Believe me I have extreme precautions, whatever this was, it pierced 
through it all

I have retry_limit's as 2-4's on all queues, and my mapreduce tasks names 
are chained, so if a top-task is replicated / or pushes a task more than 
once, that task isn't repeated as the name clashed with the 
executed/existing one

So worst case scenario, if the main trigger tasks somehow executed 4 times, 
which would be erroneous, but lets say it did, I wouldn't even noticed the 
fluke, it would boost the daily cost by 40% and not do any damage

Somehow, I'm guessing, the trigger task executed like 1000+ times, 
triggering many new mapreduce routines (custom routine, more of a map 
rather than reduce, it's proved itself for years)

So like I said, whatever this was, it pierced through it all

It's obvious there was a major issue, If google reported the issue, from 
the announcements group, like they do for many minor issues, this all could 
have been prevented, but they didn't, the lack of official announcement or 
update from google makes me think they are going to cover this one up

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to google-appengine+unsubscr...@googlegroups.com.
To post to this group, send email to google-appengine@googlegroups.com.
Visit this group at http://groups.google.com/group/google-appengine.
For more options, visit https://groups.google.com/d/optout.

Reply via email to