zookeeper-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From rammohan ganapavarapu <rammohanga...@gmail.com>
Subject Re: local call latency
Date Thu, 26 Oct 2017 01:17:19 GMT
Abe,

Under /node1 i have 30 child nodes but each child node doesn't have any
more nodes. I did took jstack at the time of latency, please find the
attached. It is happening with most of the servers i didn't check all of
them.

Thanks,
Ram

On Wed, Oct 25, 2017 at 4:06 PM, Abraham Fine <afine@apache.org> wrote:

> That's strange. ZooKeeper's reads happen locally usually quickly.
>
> Can you tell us anything else about the issue? Could you possibly
> provide a jstack output from a server that is slow to respond?  Does
> this happen with all of the servers? Does it matter if the node you are
> reading from has many children?
>
> Thanks,
> Abe
>
> On Wed, Oct 25, 2017, at 09:18, rammohan ganapavarapu wrote:
> > Hi,
> >
> > We have a multi region zk cluster, we are observing latency for just
> > local
> > zk 'ls' command intermittently nothing in logs dont know how to find
> >  reason for latency, any one have any thoughts?
> >
> > zkCli.sh ls /node1
> >
> > real 0m13.089s --When latency
> > real 0m0.689s -- No latency
> >
> > Thanks,
> > Ram
>

Mime
  • Unnamed multipart/mixed (inline, None, 0 bytes)
View raw message