Bug 1796851 - Recipe stuck in Scheduled state due to kickstart being too big to be stored in the database
Summary: Recipe stuck in Scheduled state due to kickstart being too big to be stored i...
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Beaker
Classification: Retired
Component: scheduler
Version: 26
Hardware: Unspecified
OS: Unspecified
low
medium
Target Milestone: 28.0
Assignee: Martin Styk
QA Contact: tools-bugs
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2020-01-31 12:10 UTC by Ales Zelinka
Modified: 2020-12-16 18:50 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2020-12-16 18:50:11 UTC
Embargoed:


Attachments (Terms of Use)

Description Ales Zelinka 2020-01-31 12:10:35 UTC
Description of problem:
For some recipes, the valid generated kickstart is so big it doesn't fit into the database "rendered_kickstart" table "kickstart" column (of size 65535).

beakerd tries repeatedly and fails repeatedly, the recipe stays in Scheduled state.

How reproducible:
Have a recipe with a huge kickstart, e.g. clone this one:
https://beaker.engineering.redhat.com/recipes/7853762#installation

Additional info:
The kickstart of the linked recipe is so big because 
a) it is a group job, group having +- 70 members
b) Beaker adds all the ssh keys of all the group members to the kickstart 

The ssh-keys portion of the kickstrat is 40k.

beaker-server error log:
http://pastebin.test.redhat.com/832120

pretty-printed part of the kickstart as parsed from the log message:
http://pastebin.test.redhat.com/832123

Comment 1 Martin Styk 2020-06-16 16:29:22 UTC
Space usable for Kickstart is now 2 ** 24 - 1.
In case you will breach even this limit, the job is aborted.


Note You need to log in before you can comment on or make changes to this bug.