[ovs-dev] [PATCH v2] ovs-lib: Keep internal interface ip during upgrade.

[Daniele Di Proietto]

2016-06-22 9:52 GMT-07:00 Darrell Ball <dlu998 at gmail.com>:

> On Tue, Jun 21, 2016 at 7:27 PM, Daniele Di Proietto <
> diproiettod at vmware.com
> > wrote:
>
> > Commit 9b5422a98f81("ovs-lib: Try to call exit before killing.")
> > introduced a problem where internal interfaces are destroyed and
> > recreated, losing their IP address.
> >
> > Commit 9aad5a5a96ba("ovs-vswitchd: Preserve datapath ports across
> > graceful shutdown.") fixed the problem by changing ovs-vswitchd
> > to preserve the ports on `ovs-appctl exit`.  Unfortunately, this fix is
> > not enough during upgrade from <= 2.5.0, where an old ovs-vswitchd is
> > running (without the fix) and a new ovs-lib script is performing the
> > restart.
> >
> > The problem seem to affect both RHEL and ubuntu.
> >
> > This commit fixes the upgrade by looking at the running daemon
> > version and avoid using `ovs-appctl exit` if it's < 2.5.90.
> >
> > Suggested-by: Gurucharan Shetty <guru at ovn.org>
> > Signed-off-by: Daniele Di Proietto <diproiettod at vmware.com>
>
>
> 1) Is it normal in this code base to embed specific version numbers in a
> generic library file ?
>

v1 of this patch had the check in ovs-ctl, but we thought that other
daemons might be affected (every daemon used to be killed during restart),
so we moved it to ovs-lib


> 2) If coming from < 2.5.90 then the problem that Commit 9b5422a98f81 was
> trying to fix
>     will exist ?
>

On < 2.5.90 the problem already existed on restart and with this patch it
will show once more during upgrade.


>
>    In general, do you need to document this somewhere at user level (
> install.md or somewhere else) ?
>

IMHO this is a problem of the system scripts, and with this commit the
system scripts are fixed, so that the user doesn't need to worry about this.

Thanks,

Daniele