On Wed, Jun 19, 2013 at 1:27 PM, Ben Pfaff <span dir="ltr"><<a href="mailto:blp@nicira.com" target="_blank">blp@nicira.com</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">On Wed, Jun 19, 2013 at 02:03:15AM -0700, Gurucharan Shetty wrote:<br>
> Specifically for the case, where we know that a bridge record<br>
> should exist in xapi, if we don't get any bridge records, log<br>
> the failure and retry again later.<br>
><br>
> Signed-off-by: Gurucharan Shetty <<a href="mailto:gshetty@nicira.com">gshetty@nicira.com</a>><br>
<br>
</div>Could we add a helper function for the common code here?<br></blockquote><div>I decided to abandon this patch because this makes an assumption that all the bridges that are created in xenserver are through standard xapi methods. If anyone creates an asynchronous bridge through command line (ex: using ovs-vsctl), ovs-xapi-sync will think that xapi is down because there are no records for the new bridge.</div>
<div><br></div><div>If xapi is completely down, ovs-xapi-sync will crash and restart. During the restart, it will wait for xapi to start. So we are protected against that. For the hypothetical scenario of xapi giving wrong information, there is only so much you can do.</div>
<div> </div><div>Thanks,</div><div>Guru</div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
Thanks,<br>
<br>
Ben.<br>
</blockquote></div><br>