[Spacewalk-list] metadata_expire and Spacewalk updates

[Dennis Pittman]

I have experienced that issues as well and my current plan is to build better scripts/api calls to trigger a faster regen of necessary cache data.  Currently the plan is to allow whatever distribution mechanism produce the new package/config (resource(s)) that will managed and delivered from spacewalk.  Once the new resource is ready and available then sync'd into the spacewalk server, we can then manage the spacewalk server to make certain that the latest resource(s) readily available to the managed clients.
Actions for spacewalk server:
	*	Sync resource to spacewalk server (ingest via rsync, rhnpush, satellite-sync, and etc.)
	*	Once new resource(s) is available, you need to manually update all necessary cache (using spacecmd )
	*	Last step would be to make certain that all clients are working from the latest available cache
I know this is not the answer you're looking for, but to some up the actions necessary, you'll need to consistently develop extra internal functions to make certain that your spacewalk server implementation is based configured to be able to play a key role your continuous delivery development environment.
Here's some good references: 
	* 	devops talk:   http://www.devops-blog.net/page/2
	*	missing updated metadata:  https://access.redhat.com/solutions/19303
Note:  There's other pieces of information needed here, but I assume this is a good start

Dennis J. Pittman
DP HomeBox Office ITS
(e) dennis.j.pittman at gmail.com

-----Original Message-----
From: spacewalk-list-bounces at redhat.com <spacewalk-list-bounces at redhat.com> On Behalf Of Brian Long
Sent: Friday, February 21, 2020 10:05 AM
To: spacewalk-list at redhat.com
Subject: [Spacewalk-list] metadata_expire and Spacewalk updates

I would like to know how folks are handing the default metadata_expire of six hours when they push updates via Spacewalk.  My team has run into issues for years where we will sync new updates to a channel, wait for the channel repodata to be rebuilt in Spacewalk, then select a bunch of hosts to update.  Many of the hosts will checkin immediately, see zero updates and mark the task completed.  I believe this is because they didn't refresh their metadata and they were seeing cached metadata.

How are folks forcing the metadata update when pushing updates via Spacewalk?  Do you set yum.conf metadata_expire really low across all your clients?  Or do you somehow force a metadata update before telling Spacewalk to push the latest updates?

Thank you for any suggestions.

/Brian/


_______________________________________________
Spacewalk-list mailing list
Spacewalk-list at redhat.com
https://nam03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.redhat.com%2Fmailman%2Flistinfo%2Fspacewalk-list&data=02%7C01%7C%7Cb0fb4e72cc7643193c6e08d7b6df8a4d%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637178943514502032&sdata=p0D7Cy15n%2BkbYeLKOZ%2FFr34FqgB%2BYUkVOaCSuVG15VI%3D&reserved=0