We need to return the wdqs hardware to the wdqs team.
Cloudvirt1006 is empty and ready for service (we drained it recently for a battery replacement, T246908. Cloudvirt1004 is untrustworthy because its RAID card freaked out a few weeks ago. With live migration and non-local storage, a future raid-card freakout will be much less scary.
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | Gehel | T221630 [Epic] Search platform - Hardware requests for 2019-2020 | |||
| Resolved | Gehel | T206636 Provide a way to have test servers on real hardware, isolated from production for Wikidata Query Service | |||
| Resolved | Andrew | T221631 Dedicated servers on WMCS to test WDQS scalability strategy | |||
| Resolved | Andrew | T252784 Remove Ceph from cloudvirt-wdqs100x, Add ceph to cloudvirt1004 and cloudvirt1006 |
Change 596466 had a related patch set uploaded (by Andrew Bogott; owner: Andrew Bogott):
[operations/puppet@production] Convert cloudvirt1004 and cloudvirt1006 to ceph-backed cloudvirts
Change 596466 merged by Andrew Bogott:
[operations/puppet@production] Convert cloudvirt1004 and cloudvirt1006 to ceph-backed cloudvirts
Change 596467 had a related patch set uploaded (by Andrew Bogott; owner: Andrew Bogott):
[operations/puppet@production] cloudvirt1004/1006: fix role name
Change 596467 merged by Andrew Bogott:
[operations/puppet@production] cloudvirt1004/1006: fix role name
Mentioned in SAL (#wikimedia-cloud) [2020-05-14T15:47:10Z] <andrewbogott> moving cloudvirt1004 and cloudvirt1006 to the 'ceph' aggregate for T252784
1004 and 1006 are now running ceph; I've drained the wdqs nodes but not rebuilt them yet.
Change 596816 had a related patch set uploaded (by Andrew Bogott; owner: Andrew Bogott):
[operations/puppet@production] Move cloudvirt-wdqs hosts off of ceph
Change 596816 merged by Andrew Bogott:
[operations/puppet@production] Move cloudvirt-wdqs hosts off of ceph