Description of problem: Bug 2009773 had introduced a new clean up mechanism for leftover insights archives in /var/tmp directory: ~~~ Handle SIGTERM signal Archive is created in a directory under /var/tmp with the same prefix "insights-archive-". Each time a new archive is created, it removes the previous one if it exists. Reducing the amount of directories created and stored at the same time. ~~~ It's great that we implemented this but it's not ideal. The leftover archives will be there until the next insights run (usually 24hrs). It would be better to handle the clean-up during or right after insights run fails (for whatever reason).
Hi Jan (@jsenkyri) The archive will only remain in /tmp/insights-archive when insights-client is killed with a SIGKILL signal. Keeping the file due to a failure could help investigate the cause of the problem. If we remove the archive right before the insights-client is killed, there is no room for post-mortem analysis in the archive.
Hi Alba, Right, that makes perfect sense. Thanks for the explanation.
Customer agrees to close this RFE.