<br><br><div class="gmail_quote">On Mon, May 9, 2011 at 1:00 PM, Emmanuel Dreyfus <span dir="ltr"><<a href="mailto:manu@netbsd.org">manu@netbsd.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
Hello<br>
<br>
I have been playing with replicate translator on glusterfs 3.1.4,<br>
with high availlability in mind. It works fine, if I kill one server,<br>
the client uses the other one, and vice versa. auto healing works<br>
fine, this is amazing.<br>
<br>
However things turn bad when a server die while a client holds a lock<br>
on a file. Following such an event, the file cannot be written anymore<br>
even after all servers came back online. I have to delete it to clear<br>
the situation.<br>
<br>
Is it a feature missing in the replication protocol? Or is it just<br>
missing in the implementation? Or is it a bug?<br></blockquote><div><br></div><div> </div><div>Can you describe more about the test case with the exact sequence of steps? There is code to self-heal locks after reconnection. If you can give us a test case it will be wonderful.</div>
<div><br></div><div>Thanks,</div><div>Avati</div><div><br></div></div>