sodonnel commented on PR #5488:
URL: https://github.com/apache/ozone/pull/5488#issuecomment-1804256131

   Just to be sure I am not missing something - this PR does not provide the 
complete functionality? For example, the containers pending replication are not 
yet included in the output?
   
   We already have metrics in NodeDecommissionMetrics that keeps track of all 
nodes decommissioning, and also the various counts of under-replicated, 
open-pipelines etc. The only thing the metrics do not provide, is the list of 
containers in the degraded states.
   
   I think the baseline information could therefore be provided by pulling the 
metrics from SCM (like the ozone insight commands do), and displaying it in a 
report form. Then we just need a way to get a list of the containers that are 
blocking decommission / maintenance by state, which could be obtained via a 
single new call into SCM.
   
   We would not be able to get the starting container counts vs current counts, 
but these counts are not overly important unless some containers refuse to 
close - the counts at the start are not overly interesting.
   
   I am not sure if the decommission start time is part of the metrics, but we 
could easily add it if it is not there already.
   
   I think it is worth considering this approach, as it means we have a 
consistent view of metrics and the report output, and the numbers should align. 
I also think it will be much less code.


-- 
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.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to