous answers ... and this is only for Hash
I try to explain You simple idea - something could be wrong in Job design, may be in idea of Job
Think different
regarding the humour I ment ,does the maximum value is need to be twice bigger or maybe fourth bigger from the min value ? is there any rule of how much bigger should it be?
do you have any useful link I can see the size of the types you wrote (number = 20 bytes )
you wrote "... and this is only for Hash" - can you explain?
Thnks
@Boof1977 wrote:
do you have any useful link I can see the size of the types you wrote (number = 20 bytes )
Google and Oracle Docs 😉
https://stackoverflow.com/questions/24240087/oracle-numberp-storage-size
https://asktom.oracle.com/pls/asktom/f?p=100:11:0:::11_QUESTION_ID:1856720300346322149
https://docs.oracle.com/cd/B28359_01/server.111/b28318/datatype.htm#CNCPT1824
https://docs.oracle.com/cd/B19306_01/server.102/b14237/limits001.htm#i287903
You not provide full information, so all about 20 - is just guessing
is it NUMBER(38) or NUMBER(2) - ?!!
tHASHInput - must store all data in memory for feature use
Dose talend also must have some memory to manage this data?
for example You have flow with another 100M rows and want split it to Matched and rejected ... dose Talend need any memory for make this?
Dose Your Job have only 1 component - tHASHInput and nothing more? Dose Talend need memory for this components as well? Because memory used for Hash - not released until Your Job not finished
You just describe the error message ... and nothing more
Thank for your help.