Description of problem: Old scalable app exist on stage, after server upgraded and migrated, try to change the repo and push the build via git push, will fail with exit code 1. ============= <---> remote: Warning: Identity file /var/lib/stickshift/25fa25c7cf72430b8a1c574dc7cd1947//haproxy-1.4/.ssh/haproxy_id_rsa not accessible: Permission denied. remote: rsync: connection unexpectedly closed (0 bytes received so far) [sender] remote: rsync error: error in rsync protocol data stream (code 12) at io.c(600) [sender=3.0.6] remote: Exit code: 1 <---> Version-Release number of selected component (if applicable): stage_2.0.16 How reproducible: always Steps to Reproduce: 1.Old scalable app exist on server 2.Do server upgrade and migrate 3.Change the app repo and git push Actual results: Build failed with exit code 1. Expected results: Push build should not have any problem. Additional info: This issue does not exist on INT after server upgrade. This issue does not exist for the new created scalable app on STG.
Login to the haproxy gear and ssh to its another gear in this scaling app, failed with the same error : [phps-stgdm4.stg.rhcloud.com conf]\> ssh b168350d05-stgdm4.stg.rhcloud.com Warning: Identity file /var/lib/stickshift/0aeb5a9836294074bbb6017834ace16b//haproxy-1.4/.ssh/haproxy_id_rsa not accessible: Permission denied.
Where did this app come from? Was it always on stg or was it unpacked from a backup?
Setting low priority. I believe this is due to the SELinux hiccup last sprint which was resolved except for this app.
Added comment to the release ticket to run rhc-restorecon by hand on every exec node except the C9 nodes.
The rhc-restorecon command was run on the stg node containing the bad application and it fixed the SELinux label.
Checked on latest stage env, for my existing applications, some of them have been fixed. But some of them still can get the same error. 50% of them have been fixed. I list my failed apps info. account: qgong apps: perl1s-qgongstg.stg.rhcloud.com no1s-qgongstg.stg.rhcloud.com py1s-qgongstg.stg.rhcloud.com account: bmeng+1 apps: jbas1s-bmeng1stg.stg.rhcloud.com
We think this is solved now, moving to on_qa. Please let us nkow if it's still happening. - Bug Triage Meeting
Verified on stage After upgrade on stage, git push an old existing scaling app was successful.