When running the linter against dahdi-linux.git or potentially other repositories with out-of-tree linux kernel modules, it should not complain about the LINUX_VERSION_CODE being used there:
> drivers/dahdi/dahdi-base.c:6397: WARNING:LINUX_VERSION_CODE: > LINUX_VERSION_CODE should be avoided, code should be for the version to > which it is merged
Set wipe-workspace for the onomondo-eim and onomondo-ipa jobs to true, so previous failures cannot influence new jenkins job runs. E.g. currently we had gerrit-onomondo-ipa-build failing with:
Most existing jobs call scripts/osmo-clean-workspace.sh from osmo-ci.git at the beginning of their jenkins.sh instead of using wipe-workspace. But this doesn't have any advantage here while making jenkins.sh more complex. Let's use the built-in feature from jenkins here.
When the server returns an error, the curl command does not fail - probably because the server sends a wrong http status code:
curl \ --form token="$token" \ --form email=holger@freyther.de --form file=@Osmocom.tgz \ --form version=Version --form description=AutoUpload \ https://scan.coverity.com/builds?project=Osmocom + set +x % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed … <html> <head><title>413 Request Entity Too Large</title></head> <body> <center><h1>413 Request Entity Too Large</h1></center> <hr><center>nginx/1.19.10</center> </body> </html> Finished: SUCCESS
Fail if the output doesn't contain the "Build successfully submitted" string, so the job fails if there was an error.
(The request entity too large error is probably an error on their end, scan.coverity.com still says they are performing upgrades on their servers.)
