cloudstack-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Laszlo Hornyak <laszlo.horn...@gmail.com>
Subject Re: Is it nessesary to test getBase64Keystore 100000 times in ConfigurationServerImplTest.java
Date Tue, 04 Jun 2013 14:14:12 GMT
Thanks and sorry for the inconvenience!


On Tue, Jun 4, 2013 at 1:32 PM, Prasanna Santhanam <tsp@apache.org> wrote:

> On Tue, Jun 04, 2013 at 01:23:10PM +0200, Laszlo Hornyak wrote:
> > Hi,
> >
> > Not anymore, but that was where the code broke because of a resource
> leak.
> > Please just remove that test!
> >
>
> Yup - removed y'day with 22bda5c3c960e1e6897b3b0aaacdf91add97976d
>
> >
> > On Mon, Jun 3, 2013 at 8:19 AM, Wei ZHOU <ustcweizhou@gmail.com> wrote:
> >
> > > The build of latest master branch spent much more time on
> > > ConfigurationServerImplTest.java, which includes test getBase64Keystore
> > > 100000 times.
> > >
> > > Is it neccesary?
> > >
> > > [root@weizhou-centos incubator-cloudstack]# git diff
> > > 2f29185943ac0412aa501b59493837b4055642e0
> > > 4894187991d581b72807b4282b7a29a48a8031e5
> > > +    @Test
> > > +    public void testGetBase64KeystoreZillionTimes() throws
> IOException {
> > > +        File temp = File.createTempFile("keystore", "");
> > > +        try {
> > > +            // may cause IOException with the original implementation
> > > because of too many open files
> > > +            for (int i = 0; i < 100000; i++) {
> > > +                FileUtils.writeStringToFile(temp,
> > > Base64.encodeBase64String(TEST.getBytes()));
> > > +                final String keystore =
> > > ConfigurationServerImpl.getBase64Keystore(temp.getPath());
> > > +                // let's decode it to make sure it makes sense
> > > +                Base64.decodeBase64(keystore);
> > > +            }
> > > +        } finally {
> > > +            temp.delete();
> > > +        }
> > > +    }
> > >
> >
> >
> >
> > --
> >
> > EOF
>
> --
> Prasanna.,
>
> ------------------------
> Powered by BigRock.com
>
>


-- 

EOF

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