Thanks Scott,
I have managed to fix the problem ( getting different counts on keys) by
resetting n=3 cluster just to me node and restoring from there afterwards.
Thomas
On 23 Nov 2011, at 01:14, Scott Lystig Fritchie <fritc...@snookles.com> wrote:
> Sorry about the late reply.
>
>>>>>> "tb" == thomas bikeev <thomas.bik...@mac.com> writes:
>
> tb> Oops, actually I do get the following in the monitor: Does it mean
> tb> that loose consistency because of low mem / resources?
>
> tb> 14:21:00.395 [info] monitor long_gc <0.13156.5>
> [{initial_call,{riak_core_vnode_worker_pool,init,1}},{almost_current_function,{gen_fsm,loop,7}}]
> [{timeout,57},{old_heap_block_size,0},{heap_block_size,610},{mbuf_size,
> 0},{stack_size,10},{old_heap_size,0},{heap_size,184}]
>
> No, those messages simply mean that the process had a single garbage
> collection event that took 57 milliseconds. If you're doing a lot of
> work on a lot of data, then the default settings in the etc/app.config
> file may be too low. Edit the riak_sysmon section of that file and
> revise the ceilings upward appropriately, particularly the gs_ms_limit
> item, since that's the one that you might be seeing excessive
> notifications. You may also want to change the heap_word_limit item,
> also.
>
> -Scott
_______________________________________________
riak-users mailing list
riak-users@lists.basho.com
http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
