That happens every time there is a getwork failure which means that the pool failed to give out new work in time. That's one of the things h2o is working on to fix. There is no way to completely prevent it as it's on the server side but you could probably lower the effect of it by load-balancing between pools. Also, judging from h2o's tweets, the us servers might have less of getwork failures.
ok thanks for then answer.
So you mean its better to use the US server? Is there a disadvantage since im in europe? More rejects or something like this?
And what do you mean by load balancing between pools?
The US servers might have less getwork failures but you also might get more rejects because of the higher latency. So in the end it might not make a difference. But only way to be sure is for you to test it.
What I mean is to use the load-balance option in cgminer and set it to balance the load between the eu/us servers. I don't know if it will actually help but it's been discussed here a few times. Search back some pages and you'll find it.
? 
