I've upgrade most of the servers now, and labs should behave much as
you'd expect. There are two caveats:
- Info pages on wikitech for new instances will not be present or
updated. This is because of a well-understood issue that I plan to fix
in the next few hours.
- All new instances will be scheduled on labvirt1004 for the next few
days. Once we're convinced that everything is working normally I'll
upgrade the other labvirt nodes and open up the schedule pool -- no one
should notice the difference, really, before or after.
There was a brief network outage during this window, but most of you
shouldn't have noticed much of anything. Let me know if you see any
new, bad behavior.
-Andrew
On 8/26/15 12:55 PM, Andrew Bogott wrote:
I'm still working on this, a bit. Existing labs instances and wikitech
should be fine -- the remaining issue is the creation of new
instances, which I hope to have working again in the next hour or so.
-Andrew
On 8/20/15 3:51 PM, Andrew Bogott wrote:
Thanks to Chase's help with network upgrades yesterday, the Labs
cluster is now all set for a few OpenStack upgrades. I'm going to
start the first of these upgrades (from 'Icehouse' to 'Juno') on
Wednesday at 16:00 UTC.
This should not disturb any services running within Labs.
External interactions with OpenStack (e.g. instance creation,
wikitech special pages, floating IP manipulation) may suffer
occasional interruptions.
If all goes well with this upgrade then I will attempt a second
update from Juno to Kilo the following week. Getting caught up with
the latest OpenStack version will allow for improved user interfaces
for a variety of labs features.
-Andrew
_______________________________________________
Labs-announce mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/labs-announce
_______________________________________________
Labs-l mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/labs-l
