yahoo-eng-team team mailing list archive

["Markus Zoeller \(markus_z\)" <mzoeller@xxxxxxxxxxxxxxxxxx>]

This is an automated cleanup. This bug report has been closed because it
is older than 18 months and there is no open code change to fix this.
After this time it is unlikely that the circumstances which lead to
the observed issue can be reproduced.

If you can reproduce the bug, please:
* reopen the bug report (set to status "New")
* AND add the detailed steps to reproduce the issue (if applicable)
* AND leave a comment "CONFIRMED FOR: <RELEASE_NAME>"
  Only still supported release names are valid (LIBERTY, MITAKA, OCATA, NEWTON).
  Valid example: CONFIRMED FOR: LIBERTY


** Changed in: nova
   Importance: Medium => Undecided

** Changed in: nova
       Status: Confirmed => Expired

-- 
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/1382153

Title:
  n-cond shoul not joining to servicegroup an all worker

Status in OpenStack Compute (nova):
  Expired

Bug description:
  All nova conductor worker process attempts to join to the service on the same host. It does not seams required.
  If you have 48 conductor worker on a node, it means it tries to maintain the membership with all 48 worker.

  Since the workers are started almost at the same time, it means 48
  burst update attempt close to each other.

  The situation even worse with zk driver,  it does not works with multiple workers >1 , because all worker thread inherited the same zookeeper connection from it's parent.  (4096 connection allowed from the same ip on my zk servers)
  (The api service does not do status report, so it can work with multiple workers)

  The  "lsof -P |grep cond | grep 2181"  indicates all conductor worker
  uses the same tcp source port  --> the same socket inherited.

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1382153/+subscriptions