@Araron you're right and i was wrong! 2011/10/20 Yang <teddyyyy...@gmail.com>
> found it , https://issues.apache.org/jira/browse/CASSANDRA-3387 > > On Thu, Oct 20, 2011 at 1:37 PM, aaron morton <aa...@thelastpickle.com> > wrote: > > It's unlikely that HH is the issue. (Disclaimer, am not familiar with HH > in > > 1.0, i know it's changes a bit) > > Take a look at the TP Stats, what's happening ? > > Cheers > > ----------------- > > Aaron Morton > > Freelance Developer > > @aaronmorton > > http://www.thelastpickle.com > > On 20/10/2011, at 10:10 AM, Jérémy SEVELLEC wrote: > > > > Ok. > > I think a degration could be normal because your cluster is in a degraded > > state when a node is down. > > With a replication_factor of 3 and with a 3 nodes cluster, each data you > > write is replicated on each node. As One node is down, when writing, it's > > impossible to send a replica on the down node and hints are send > > : > http://www.datastax.com/docs/1.0/dml/about_writes#hinted-handoff-writes > > And it could be more expensive to achieve QUORUM when you read in that > > context. > > It may be one explanation. You can turn cassandra log into debug level to > > see what happen when when there is a down node. > > 2011/10/19 Yang <teddyyyy...@gmail.com> > >> > >> 3 > >> > >> sorry forgot this important info > >> > >> On Oct 19, 2011 11:31 AM, "Jérémy SEVELLEC" <jsevel...@gmail.com> > wrote: > >>> > >>> Hi, what is your replication_factor? > >>> > >>> 2011/10/19 Yang <teddyyyy...@gmail.com> > >>>> > >>>> I'm using a cassandra version compiled from 1.0.0 github HEAD. > >>>> > >>>> I have 3 nodes, A B and C, on node A I run a client, which talks only > >>>> to B as the coordinator. > >>>> > >>>> the performance is pretty good, a QUORUM read+write takes < 10ms. > >>>> > >>>> but then I shutdown C, quickly the performance starts to degrade, and > >>>> QUORUM read+write time steadily increase to about 300ms. > >>>> > >>>> if I shutdown A and keep C, I observe the same effect. > >>>> > >>>> > >>>> I understand that "2 out of 3" is going to give you faster response > >>>> than "2 out of 2", but the difference should not be that dramatic as > >>>> 10ms vs 300ms. > >>>> > >>>> any possible reasons for this?(or how to debug this?) > >>>> > >>>> Thanks > >>>> Yang > >>> > >>> > >>> > >>> -- > >>> Jérémy > > > > > > > > -- > > Jérémy > > > > > -- Jérémy