Re: how to handle dataregion out of memmory gracefully

2018-11-02 Thread Denis Magda
Enable Ignite persistence [1], an eviction algorithm [2] or swap space [3] to avoid OOM in case of the data region overflow. [1] https://apacheignite.readme.io/docs/distributed-persistent-store [2] https://apacheignite.readme.io/docs/evictions [3] https://apacheignite.readme.io/docs/swap-space

Re: how to handle dataregion out of memmory gracefully

2018-10-31 Thread wt
Thanks for this but what concerns me is if i setup several data regions and one of them happens to be a play area i guess for data scientists and they hit the memory limit of that region then the whole server goes down. in the case for a cluster all the nodes hosting that region will go down. I

Re: how to handle dataregion out of memmory gracefully

2018-10-31 Thread Павлухин Иван
Hi Wayne, You can see a message written to a console during Ignite startup with a calculated amount of memory required for a server. It looks as follows: Nodes started on local machine require more than 80% of physical RAM what can lead to significant slowdown due to swapping (please decrease

Re: how to handle dataregion out of memmory gracefully

2018-10-29 Thread Ilya Kasnacheev
Hello! I'm afraid that Ignite is not usable currently after suffering Out Of Memory error. You should be careful to prevent that from happening. Currently there is no graceful way of dealing with it. Regards, -- Ilya Kasnacheev вс, 28 окт. 2018 г. в 11:58, wt : > in testing i managed to

how to handle dataregion out of memmory gracefully

2018-10-28 Thread wt
in testing i managed to exceed a data regions space and this region is memory only. When this happens an unhandled exception is thrown from the underlying ignite dlls and the process crashes. How can i handle this gracefully without losing the server? -- Sent from: