yahoo-eng-team team mailing list archive

[Kanagaraj Manickam <mkr1481@xxxxxxxxx>]

Public bug reported:

in the manager.py, ComputeManager has a method "_allocate_network_async"
and uses the CONF parameter "network_allocate_retries". While this
method retries, the logic used is not proper as listed below:

                retry_time *= 2
                if retry_time > 30:
                    retry_time = 30

This bug is filed to correct it as follows:
if retry_time > 30:
    retry_time = 30
else
    retry_time *= 2

This will avoid the calculation of retry time out when the timeout
reaches beyond 30 sec.

** Affects: nova
     Importance: Undecided
     Assignee: Kanagaraj Manickam (kanagaraj-manickam)
         Status: New

** Changed in: nova
     Assignee: (unassigned) => Kanagaraj Manickam (kanagaraj-manickam)

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1334233

Title:
  compute_manager network allocation retries not handled properly

Status in OpenStack Compute (Nova):
  New

Bug description:
  in the manager.py, ComputeManager has a method
  "_allocate_network_async" and uses the CONF parameter
  "network_allocate_retries". While this method retries, the logic used
  is not proper as listed below:

                  retry_time *= 2
                  if retry_time > 30:
                      retry_time = 30

  This bug is filed to correct it as follows:
  if retry_time > 30:
      retry_time = 30
  else
      retry_time *= 2

  This will avoid the calculation of retry time out when the timeout
  reaches beyond 30 sec.

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1334233/+subscriptions