<div dir="ltr">But I think using ifdown isn't the correct way to test the cluster, this topic was discussed many times<br></div><div class="gmail_extra"><br><div class="gmail_quote">2018-04-26 9:53 GMT+02:00 范国腾 <span dir="ltr"><<a href="mailto:fanguoteng@highgo.com" target="_blank">fanguoteng@highgo.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div link="#0563C1" vlink="#954F72" lang="ZH-CN">
<div class="m_-5680550059889405999WordSection1">
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">1. There is no failure in initial status. sds1 is master<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US"><img style="width:6.927in;height:5.5in" id="m_-5680550059889405999图片_x0020_2" src="cid:image001.png@01D3DD75.3F4BF110" width="665" height="528"></span><span lang="EN-US"><u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US"><u></u> <u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">2. ifdown the sds1 VIP network card.<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US"><img style="width:6.9583in;height:5.7812in" id="m_-5680550059889405999图片_x0020_3" src="cid:image002.png@01D3DD75.71D5DE70" width="668" height="555"></span><span lang="EN-US"><u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">3. ifup the sds1 VIP network card and then ifdown sds2 VIP network card<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US"><img style="width:4.6562in;height:5.6145in" id="m_-5680550059889405999图片_x0020_4" src="cid:image003.png@01D3DD76.26C5E820" width="447" height="539"></span><span lang="EN-US"><u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US"><u></u> <u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US"><u></u> <u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span class=""><span lang="EN-US">-----</span><span style="font-family:等线">邮件原件</span><span lang="EN-US">-----<br>
</span><span style="font-family:等线">发件人</span><span lang="EN-US">: Jehan-Guillaume de Rorthais [mailto:<a href="mailto:jgdr@dalibo.com" target="_blank">jgdr@dalibo.com</a>]
<br>
</span></span><span style="font-family:等线">发送时间</span><span lang="EN-US">: 2018</span><span style="font-family:等线">年</span><span lang="EN-US">4</span><span style="font-family:等线">月</span><span lang="EN-US">26</span><span style="font-family:等线">日</span><span lang="EN-US">
15:07<br>
</span></p><div><div class="h5"><span style="font-family:等线">收件人</span><span lang="EN-US">: </span><span style="font-family:等线">范国腾</span><span lang="EN-US"> <<a href="mailto:fanguoteng@highgo.com" target="_blank">fanguoteng@highgo.com</a>><br>
</span><span style="font-family:等线">抄送</span><span lang="EN-US">: Cluster Labs - All topics related to open-source clustering welcomed <<a href="mailto:users@clusterlabs.org" target="_blank">users@clusterlabs.org</a>>;
</span><span style="font-family:等线">李梦怡</span><span lang="EN-US"> <<a href="mailto:limengyi@highgo.com" target="_blank">limengyi@highgo.com</a>><br>
</span><span style="font-family:等线">主题</span><span lang="EN-US">: Re: [ClusterLabs] the PAF switchover does not happen if the VIP resource is stopped</span></div></div><p></p><div><div class="h5">
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US"><u></u> <u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">On Thu, 26 Apr 2018 02:53:33 +0000<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span style="font-family:等线">范国腾</span><span lang="EN-US"> <<a href="mailto:fanguoteng@highgo.com" target="_blank"><span style="color:windowtext;text-decoration:none">fanguoteng@highgo.com</span></a>> wrote:<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US"><u></u> <u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> Hi Rorthais</span><span style="font-family:等线">，</span><span lang="EN-US"><u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> Thank you for your help. <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> The replication works at that time. <u></u>
<u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> I try again today.<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> (1) If I run "ifup enp0s3" in node2, then run "ifdown enp0s3" in
<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> node1, the switchover issue could be reproduced. (2) But if I run
<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> "ifup enp0s3" in node2, run "pcs resource cleanup mastergroup" to
<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> clean the VIP resource, and there is no Failed Actions in "pcs
<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> status", then run "ifdown enp0s3" in node1, it works. The switchover could happened again.<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> Is there any parameter to control this behaviors so that I don't need
<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> to execute the "pcs cleanup" command every time?<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US"><u></u> <u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">Check the failcounts for each resource on each nodes (pcs resource failcount [...]).<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">Check the scores as well (crm_simulate -sL).<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US"><u></u> <u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> -----</span><span style="font-family:等线">邮件原件</span><span lang="EN-US">-----<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> </span><span style="font-family:等线">发件人</span><span lang="EN-US">: Jehan-Guillaume de Rorthais [<a href="mailto:jgdr@dalibo.com" target="_blank"><span style="color:windowtext;text-decoration:none">mailto:jgdr@dalibo.com</span></a>]<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> </span><span style="font-family:等线">发送时间</span><span lang="EN-US">: 2018</span><span style="font-family:等线">年</span><span lang="EN-US">4</span><span style="font-family:等线">月</span><span lang="EN-US">25</span><span style="font-family:等线">日</span><span lang="EN-US">
18:39<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> </span><span style="font-family:等线">收件人</span><span lang="EN-US">:
</span><span style="font-family:等线">范国腾</span><span lang="EN-US"> <<a href="mailto:fanguoteng@highgo.com" target="_blank"><span style="color:windowtext;text-decoration:none">fanguoteng@highgo.com</span></a>><u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> </span><span style="font-family:等线">抄送</span><span lang="EN-US">: Cluster Labs - All topics related to open-source clustering
<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> welcomed <<a href="mailto:users@clusterlabs.org" target="_blank"><span style="color:windowtext;text-decoration:none">users@clusterlabs.org</span></a>>;
</span><span style="font-family:等线">李梦怡</span><span lang="EN-US"> <<a href="mailto:limengyi@highgo.com" target="_blank"><span style="color:windowtext;text-decoration:none">limengyi@highgo.com</span></a>>
</span><span style="font-family:等线">主题</span><span lang="EN-US">: Re: <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> [ClusterLabs] the PAF switchover does not happen if the VIP resource
<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> is stopped<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> On Wed, 25 Apr 2018 08:58:34 +0000<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> </span><span style="font-family:等线">范国腾</span><span lang="EN-US"> <<a href="mailto:fanguoteng@highgo.com" target="_blank"><span style="color:windowtext;text-decoration:none">fanguoteng@highgo.com</span></a>> wrote:<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > Our lab has two resource: (1) PAF (master/slave) (2) VIP (bind to the<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > master PAF node). The configuration is in the attachment.<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > Each node has two network card: One(enp0s8) is for the pacemaker
<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > heartbeat in internal network, the other(enp0s3) is for the master
<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > VIP in the external network.<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > We are testing the following case: if the master VIP network card is
<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > down, the master postgres and VIP could switch to another node.<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > 1. At first, node2 is master, I run "ifdown enp0s3" in node2, then<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > node1 become the master, that is ok.<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > 2. Then I run "ifup enp0s3" in node2, wait for 60 seconds,<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> Did you check PostgreSQL instances were replicating again?<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> <u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > then run "ifdown enp0s3" in node1, but the node1 still be master.
<u></u><u></u></span></p>
<p class="m_-5680550059889405999MsoPlainText"><span lang="EN-US">> > Why does switchover doesn't happened? How to recover to make system work?<u></u><u></u></span></p>
</div></div></div>
</div>
<br>______________________________<wbr>_________________<br>
Users mailing list: <a href="mailto:Users@clusterlabs.org">Users@clusterlabs.org</a><br>
<a href="https://lists.clusterlabs.org/mailman/listinfo/users" rel="noreferrer" target="_blank">https://lists.clusterlabs.org/<wbr>mailman/listinfo/users</a><br>
<br>
Project Home: <a href="http://www.clusterlabs.org" rel="noreferrer" target="_blank">http://www.clusterlabs.org</a><br>
Getting started: <a href="http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf" rel="noreferrer" target="_blank">http://www.clusterlabs.org/<wbr>doc/Cluster_from_Scratch.pdf</a><br>
Bugs: <a href="http://bugs.clusterlabs.org" rel="noreferrer" target="_blank">http://bugs.clusterlabs.org</a><br>
<br></blockquote></div><br><br clear="all"><br>-- <br><div class="gmail_signature" data-smartmail="gmail_signature"> .~.<br> /V\<br> // \\<br>/( )\<br>^`~'^</div>
</div>