Bug 811382

Unable to reproduce on dev.

I've tried to test this on devenv_1757.

Since only root user have sudo permission, create another user and try to sudo will get "bash: /usr/bin/sudo: Permission denied"

[root@ip-10-100-211-171 ~]# ls -lh /usr/bin/sudo
---x--x---. 2 root wheel 208K Jul 21  2011 /usr/bin/sudo

And sudo with root user itself, can not reproduce it  as well


#sudo rhc-admin-ctl-district -h
== Synopsis

rhc-admin-ctl-district: Control districts

== Usage

rhc-admin-ctl-district OPTIONS

Options:
-u|--uuid     <district uuid>
    District uuid  (alphanumeric)
-c|--command <command>
    (add-node|remove-node|deactivate-node|activate-node|add-capacity|remove-capacity|create|destroy)
-n|--name <district name>
    District name (Used on create or in place of uuid on other commands)
-p|--node_profile <profile_size>
    (small(default)|jumbo|exlarge|large|medium|micro) Only needed for create
-i|--server_identity

Hi, Thomas

Can you help to check if that's what you want and if it works for you now?
 

Thanks

Yeah, I tried this out on Krishna's devenv on Friday and it seems to be working now.

I believe whoever fixed the ruby CONST errors also fixed this problem.

NOTE: this still fails in STG and PROD right now, so I'll double check this once we push the next release to STG.

Hi, Thomas

Does this work for you on STG and PROD?

Thanks

Unfortunately it still doesn't work in STG and PROD.

I worked with Krishna on this, and I tested it in his devenv and it worked fine. So there is something different between STG/PROD and the devenvs.

Note: Another difference is that when we run commands in STG and PROD, we still get those annoying "warning: already initialized constant" messages from the rhc-admin-* commands. We don't get those in the devenv anymore. So whatever fixed those messages in the devenv might be what fixed this problem too.

Moving back to assigned.


[root@REDACTED ~]# sudo rhc-admin-ctl-district
/usr/lib/ruby/gems/1.8/gems/mongo-1.5.2/lib/mongo/util/pool_manager.rb:282:in `get_valid_seed_node': Cannot connect to a replica set using seeds localhost:27017, : (Mongo::ConnectionFailure)
	from /usr/lib/ruby/gems/1.8/gems/mongo-1.5.2/lib/mongo/util/pool_manager.rb:158:in `connect_to_members'
	from /usr/lib/ruby/gems/1.8/gems/mongo-1.5.2/lib/mongo/util/pool_manager.rb:30:in `connect'
	from /usr/lib/ruby/gems/1.8/gems/mongo-1.5.2/lib/mongo/repl_set_connection.rb:160:in `connect'
	from /usr/lib/ruby/gems/1.8/gems/mongo-1.5.2/lib/mongo/repl_set_connection.rb:500:in `setup'
	from /usr/lib/ruby/gems/1.8/gems/mongo-1.5.2/lib/mongo/repl_set_connection.rb:144:in `initialize'
	from /usr/lib/ruby/gems/1.8/gems/stickshift-controller-0.10.8/lib/stickshift-controller/lib/stickshift/mongo_data_store.rb:121:in `new'
	from /usr/lib/ruby/gems/1.8/gems/stickshift-controller-0.10.8/lib/stickshift-controller/lib/stickshift/mongo_data_store.rb:121:in `db'
	from /var/www/stickshift/broker/lib/express/broker/mongo_data_store.rb:114:in `district_collection'
	from /var/www/stickshift/broker/lib/express/broker/mongo_data_store.rb:22:in `find_all_districts'
	from /usr/lib/ruby/gems/1.8/gems/stickshift-controller-0.10.8/lib/stickshift-controller/lib/stickshift/mongo_data_store.rb:168:in `rescue_con_failure'
	from /var/www/stickshift/broker/lib/express/broker/mongo_data_store.rb:21:in `find_all_districts'
	from /var/www/stickshift/broker/app/models/district.rb:37:in `find_all'
	from /usr/bin/rhc-admin-ctl-district:180
[root@REDACTED ~]#

FYI,

if I edit the /etc/sudoers

and change:

Defaults    env_reset

to

Defaults    !env_reset

It seems the command starts working fine.

(In reply to comment #6)

> 
> It seems the command starts working fine.

Moving to On_QA

Hi Thomas,

Since the issue occurs on STG and PROD only, and for QE, we do not have permission to test the bug. Could you help test this on STG and PROD.

Thanks.

Yep, this is working now.

Closing bug.