Tomas Hurka

Thanks for data, it shows that the `0x1c` segment is not finished - the size of segment is written to the disk once the whole segment is done. Based on...

> I did not hear the customer report a disk space issue around this time. So I am not sure if that's the case. > I can have them verify...

> Also, does the `0x1c` segment have its data at the end of the hprof file to make this a truncated one? Yes, and after `0x1c` there should be `0x2c`...

> As you requested, here's the last 256 bytes This data shows that the heap dump file ends abruptly in the middle of instance object dump.

NPE is fixed by GH-368. BTW: You are running VisualVM 2.1, not 2.1.1 as mentioned above.

You need to build it with JDK 8. See [README.md.](https://github.com/oracle/visualvm)

Thanks for the report. It would be great if you can come up with the steps how to reproduce this problem. There are some strange things in the log file...

You can also write results directly to file. See [this](https://stackoverflow.com/questions/43376493/how-to-save-the-visualvm-oql-result-directly-into-file) answer for more information.

Can you please provide the CPU snapshot or steps to reproduce?

Above snapshot: [GH-346-snapshot_console_logging.nps.zip](https://github.com/oracle/visualvm/files/7210856/GH-346-snapshot_console_logging.nps.zip)