pollinate at 4.25-0ubuntu1~16.04.1 and 4.25-0ubuntu1~14.04.1 is in xenial and trusty queue: https://launchpad.net/ubuntu/xenial/+queue?queue_state=1&queue_text=pollinate https://launchpad.net/ubuntu/trusty/+queue?queue_state=1&queue_text=pollinate
** Description changed: === Begin SRU Template === [Impact] - The key change is that the user agent string as of Zesty and Artful now contains the contents of /proc/uptime. In aggregate, this gives us important performance data to detect improvements and regressions in Ubuntu boot times. + The key change is that the user agent string as of Zesty and Artful now + contains the contents of /proc/uptime. In aggregate, this gives us + important performance data to detect improvements and regressions in + Ubuntu boot times. [Test Case] + Pollinate does not provide an easy way to test either of the changes + that were added. + + The two changes are: + a.) the defaults file /etc/default/pollinate file was sourced twice during + execution. That was unlikely to cause issue, but the cleanup removed + unnecessary duplication. + + b.) uptime is read from /proc/cpuinfo and added to the USER_AGENT + variable that is fed to curl. + + So realistically the best test is to just check for regression. Note + that this cannot be done in a container as pollinate does not run + in a container. + + a.) start an instance + b.) enable proposed, install updated pollinate. + c.) rm -Rf /var/cache/pollinate/ + d.) reboot + e.) look at status of pollinate: + service pollinate status. [Regression Potential] + The script runs with 'set -e', and reads from /proc/uptime with: + read up idle < /proc/uptime + + That means that the 'pollinate' script will exit failure if either: + a.) /proc/uptime did not exist (possibly proc not mounted) + b.) a read error occurred. + + These scenarios seem unlikely, especially in Ubuntu environments. + The upstream code should check to at least check the readability of + /proc/uptime first. [Other Info] === End SRU Template === ** Description changed: === Begin SRU Template === [Impact] The key change is that the user agent string as of Zesty and Artful now contains the contents of /proc/uptime. In aggregate, this gives us important performance data to detect improvements and regressions in Ubuntu boot times. [Test Case] Pollinate does not provide an easy way to test either of the changes that were added. The two changes are: a.) the defaults file /etc/default/pollinate file was sourced twice during execution. That was unlikely to cause issue, but the cleanup removed unnecessary duplication. b.) uptime is read from /proc/cpuinfo and added to the USER_AGENT variable that is fed to curl. So realistically the best test is to just check for regression. Note that this cannot be done in a container as pollinate does not run in a container. a.) start an instance b.) enable proposed, install updated pollinate. c.) rm -Rf /var/cache/pollinate/ d.) reboot e.) look at status of pollinate: - service pollinate status. + service pollinate status. [Regression Potential] The script runs with 'set -e', and reads from /proc/uptime with: - read up idle < /proc/uptime + read up idle < /proc/uptime That means that the 'pollinate' script will exit failure if either: a.) /proc/uptime did not exist (possibly proc not mounted) b.) a read error occurred. These scenarios seem unlikely, especially in Ubuntu environments. The upstream code should check to at least check the readability of /proc/uptime first. [Other Info] + The reading of up/idle from uptime was originally added under bug 1638552. === End SRU Template === -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1708192 Title: SRU pollinate 4.25 To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/pollinate/+bug/1708192/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
