hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Andrew Purtell <apurt...@apache.org>
Subject Re: JVM seg fault in HBase region server
Date Thu, 02 May 2013 20:49:15 GMT
In that blog post BenoƮt does a fair amount of showing off to end up with
"So this segfault remains a mystery so far" :-)

My take is don't suspect the JVM, suspect your libc. I am not a fan of
using Ubuntu on the server side. Perhaps you can try a distribution based
on a different kernel+glibc? (RHEL comes to mind.) I realize that could
be... inconvenient.


On Thu, May 2, 2013 at 1:39 PM, Varun Sharma <varun@pinterest.com> wrote:

> I don't have one unfortunately - We did not have the -XX:ErrorLog turned on
> :(
>
> But I did some digging following what Benoit wrote in his Blog. Basically
> the segfault happens in the same place inside a clearerr() function in
> glibc which accepts a file descriptor.
>
> The JVM version is sun java 1.6.0u38. Is this possibly too old ?
>
> Thanks
> Varun
>
>
> On Thu, May 2, 2013 at 12:08 PM, Andrew Purtell <apurtell@apache.org>
> wrote:
>
> > Can you pastebin or post somewhere the entire hs_err* file?
> >
> >
> > On Wed, May 1, 2013 at 1:54 PM, Varun Sharma <varun@pinterest.com>
> wrote:
> >
> > > Hi,
> > >
> > > I am seeing the following which is a JVM segfault:
> > >
> > > hbase-regionser[28734]: segfault at 8 ip 00007f269bcc307e sp
> > > 00007fff50f7e638 error 4 in libc-2.15.so[7f269bc51000+1b5000]
> > >
> > > Benoit Tsuna reported a similar issue a while back -
> > >
> > >
> >
> http://blog.tsunanet.net/2011/05/jvm-u24-segfault-in-clearerr-on-jaunty.html
> > >
> > > I wonder if there was ever a resolution on the issue. We are running
> java
> > > version "1.6.0_38"
> > > , ubuntu precise 12.04 - kernel 3.2. The libc version from the above
> line
> > > seems to be 2.15.
> > >
> > > I wonder if people have seen this in the past and if yes, were they
> able
> > to
> > > resolve it and if yes, then how ? Was it a VM upgrade, a kernel upgrade
> > or
> > > a libc upgrade ?
> > >
> > > Thanks
> > > Varun
> > >
> >
> >
> >
> > --
> > Best regards,
> >
> >    - Andy
> >
> > Problems worthy of attack prove their worth by hitting back. - Piet Hein
> > (via Tom White)
> >
>



-- 
Best regards,

   - Andy

Problems worthy of attack prove their worth by hitting back. - Piet Hein
(via Tom White)

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