- May 23, 2017
-
-
Eric Cano authored
Added many conditional printouts in unit tests. Those printouts are controlled by #defines in a central file.
-
Steven Murray authored
-
Michael Davis authored
-
Michael Davis authored
-
Michael Davis authored
-
Michael Davis authored
-
Michael Davis authored
-
Michael Davis authored
-
- May 22, 2017
-
-
Steven Murray authored
The cta-frontend and cta-taped rpms now create /var/log/cta
-
Julien Leduc authored
-
Steven Murray authored
Packaged cta-taped.sysconfig and cta-taped.service
-
Steven Murray authored
Packaged cta-frontend.logrotate and cta-taped.logrotate
-
Steven Murray authored
Packaged the man page of cta-taped
-
Vladimir Bahyl authored
-
Julien Leduc authored
-
- May 20, 2017
-
-
Eric Cano authored
-
- May 19, 2017
-
-
Steven Murray authored
Revert "Revert "Removed 'Do nothing' logic from cta::server::Daemon which unwantedly kicks in if systemd is used"" This reverts commit 2d27ddb0. Reverting the revert because nothing changed. The GitLab CI still does not succeeed in running the "init" pod. The commit I made and then reverted appears to have nothing to do with the problems of the GitLab CI.
-
Steven Murray authored
Revert "Removed 'Do nothing' logic from cta::server::Daemon which unwantedly kicks in if systemd is used" This reverts commit 4c7439cf. The current pipeline in the GitLab CI keeps failing to complete the "init" pod's work. The same Docker image is used on stevedevctai.cern.ch to successfully run systest.sh -n ctatest. The GitLab CI does not provide any logs from the "init" pod and therefore cannot be debugged. This revert is a desperate attempt at fixing ithe problem. If it does work then it means the GitLab CI is benefiting from cta-taped remaining as the root user.
-
Steven Murray authored
-
Steven Murray authored
-
Steven Murray authored
-
Eric Cano authored
Fixed calls to promise::get_future() after possible access form other thread. They are now guaranteed to happen before. Added helgrind annotations for promise based synchronisation. Added macros enabling helgrind annotations for shared_ptr. Added suppression for shared_ptr used inside other standard lib object and not covered by the previous macros. Added unit test for lower level . Added suppressions for reported race conditions in Rados library. Review heavily MemArchiveQueue and fixed missing commit in object store, leading to potentially orphaned objects. Enabled formerly disabled test as it is now fast enough.
-
Eric Cano authored
-
Steven Murray authored
-
Steven Murray authored
-
Steven Murray authored
Changed user stage:st to cta:cta
-
- May 18, 2017
-
-
Steven Murray authored
cta-taped now logs the user and group that it is setting for the current process. For example: May 18 16:54:13 itdssbuild01 cta-taped: <158>2017-05-18T16:54:13.402884+02:00 cta-taped[17210]: LVL="Info" TID="17210" MSG="Setting user name and group name of current process" userName="stage" groupName="st"
-
Julien Leduc authored
-
Julien Leduc authored
Add publish_tagged_rpm and needed helper derived from https://gitlab.cern.ch/ci-tools/ci-web-deployer as we need to run the update_repo.sh hook after the file transfert.
-
- May 17, 2017
-
-
Julien Leduc authored
-
Julien Leduc authored
Adding cta_tagged_rpm stage. CTA_RELEASE and CTA_VERSION are not exported as is to avoid overriding environment variables used by cmake. Therefore I am using TAG_VERSION and TAG_RELEASE.
-
Michael Davis authored
-
Michael Davis authored
-
Julien Leduc authored
Fixing issue #108: rpmbuild runs cmake underneath, but this second instance needs to propagate the variables passed to the initial cmake.
-
Steven Murray authored
Fixed. The cta command-line tool will now timeout after 15 seconds when trying to connect to the CTA front-end daemons. The previous timeout was 8 minutes.
-
Michael Davis authored
-
Eric Cano authored
The test remains disabled as it still take quite a bit of time.
-
Eric Cano authored
This is done by replacing unique_ptr/bare pointer with 2 shared_ptr. Also added suppressions for helgrind complains about rados.
-
- May 16, 2017
-
-
Julien Leduc authored
Print CTA_BUILD_ID in the jobs and not the full SHA of the commit of course. The generated rpm version was OK but the logs were useless.
-