TAJO-1896: Fix the travis test failure.

[jinossy]

No description provided.

[jihoonson]

Thanks for your great work. This is the really necessary and urgent issue to be resolved as soon as possible.
Your patch looks good to me overall. I have only one question. Would you share your intention why you removed the restriction on MaxMetaspaceSize?

[jinossy]

Garbage collection of the dead classes and classloaders is triggered once the class metadata usage reaches the “-XX:MetaspaceSize”
So we don't need limitation.

https://blogs.oracle.com/poonam/entry/about_g1_garbage_collector_permanent
https://dzone.com/articles/java-8-permgen-metaspace

[jihoonson]

Thanks for your reply. I also found the following phrase from the link.

By default class metadata allocation is limited by the amount of available native memory.

So, we can expect that our test can use every available native memory with GC. If our test is failed due to lack of available memory, it means that our test exhaust too much memory, I think.
Here is my +1. Appreciate for your work.