<div dir="ltr"><br clear="all"><div>Hello</div><div><br></div><div>After 2 commit below</div><div>585554e drbd: resize: reduce own sanity checks, trust peer</div><div>c09ce4b drbd: resize loop avoidance band-aid</div><div><br></div><div>It can't reconnect just in case different volume's size.<br></div><div>That was only occurred over the 3 nodes configuration.</div><div><br></div><div>The scenario is below</div><div><br></div><div>A(100mb) - B(100mb) - C(90mb)</div><div>1. Success to connect at Inconsistent state</div><div>2. C node does "Primary --force "</div><div>3. All nodes agree 90mb size</div><div>4. C node does down</div><div>5. A node does down</div><div>6. A node does up</div><div>7. A and b agree 100mb size</div><div>8. C node can't connect because of lower size</div><div><br></div><div><br></div><div><br></div><div><div>I was looking for the code,and I think it was recalculated forcibly because below</div><div><br></div><div>in receive_sizes()</div><div><div><span class="" style="white-space:pre"> </span>/* Maybe the peer knows something about peers I cannot currently see. */</div><div><span class="" style="white-space:pre"> </span>if (is_handshake)</div><div><span class="" style="white-space:pre"> </span>ddsf |= DDSF_FORCED;</div></div></div><div><br></div><div><br></div><div>Thank you,</div><div><br></div>
</div>