Leandro Motta Barros
Leandro Motta Barros
> we can probably close this once #1397 is merged Yes, thanks. Just added a "Resolves" link.
> It looks like we could also remove the balena-healthcheck-image related bits from `meta-balena-common/recipes-containers/balena/balena/balena-healthcheck`, `meta-balena-common/recipes-containers/docker-disk/files/entry.sh`, and `meta-balena-common/recipes-containers/docker-disk/docker-disk.bb`. @jakogut I think I did this already on this PR. Maybe a missed...
> hi @lmbarros there's some cleanup left Thanks @alexgg, I updated the PR. > also, did you find out what it takes to starve the engine with this new healthcheck?...
@jakogut, I am happy enough with the PR since the last update. Making it "ready for review".
It's there already (using "Resolves")!
Tried setting `WatchdogSec=0` on an oldish balenaOS 2.83.21 (based on Systemd 244) and got the same error. So, this doesn't look like a recent Systemd change as I thought it...
Checked our Jenkins workers as well, because they use `WatchdogSec=0` to disable the watchdog. This is working there only because we _also_ [redefine `ExecStart`](https://github.com/product-os/balena-jenkins/blob/b593652d56d5780574b15f1e7d55cb63da613051/engine-drop-in/run.sh#L29) so that `healthdog` is not used....
Did a couple more quick tests: * `SIGKILL` leaves the socket unusable in the container, as we already knew. * `SIGABRT` gives the same result as above. (This case might...
The integration tests for delta sizes show that this PR had the positive side-effect of reducing the size in some of the test cases. The new numbers actually make more...
Totally untested as of now, therefore keeping PR as a draft.