<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div style="font-family: "TW Cen MT", "Century Gothic", sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);">
He is an extract of "corosync.log"...</div>
<div style="font-family: "TW Cen MT", "Century Gothic", sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);">
<br>
</div>
<div style="font-family: "TW Cen MT", "Century Gothic", sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);">
Thierry</div>
<div style="font-family: "TW Cen MT", "Century Gothic", sans-serif; font-size: 12pt; color: rgb(0, 0, 0); background-color: rgb(255, 255, 255);">
<br>
</div>
<div id="appendonsend"></div>
<hr style="display:inline-block;width:98%" tabindex="-1">
<div id="divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" style="font-size:11pt" color="#000000"><b>De :</b> Users <users-bounces@clusterlabs.org> de la part de Ulrich Windl <Ulrich.Windl@rz.uni-regensburg.de><br>
<b>Envoyé :</b> mercredi 9 mars 2022 17:13<br>
<b>À :</b> users@clusterlabs.org <users@clusterlabs.org><br>
<b>Objet :</b> [ClusterLabs] Antw: Re: Antw: [EXT] Cluster timeout</font>
<div> </div>
</div>
<div class="BodyFragment"><font size="2"><span style="font-size:11pt;">
<div class="PlainText">>>> FLORAC Thierry <thierry.florac@onf.fr> schrieb am 09.03.2022 um 16:56 in<br>
Nachricht<br>
<PR2P264MB07678DBB0517CB8C7695627CF50A9@PR2P264MB0767.FRAP264.PROD.OUTLOOK.COM>:<br>
<br>
>>>> FLORAC Thierry <thierry.florac@onf.fr> schrieb am 09.03.2022 um 11:46 in<br>
> Nachricht<br>
><br>
<PR2P264MB076751671FC57F33B995F851F50A9@PR2P264MB0767.FRAP264.PROD.OUTLOOK.CO<br>
<br>
> M>:<br>
> <br>
>> Hi,<br>
>><br>
>> I manage an active/passive PostgreSQL cluster using DRBD, LVM, Pacemaker<br>
and<br>
> <br>
>> Corosync on a Debian GNU/Linux operating system.<br>
>> Everything is OK, but my platform seems to be quite "sensitive" to small<br>
>> network timeouts which are generating a cluster migration start from<br>
active<br>
> <br>
>> to passive node; generally, the process doesn't go through to the end: as<br>
>> soon as the connection is back again, the migration is cancelled and the<br>
>> database restarts!<br>
> <br>
> Could it be you run without fencing? Maybe show some logs!<br>
> <br>
> Logs are quite verbose and not very easy to understand...<br>
> What log would you need?<br>
<br>
Those showing what happens when the network goes down, and what happens when<br>
the network comes up.<br>
Usually the DC writes some good "action summaries" (typically after<br>
"pacemaker-controld[7236]: notice: State transition S_IDLE -><br>
S_POLICY_ENGINE"). Those would be helpful.<br>
<br>
> <br>
>> That should be OK but on the application side, some database connections<br>
(on<br>
> <br>
>> a Java WildFly server) can become "invalid"! So I would like to avoid<br>
these<br>
> <br>
>> migrations when this kind of small timeout occurs...<br>
>><br>
>> So my question is: which cluster settings can I change to increase the<br>
>> timeout before starting a cluster migration?<br>
>><br>
>> Best regards,<br>
>> Thierry<br>
>><br>
>><br>
>><br>
>> Thierry Florac<br>
>> Resp. Pôle Architecture Applicative et Mobile<br>
>> DSI ‑ Dépt. Études et Solutions Tranverses<br>
>> 2, avenue de Saint‑Mandé ‑ 75570 Paris cedex 12<br>
>> Tél : 01 40 19 59 64<br>
>> www.onf.fr <<a href="https://www.onf.f<https://www.onf.fr">https://www.onf.f<https://www.onf.fr</a>>r<br>
<br>
<br>
<br>
_______________________________________________<br>
Manage your subscription:<br>
<a href="https://lists.clusterlabs.org/mailman/listinfo/users">https://lists.clusterlabs.org/mailman/listinfo/users</a><br>
<br>
ClusterLabs home: <a href="https://www.clusterlabs.org/">https://www.clusterlabs.org/</a><br>
</div>
</span></font></div>
</body>
</html>