In particular that it works only for 'drbd' and 'plain', and that it doesn't perform hardware failure detection, which are both common user expectations.
Signed-off-by: Petr Pudlak <[email protected]> --- man/harep.rst | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/man/harep.rst b/man/harep.rst index 5b38035..86f353b 100644 --- a/man/harep.rst +++ b/man/harep.rst @@ -65,6 +65,18 @@ the cluster has both ``fix-storage`` and ``reinstall``. The I1 instance will be allowed to ``failover``, the I2 instance only to ``fix-storage``. +LIMITATIONS +----------- + +Harep doesn't do any hardware failure detection on its own, it relies on +nodes being marked as offline by the administrator. + +Also harep currently works only for instances with the ``drbd`` and +``plain`` disk templates. + +Both these issues will be addressed by a new maintenance daemon in +future Ganeti versions, which will supersede harep. + OPTIONS ------- -- 2.6.0.rc2.230.g3dd15c0
