Description of problem: When running snippet update, it's not unusual to have the file open in an external editor for an extended period of time. But if the session times out during this time, the edits are lost when one exits the editor Version-Release number of selected component (if applicable): spacecmd-1.8.15-1 How reproducible: Every time Steps to Reproduce: 1. snippet-update my-favourite-snippet 2. Do some extensive editing 3. ERROR: redstone.xmlrpc.XmlRpcFault: unhandled internal exception: Could no find session with id: 1330295 Actual results: ERROR: redstone.xmlrpc.XmlRpcFault: unhandled internal exception: Could no find session with id: 1330295 Lots of bad words being uttered... Expected results: Language being much more civilised (-: Additional info: If it' matters, I've been running nightly spacecmd on RHEL6
Forwarding to upstream.
Reassigning to spacecmd maintainer.
So this issue could be fixed in spacecmd by not opening the API session until after the edit is completed, but IMO a probably simpler solution is a workflow adjustment: 1. Keep the snippets in a local git/svn repo, edit, then update the snippet after the edit with a non-interactive snippet_update -f snippetfile -n snippetname or 2. Keep the snippets in a local git/svn repo, and setup a trigger script which updates a checkout under the /var/lib/rhn/kickstarts/snippets directory, changing /var/lib/rhn/kickstarts/snippets/<orgnum> to be a symlink to the checkout Note that for test environments you could just edit the snippets directly under /var/lib/rhn/kickstarts/snippets/<orgnum> instead of using the spacecmd interactive mode, which again would solve this problem. Also, I think the same problem would occur doing a lengthy edit via the web UI, in which case the only solution would be to extend the web session cookie timeout (which would also extend the time before the problem described in this bug occurs): https://access.redhat.com/knowledge/solutions/11001 So with the above in mind, IMO reworking the spacecmd session management logic to solve this problem is probably not justified - propose closing this bug WONTFIX if the reported is happy with the workflow suggestions above.
Reassigning to Aron Parsons, spacecmd maintainer
(In reply to comment #3) > So this issue could be fixed in spacecmd by not opening the API session > until after the edit is completed, but IMO a probably simpler solution is a > workflow adjustment: Agreed. One should not be relying on a vim session as their only record of a snippet modification. I don't think many people are going to hit this, so closing as WONTFIX.
This BZ closed some time during 2.5, 2.6 or 2.7. Adding to 2.7 tracking bug.