[GitHub] [trafficcontrol] jhg03a commented on issue #3687: Ignoring ADMIN_DOWN Cache Servers while setting parent_pending flag

[GitBox <gi...@apache.org>]

jhg03a commented on issue #3687: Ignoring ADMIN_DOWN Cache Servers while setting parent_pending flag
URL: https://github.com/apache/trafficcontrol/issues/3687#issuecomment-505080703
 
 
   So what is the intended behavior in the following scenario:
   
   1 edge + 2 mids (both reported):
   1. User requests content A/B which hashes to across both mids
   2. Mid 2 is set to ADMIN_DOWN
   3. DS Owner submits broad `/.*` invalidation job
   4. User re-requests content A/B which this time is hashed across only Mid 1
   5. Mid 2 is set back to REPORTED
   6. User re-requests content A/B which rehashes back to where it was in step 1
   
   I'm assuming that means the pending flag on Mid 2 is set at this point which means there is a race condition between syncds and revalidate.  I'm also wondering if this ought to trigger a retroactive parent_pending flag on the edges as well, just to prevent this.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services