<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body>
<font size="2">Hi<br>
<br>
We are running a 2 node test cluster with the following configuration:<br>
<br>
OS: Rocky 9.2<br>
Slurm: 23.02.5<br>
pmix-5.0.1<br>
<br>
Jobs seem to run OK, however when they do the munge daemon always logs:<br>
2023-10-06 15:37:41 +0200 Info: Unauthorized credential for client UID=0 GID=0<br>
<br>
When running munge -n tests the daemon does <b><u>not</u></b> log these errors.<br>
<br>
<br>
Things we have already done:<br>
<br>
Time <u><b>is</b></u> synced on the two servers using ntp. My colleague and I both checked this with chronyc sources -v:<br>
MS Name/IP address Stratum Poll Reach LastRx Last sample<br>
===============================================================================<br>
^* hidden 2 8 377 254 +11us[ +14us] +/- 11ms<br>
<br>
User gids\uids have been double checked, they are <b><u>identical</u></b> between nodes. This occurs with any user we create.<br>
The munge and slurm user gids\uids are also identical between nodes.<br>
We recreated and redistributed the munge key making sure permissions are set to 400.<br>
Restarted nodes and daemons.<br>
<br>
Q1: Is this being caused by slurmd running as root and the job being run as a normal userid?<br>
<br>
Q2: Is this just a cosmetic message that we can ignore?<br>
<br>
<br>
Andy<br>
</font>Disclaimer - University of Cape Town This email is subject to UCT policies and email disclaimer published on our website at http://www.uct.ac.za/main/email-disclaimer or obtainable from +27 21 650 9111. If this email is not related to the business of
UCT, it is sent by the sender in an individual capacity. Please report security incidents or abuse via https://csirt.uct.ac.za/report-incident
</body>
</html>