Please check your -Xmx switch.
Memory will not be released because that is not how Java works. It
allocates the memory it needs and periodically garbage collects within
that. You have given it too much memory and you should not expect Java to
release it ever. The solution is to give it less. A rule of thumb is to
leave 10gb free for system usage and divide the remainder among your Java
processes.
Thanks,
Karl
On Fri, Oct 2, 2020 at 11:21 AM Bisonti Mario
wrote:
> Yes, buti t seems that, when the indexing finished, the memory is not
> released
>
>
>
>
>
> *Da:* Karl Wright
> *Inviato:* venerdì 2 ottobre 2020 17:14
> *A:* user@manifoldcf.apache.org
> *Oggetto:* Re: Memory problem on Agent ?
>
>
>
> Hi Mario,
>
> Java processes only use the memory you hand them.
>
>
> It looks like you are handing Java more memory than your machine has.
>
> This will not work.
>
>
> Karl
>
>
>
>
>
> On Fri, Oct 2, 2020 at 10:45 AM Bisonti Mario
> wrote:
>
>
>
> Hallo.
>
>
>
> When I scan the content of Repository , I note that memory used is very
> high and it isn’t released
>
>
>
> i.e. 60GB on 70GB available
>
>
>
> I tried to free shutting down tjhe agent but I am not able:
>
>
>
> /opt/manifoldcf/multiprocess-zk-example-proprietary/stop-agents.sh
>
> OpenJDK 64-Bit Server VM warning: INFO:
> os::commit_memory(0x7f4d5800, 34359738368, 0) failed; error='Not
> enough space' (errno=12)
>
> #
>
> # There is insufficient memory for the Java Runtime Environment to
> continue.
>
> # Native memory allocation (mmap) failed to map 34359738368 bytes for
> committing reserved memory.
>
> # An error report file with more information is saved as:
>
> # /opt/manifoldcf/multiprocess-zk-example-proprietary/hs_err_pid2796.log
>
>
>
> So, to free memory, I have to restart the server
>
> How could I solve this?
>
>
>
> Thanks a lot
>
> Mario
>
>
>
>
