Re: [PATCH v6 2/4] tests/test-replication.c: Add test for for secondary node continuing replication

[Lukas Straub <lukasstraub2@web.de>]

On Wed, 9 Oct 2019 06:03:03 +0000
"Zhang, Chen" <chen.zhang@intel.com> wrote:

> > -----Original Message-----
> > From: Lukas Straub <lukasstraub2@web.de>
> > Sent: Saturday, October 5, 2019 9:06 PM
> > To: qemu-devel <qemu-devel@nongnu.org>
> > Cc: Zhang, Chen <chen.zhang@intel.com>; Jason Wang
> > <jasowang@redhat.com>; Wen Congyang <wencongyang2@huawei.com>;
> > Xie Changlong <xiechanglong.d@gmail.com>; Kevin Wolf
> > <kwolf@redhat.com>; Max Reitz <mreitz@redhat.com>; qemu-block
> > <qemu-block@nongnu.org>
> > Subject: [PATCH v6 2/4] tests/test-replication.c: Add test for for secondary
> > node continuing replication
> >
> > This simulates the case that happens when we resume COLO after failover.
> >
> > Signed-off-by: Lukas Straub <lukasstraub2@web.de>
> > ---
> >  tests/test-replication.c | 52
> > ++++++++++++++++++++++++++++++++++++++++
> >  1 file changed, 52 insertions(+)
> >
> > diff --git a/tests/test-replication.c b/tests/test-replication.c index
> > f085d1993a..8e18ecd998 100644
> > --- a/tests/test-replication.c
> > +++ b/tests/test-replication.c
> > @@ -489,6 +489,56 @@ static void test_secondary_stop(void)
> >      teardown_secondary();
> >  }
> >
> > +static void test_secondary_continuous_replication(void)
> > +{
> > +    BlockBackend *top_blk, *local_blk;
> > +    Error *local_err = NULL;
> > +
> > +    top_blk = start_secondary();
> > +    replication_start_all(REPLICATION_MODE_SECONDARY, &local_err);
> > +    g_assert(!local_err);
> > +
> > +    /* write 0x22 to s_local_disk (IMG_SIZE / 2, IMG_SIZE) */
> > +    local_blk = blk_by_name(S_LOCAL_DISK_ID);
> > +    test_blk_write(local_blk, 0x22, IMG_SIZE / 2, IMG_SIZE / 2, false);
> > +
> > +    /* replication will backup s_local_disk to s_hidden_disk */
> > +    test_blk_read(top_blk, 0x11, IMG_SIZE / 2,
> > +                  IMG_SIZE / 2, 0, IMG_SIZE, false);
> > +
> > +    /* write 0x33 to s_active_disk (0, IMG_SIZE / 2) */
> > +    test_blk_write(top_blk, 0x33, 0, IMG_SIZE / 2, false);
> > +
> > +    /* do failover (active commit) */
> > +    replication_stop_all(true, &local_err);
> > +    g_assert(!local_err);
> > +
> > +    /* it should ignore all requests from now on */
>
> Should we need add teardown_secondary() here?

I don't think so. It is used to remove the block node after each test and I'm doing that below.

Regards,
Lukas Straub

> Thanks
> Zhang Chen
>
> > +
> > +    /* start after failover */
> > +    replication_start_all(REPLICATION_MODE_PRIMARY, &local_err);
> > +    g_assert(!local_err);
> > +
> > +    /* checkpoint */
> > +    replication_do_checkpoint_all(&local_err);
> > +    g_assert(!local_err);
> > +
> > +    /* stop */
> > +    replication_stop_all(true, &local_err);
> > +    g_assert(!local_err);
> > +
> > +    /* read from s_local_disk (0, IMG_SIZE / 2) */
> > +    test_blk_read(top_blk, 0x33, 0, IMG_SIZE / 2,
> > +                  0, IMG_SIZE / 2, false);
> > +
> > +
> > +    /* read from s_local_disk (IMG_SIZE / 2, IMG_SIZE) */
> > +    test_blk_read(top_blk, 0x22, IMG_SIZE / 2,
> > +                  IMG_SIZE / 2, 0, IMG_SIZE, false);
> > +
> > +    teardown_secondary();
> > +}
> > +
> >  static void test_secondary_do_checkpoint(void)
> >  {
> >      BlockBackend *top_blk, *local_blk;
> > @@ -584,6 +634,8 @@ int main(int argc, char **argv)
> >      g_test_add_func("/replication/secondary/write", test_secondary_write);
> >      g_test_add_func("/replication/secondary/start", test_secondary_start);
> >      g_test_add_func("/replication/secondary/stop",  test_secondary_stop);
> > +    g_test_add_func("/replication/secondary/continuous_replication",
> > +                    test_secondary_continuous_replication);
> >      g_test_add_func("/replication/secondary/do_checkpoint",
> >                      test_secondary_do_checkpoint);
> >      g_test_add_func("/replication/secondary/get_error_all",
> > --
> > 2.20.1
>