This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
This result is reported by customer, the application was running for more than 23 hours: >Our application was run with setting -Xmx192m. It is my assumption that if >we sample 1 in 20 allocations, that we should multiply the number of >bytes/instances shown in the snapshot by 20 to get an estimate of the actual >instances in memory. If this is true, java.util.HashMap$Entry is showing >almost 10MB, which would indicate approximately 200MB for this data type >alone, which is clearly wrong because there are also other data types in >memory, and this exceeds the heap size limit of 192 MB. > >Other things to note: > >- the total allocated objects for java.util.HashMap$Entry is -2,139,728,559. >- when showing allocation stack traces for java.util.HashMap$Entry, note >that the "quotient" (live bytes / live objects) is different, depending on >which line you examine.
Created attachment 54193 [details] screenshot
This does not look like P2 to me. Total Alloc. Obj column is not critical information. Moreover it is clear that there was a huge number of java.util.HashMap$Entry allocated anyway the exact allocated size is not that important. Next time please try to file ordinary bug report. The description in this report is more suited for feedback mailing list.
changing milestone