[tracking] netapp c/d upgrade

RESOLVED FIXED

Status

Release Engineering
Other
--
minor
RESOLVED FIXED
8 years ago
4 years ago

People

(Reporter: mrz, Unassigned)

Tracking

Firefox Tracking Flags

(Not tracked)

Details

Attachments

(2 attachments)

Comment hidden (empty)
(Reporter)

Updated

8 years ago
Assignee: server-ops → nobody
Component: Server Operations → Release Engineering
QA Contact: mrz → release
(Reporter)

Comment 1

8 years ago
Created attachment 414693 [details]
VMs on netapp-c
(Reporter)

Comment 2

8 years ago
Created attachment 414694 [details]
VMs on netapp-d
(Reporter)

Comment 3

8 years ago
Filing this bug to track netapp firmware upgrades.  Have been putting these off for months, hoping to take advantage of the Christmas/New Year holiday slowness.

I've attached lists of VMs that will be impacted on both netapp-c & netapp-d.  Normal firmware/OS upgrades are hitless but may require resetting the drive shelves.

VMs that can't go offline can be vmotion'd to DELL storage.
Comment on attachment 414693 [details]
VMs on netapp-c

>production-patrocles
>fx-linux-1.9-slave08
>fx-win32-1.9-slave07
>crazyhorse
>fx-linux-tbox
>production-prometheus-vm
>fx-win32-tbox
>fx-linux-1.9-slave07

These are 1.8 or 1.9 machines which aren't pooled, we should migrate them.

>staging-stage
>staging-puppet
>staging-master

These should stay, if possible.

>bm-vpn01

This should stay up if we've started using it by then.

Everything else can go down IMHO.
(In reply to comment #4)
> These are 1.8 or 1.9 machines which aren't pooled, we should migrate them.

Alternatively, closing the Fx3.0 and Mozilla1.8 trees for a few hours probably wouldn't be a big deal. production-1.8-master would impact on doing a Tb2 firedrill, but that seems pretty unlikely.
 
>> qm-crash-linux01
>> qm-crash-xp01
Ask QA about these ?

crazyhorse is listed on both netapps ?

Which netapp shelf is currently rebuilding after a drive failure ?
(Reporter)

Comment 6

8 years ago
> crazyhorse is listed on both netapps ?

one's inactive.

> Which netapp shelf is currently rebuilding after a drive failure ?

?  The shelf with the drive failure is the one that rebuilds after a drive failure.
Alright, I should have asked the direct question. I heard on IRC that some of the network storage is suffering slower performance than normal because it's rebuilding, does that affect the VMs listed here ? Does it affect any other build VMs directly ?
(Reporter)

Comment 8

8 years ago
It's not the netapp or equallogic arrays.  I think I heard the Sun box had taken a drive failure but we're not doing VMs off that array.
I'd like to schedule these upgrades for Jan 2'nd.  Please let me know if you folks have any objections to the time or to the work.
If you can migrate the 1.8, 1.9 machines in comment#4, great. Otherwise we can send out notice a few days in advance about closing those trees for the duration.

bm-vpn01 should stay up, can that be migrated in advance?
(Reporter)

Comment 11

7 years ago
These?

75GB production-patrocles
60GB crazyhorse
22GB fx-linux-tbox
73GB production-prometheus-vm
45GB fx-win32-tbox
30GB fx-linux-1.9-slave07
30GB fx-linux-1.9-slave08
60GB fx-win32-1.9-slave07

~400GB.

Could move them.  When?
(In reply to comment #11)
> 75GB production-patrocles
> 60GB crazyhorse
> 22GB fx-linux-tbox
> 73GB production-prometheus-vm
> 45GB fx-win32-tbox
> 30GB fx-linux-1.9-slave07
> 30GB fx-linux-1.9-slave08
> 60GB fx-win32-1.9-slave07

These all got moved in the recent downtime, yes?
(Reporter)

Comment 13

7 years ago
no, nothing was moved.

this bug is done though.

Updated

7 years ago
Status: NEW → RESOLVED
Last Resolved: 7 years ago
Resolution: --- → FIXED
(Assignee)

Updated

4 years ago
Product: mozilla.org → Release Engineering
You need to log in before you can comment on or make changes to this bug.