search

fedora-copr / copr

RPM build system - upstream for https://copr.fedorainfracloud.org/
115 stars 61 forks source link

Builders fail to spawn - timeouting on subscription manager after 10 minutes #3108

Closed praiskup closed 8 months ago

praiskup commented 8 months ago

TASK [install the subscription-manager script] ***** Wednesday 17 January 2024 13:43:22 +0000 (0:00:00.098) 0:00:08.367 * skipping: [2620:52:3:1:dead:beef:cafe:c149]

TASK [Activate Red Hat Subscription] *** Wednesday 17 January 2024 13:43:22 +0000 (0:00:00.090) 0:00:08.458 * DEBUG:root: -> exit_status=124, time=600.011s DEBUG:root:Cleaning up ...

praiskup commented 8 months ago

It is fine on AWS, only 16 seconds:

Wednesday 17 January 2024  14:49:59 +0000 (0:00:00.081)       0:00:52.791 ***** 
===============================================================================
Activate Red Hat Subscription ------------------------------------------ 16.56s
praiskup commented 8 months ago

Is this again ipv6 related?

praiskup commented 8 months ago

Rebooting vmhost-x86-copr02.rdu-cc.fedoraproject.org helped.

praiskup commented 8 months ago

Ok, seems like restarting libvirtd daemon was enough, for some reason. I did that on all hypervisors. I'll close tomorrow after checking that everything is fine.

NeilHanlon commented 8 months ago

Is this again ipv6 related?

it could be, since RHSM isn't available via v6.