[ClusterLabs] cluster unable to start a resource which otherwise starts

Windl, Ulrich u.windl at ukr.de
Wed Mar 4 07:48:22 UTC 2026


Hi!

Another thing is:
monitor: KEA-DHCP4-monitor-interval-60s
      interval=60s timeout=100s

IMHO having an interval shorter than the timeout makes little sense.

Kind regards,
Ulrich Windl

From: Users <users-bounces at clusterlabs.org> On Behalf Of Andreas Mock via Users
Sent: Thursday, February 26, 2026 9:55 AM
To: Cluster Labs - All topics related to open-source clustering welcomed <users at clusterlabs.org>
Cc: Andreas Mock <andreas.mock at drumedar.de>
Subject: [EXT] [EXT] Re: [ClusterLabs] cluster unable to start a resource which otherwise starts

Hi,

when the systemctl way works (I assume as root) and otherwise not than there a commonly two problems:

  1.  Environment
  2.  User Rights
  3.  SELinux (if enabled)

I would start to investigate with b)

Best regards
Andreas


Von: Users <users-bounces at clusterlabs.org> Im Auftrag von lejeczek via Users
Gesendet: Dienstag, 24. Februar 2026 11:28
An: users at clusterlabs.org
Cc: lejeczek <peljasz at yahoo.co.uk>
Betreff: [ClusterLabs] cluster unable to start a resource which otherwise starts

Hi guys.

I have a simple resource

-> $ pcs resource config KEA-DHCP4
Resource: KEA-DHCP4 (class=systemd type=kea-dhcp4.service)
  Meta Attributes: KEA-DHCP4-meta_attributes
    failure-timeout=25s
  Operations:
    monitor: KEA-DHCP4-monitor-interval-60s
      interval=60s timeout=100s
    start: KEA-DHCP4-start-interval-0s
      interval=0s timeout=100s
    stop: KEA-DHCP4-stop-interval-0s
      interval=0s timeout=100s

which when cluster attempts to star it fails.
I have not tried all the types of resources yet but it seems there might be a pattern.
When the resource is configured to connect to mysql then there is:

 INIT_FAIL failed to initialize Kea server: configuration error using file '/etc/kea/kea-dhcp4.conf': Unable to open database: Can't connect to server on '127.0.0.1' (115)

I've only migrated to Centos 10, this still works on Centos 9 with pacemaker-libs-2.1.10-2.el9
c10 runs with pacemaker-libs-3.0.1-5.el10
kea-dhcp4.service starts with _systemctl_ - no issue.

Do you know what might be the issue?

thanks, L.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/users/attachments/20260304/b22fb81f/attachment.htm>


More information about the Users mailing list