My guess is that this is the problem: https://github.com/ManageIQ/manageiq/pull/8707 We changed the server setproctitle to "MIQ Server" but the "rake evm:start" (start evmserverd) mechanism which prevents multiple server processes was looking for evm_server.rb. I'll try to get that PR merged and we can try backporting it to 5.6 to see if it helps.
New commit detected on ManageIQ/manageiq/master: https://github.com/ManageIQ/manageiq/commit/fe8f79677f1ba386d8cb581c848d800131deb549 commit fe8f79677f1ba386d8cb581c848d800131deb549 Author: Joe Rafaniello <jrafanie> AuthorDate: Sat May 14 23:00:38 2016 -0400 Commit: Joe Rafaniello <jrafanie> CommitDate: Mon May 16 10:29:35 2016 -0400 evm.pid / process existence is the safe way to ask if we're running? When we changed the process title using setproctitle to "MIQ Server" [1], we broke rake evm:start's detection of a previously running server. While a process title/command line is somewhat helpful to check, only the server should be writing to evm.pid so checking for a process from evm.pid is enough. https://bugzilla.redhat.com/show_bug.cgi?id=1335391 [1] 1ba7801ee1d2591f17fe05861d726ac0d9a20fa1 app/models/miq_server.rb | 2 +- spec/models/miq_server_spec.rb | 9 +++++++++ 2 files changed, 10 insertions(+), 1 deletion(-)
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2016:1348