Cloud-init randomly fail to get nova metadata
Bug #1766824 reported by
Damien Ciabrini
This bug affects 1 person
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| novajoin |
Fix Released
|
Undecided
|
Damien Ciabrini | ||
Bug Description
From time to time it seems that we experience contention on novajoin, which makes cloud-init fail to retrieve metadata when doing:
curl -s http://
To workaround the chances of contention we currently have a random sleep before that call so lower the odds that many nodes call novajoin at the same time.
This random sleep reduces the failure rate at deploy time but we can still experience it fairly often.
| Changed in novajoin: | |
| assignee: | nobody → Damien Ciabrini (dciabrin) |
| description: | updated |
| Changed in novajoin: | |
| status: | New → In Progress |
Revision history for this message
| OpenStack Infra (hudson-openstack) wrote Fix merged to novajoin (master) | #1 |
| Changed in novajoin: | |
| status: | In Progress → Fix Released |
To post a comment you must log in.
Reviewed: https:/
Committed: https:/
Submitter: Zuul
Branch: master
commit ed5de4f1383bda6
Author: Damien Ciabrini <email address hidden>
Date: Wed Apr 25 10:53:02 2018 +0200
Add a retry loop when fetching vendor_data from novajoin
In [1] we added a random sleep in the script called by cloud-init
to fetch vendor_data, to lower chances that many nodes connect to
novajoin concurrently. Still, curl failures still happen from
time to time.
Add a retry loop around the call to curl to further increase the
chances of success, as this seems to give good results in practice
when deploying overclouds with tripleo.
[1] 188f09ef22a2333
Closes-Bug: #1766824
Change-Id: I728c2cb0c8a743