Hi John

Thanks for the quick reply - I added some more detailed info here: 
https://gist.github.com/stackdump/217b2c87f62540fa4807

I didn't include the process that logs f:machine before the sequence, but 
you can see the log entry just after the sequence begins running.

The machine hash already contained a remote_id:

2014-02-28 19:20:26  Env: 5310bda1e1d14826e00000bc Thread: 22454140  - 
Participants::Log: {"CREATING MACHINE"=>{"availability_zone"=>"nova", 
"flags"=>{"migrate"=>true, "wipe"=>true}, "flavor_id"=>"18", 
"image_id"=>"de0bf0b8-8f16-4e0e-bc23-cfc27c52283c", 
"machine_id"=>"5310be96749f06d9ef000051", "name"=>"mlb14-goo-balancer4", 
"puppet_role"=>"role::sdod::playerconnect", 
"remote_id"=>"020235ef-9541-46b3-9f05-b1832daf440d", 
"security_groups"=>["server.balancer"], 
"services"=>{"playcore"=>{"balancer"=>{}}}, "state"=>"ECO_CREATED", 
"tenant_name"=>"GPAD_SD1", 
"user_data"=>"application=mlb14&environment=production pe_eco_environment=ote 
pe_eco_message_broker=eco-ote-messaging.eco.usw1.cld.scea.com", 
"_id"=>"5310e13de1d1483dd90000c0"}, "ref"=>"log"}


Thanks for the debugging tip - I'll keep investigating.

--Thanks
Matt York


On Friday, February 28, 2014 2:52:08 PM UTC-8, John Mettraux wrote:
>
>
> On Fri, Feb 28, 2014 at 02:41:59PM -0800, Matthew York wrote: 
> > 
> > So some strangeness today: 
> > 
> > I have a bit of process def that has this: 
> > 
> >     sequence :unless => '$f:machine.remote_id' do 
> > 
> > I've encountered a scenario where this comparison fails - from my logs I 
> > see a dump of f:machine that does have 'remote_id' populated just before 
> > this process is run. 
> > Somehow the sequence is still executed. 
> > 
> > Note that this particular process is nested under a concurrent_iterator 
> - 
> > and in the other branches 29 other machines (properly) skipped this 
> > sequence, but one did not. 
> > 
> > I've looked over some of the older posts about comparisons, and I don't 
> > think i'm doing anything too strange here. 
> > 
> > Do you think this would be an issue with ruote-mon? ( which I'm still 
> using 
> > with 1 worker ) 
> > I haven't tried to replicate this issue using another storage. 
> > 
> > I'm just now starting to attempt to replicate this issue with a simple 
> test 
> > case, but something tells me this isn't simple. 
>
> Hello Matt, 
>
> I'm sorry, from the information you are giving me, I cannot say much. 
>
> What is remote_id right before the sequence gets hold of it, could it be 
> "" 
> or "false"? The comparison/unless code isn't that smart, it takes a 
> decision, 
> but not a random decision. 
>
> You could had some debug output around line 368 or 
> lib/ruote/exp/flow_expression.rb or in the apply? method of class 
> Condition 
> in lib/ruote/exp/condition.rb to see what's ending up there... 
>
> I hope this will help. 
>
>
> Best regards, 
>
> John 
>
>

-- 
-- 
you received this message because you are subscribed to the "ruote users" group.
to post : send email to [email protected]
to unsubscribe : send email to [email protected]
more options : http://groups.google.com/group/openwferu-users?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"ruote" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to