Hi all,
I'm being bitten by the weird "waiting for Pulp to start the task" problem.
I freshly installed katello on a CentOS7 server, and subscribed a CentOS5
server to it.
I can see the CentOS 5 server as a Content host, and added a content view
etc … all repo syncing also works just fine. But whenever I try to
install/remove a package
on the server, it just won't start and seems blocked at "waiting for Pulp
to start the task".
I already restarted all pulp services (heck, even rebooted the katello
server), still nothing.
Tried updating the EL5 server to the latest katello client packages
(version 2.4, which is a burden since they are not correctly signed for EL5
servers).
So any help on this would be appreciated
Below you can find the dynflow console output:
3: Actions::Pulp::Consumer::ContentUninstall (waiting for Pulp to start the
task) [ 1490.61s / 4.11s ] Cancel
<https://btcab0004ap.unix.banksys.be/foreman_tasks/dynflow/35581a75-05a1-4db9-b3d6-23aa4eb8546c/cancel/3>
Started at: 2016-01-25 13:13:14 UTC
Ended at: 2016-01-25 13:38:05 UTC
Real time: 1490.61s
Execution time (excluding suspended state): 4.11s
Input:
···
--- consumer_uuid: a83cde10-6097-4240-8a9a-f82398c5c8a1 type: rpm args:- afick remote_user: admin remote_cp_user: admin locale: enOutput:
pulp_tasks:- exception:
task_type:
_href: /pulp/api/v2/tasks/75ad7835-61fd-4121-af1b-ceb53782889e/
task_id: 75ad7835-61fd-4121-af1b-ceb53782889e
tags:
- pulp:consumer:a83cde10-6097-4240-8a9a-f82398c5c8a1
- pulp:action:unit_uninstall
finish_time:
_ns: task_status
start_time:
traceback:
spawned_tasks: []
progress_report: {}
queue: agent.dq
state: waiting
worker_name: agent
result:
error:
_id:
$oid: 56a61f6a41bc8a5d169ae24a
id: 56a61f6a41bc8a5d169ae24a
poll_attempts:
total: 113
failed: 0
With friendly regards,
Franky