[Wg-test-framework] Xen Infrastructure Meeting Minutes Feb 25 2016

[Lars Kurth <lars.kurth@xxxxxxxxxx>]

Please fine attached the etherpad for the meeting at: https://beta.etherpad.org/p/xen-infrastructure-25-Feb-16

And also in-lined below

Regards

Lars

= Attendees =

Xen: Lars Kurth

Credativ: Felix Geyer, Michael Sprengel, Yogesh Patel

See https://beta.etherpad.org/p/xen-infrastructure-11-Feb-16 for older minutes

= New things / Input required / Blocked Items =

65862 BACKUP: we will very probaly need to decide what to back-up - goal: find out how much space we need

• Create a ticket with host and directory information - and then get Ian to respond

• ACTION: Felix to create ticket and prod Ian for input

• Still Open

65862 Rackspace Volume: How much can we use from the free RAX infrastructure

• Would we be able to create a large volume (up to 1 TB) for backup under the existing agreement (comp'ed account) with Rackspace

• ACTION: Felix to provide information on how much disk space we have now, account ID and the number of VMs we use + have retired

• ACTION: Lars to take up with Rackspace, after receiving missing information

1. Updated ticket 65862 with a request for more information, such that I can target the request better

2. Have a contact at Rackspace to ask the question

71794 DNS Coordination

• Have the contact from both on Citrix and LF, which are part of th eticket

• Issue: xen.org and ... - Citrix does not seem to manage the two domains, but DNS records show that Citrix is managing these

• ACTION: Ian to poke Citrix

• Issue: LF wanted more detailed instructions

• ACTION: Felix to follow up with LF to figure out what they need to do

• Need to agree a new date+time once all the issues are sorted (Citrix & LF are based in the US; morning may not work for them)

• ACTION Lars: Agree a time and date with all stakeholders that works for everyone, once the other two actions have been closed

71972 List archives:

• Lars: Happy with the proposal

• ACTION: Credativ to progress

• 
  

70806 New rails for ARM crate

Summary: We do not have the exact part number of the rails, but we do know the chassis (see http://www.pcicase.co.uk/index.php?pag=productd&cid=89&pid=218&p=c4srmc4s24umidsizeentrychassis.html) and would need rails that match the chassis outlined. 

ACTION (Yogesh): See whether the rail can be identified based on  http://www.pcicase.co.uk/index.php?pag=productd&cid=89&pid=218&p=c4srmc4s24umidsizeentrychassis.html & http://lists.xenproject.org/archives/html/wg-test-framework/2016-01/msg00024.html

• See ticket 70806

• Updated the ticket : best assessment without going to the COLO. Next steps: Order

• ACTION Lars/Ian to confirm and then order rails

== Back-ups and Monitoring ==

See https://beta.etherpad.org/p/xen-infrastructure-28-Jan-16 for older minutes

Note that some actions were closed

The urgent actions are tracked under New things/Input required/Blocked Items

== New: Xen Project CI Loop ==

The Xen Project is also operating a CI Loop that tests every OpenStack change against the CI Loop

This infrastructure is part of a separate set of VM's that were set up by Bob Ball

We did have a security breach: http://lists.xenproject.org/archives/html/wg-openstack/2016-02/msg00003.html - search for Security for the relevant section

There is one control VM, which runs Ubuntu which needs to a) be monitored for bandwidth (and other issues) and b) we need to regularly apply security patches. May need to look at back-ups also in futire

All the other VMs are throwaway VMs and get created on demand. So only the control VM needs to monitored

Lars asked whether the fact that this is a Ubuntu VM

Felix stated that it wouldn't be an issue if it is a newer Ubuntu version

Lars: there is a password manager dependency, as all the log-in credentials are different from existing 

Felix stated that the minitoring tools don't do bandwidth monitoring

Lars: Bob stated that Rackspace provides some tooling via their control panael

Longer term: I would also want to investigate whether we can get better global coverage to be easily to take the CI Loop off-line. Right now, we have timezone coverage for Europe and the US, but not Asia. 

ACTION (Lars): Create a a ticket

= Open / Ongoing / Unkown =

== Other itty-bitty bits ==

ACTION: Felix to ensure that Jan & Feb reports are sent

• Note from Lars: this looks sensible, but need to check how much we spend, such that we can stay within budget

• Felix: next report to be sent in the next few days

• Lars created https://docs.google.com/spreadsheets/d/1UQeb2FAmnH29f4oexDHn_-0a2vkTivp9cHZiNIeaTWg/edit#gid=0 to help manage the hour budget (feel free to request access)

== Open Tickets ==

CREATING TICKETS:

• Send mail to xen@support.credativ.com - creates a new ticket. 

• Creating tickets via the web front-end causes issues

• Add CC'ed people from web front-end 

Note that we should close tickets which are completed. Need to respond whether these are good.

ACTION: All please mark tickets which require sign-off by adding a comment here. Maybe just mark the tickets with a * after the ticket number 

65862 Xen Project test colo and rackspace VMs have no backups

67662 Xen test colo software/OS setup needs looking at

69131 Servers to be moved back to rack (elbling0, elbling1, merlot0)

69583 Server(s) (oseleta0/oseleta1) to visited by Dell technician

70045 Upgrade Debian Squeeze VMs to Debian Jessie (xenbits)

70055 Handover Rackspace VMs

70714 Please re-obtain DSETs from oseleta0

70801* Web access to ticket system

70806* Verify Rails for ARM Box

71794 jessie upgrade mail.xenproject.org [linuxfoundation.org #19028]

71972 Lost/unarchived e-mails on xen-devel (when sent from cn.fujitsu.com)

72038* Create few aliases on the mail server

• ACTION LARS (Done): Add the 2 missing mails to the log of the ticket, after checking they have not been deleted accidentially

_______________________________________________
Wg-test-framework mailing list
Wg-test-framework@xxxxxxxxxxxxxxxxxxxx
http://lists.xenproject.org/cgi-bin/mailman/listinfo/wg-test-framework