per T241852#5945821 15 new appservers in codfw can't be racked until at least 15 old servers have been removed.
Therefore rack C3 should have priority before other "decom old servers" tasks and we should start to take some out of there to make room.
per T241852#5945821 15 new appservers in codfw can't be racked until at least 15 old servers have been removed.
Therefore rack C3 should have priority before other "decom old servers" tasks and we should start to take some out of there to make room.
Change 579073 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/puppet@production] decom 15 codfw appservers
Mentioned in SAL (#wikimedia-operations) [2020-03-11T22:28:02Z] <mutante> depooled mw2167 through mw2172 - rack C3 (T247018)
mw2158 through mw2172 are permanently depooled (state=inactive) now. That's exactly 15 servers from the middle of C3. Set them to status "decommissioning" in netbox.
Icinga downtime for 12:00:00 set by dzahn@cumin1001 on 15 host(s) and their services with reason: decom
mw[2158-2172].codfw.wmnet
Reverted this. They are back in the pool for now because this got stalled for a week or 2, so why not keep using them for right now.
Mentioned in SAL (#wikimedia-operations) [2020-05-21T12:13:19Z] <mutante> depooled mw2158 through mw2172 to make room again in C3 as planned (T247018)
Change 597771 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/puppet@production] site: remove 13 old jobrunners from codfw rack C3
Icinga downtime for 4 days, 0:00:00 set by dzahn@cumin1001 on 15 host(s) and their services with reason: decom
mw[2158-2172].codfw.wmnet
Change 597771 merged by Dzahn:
[operations/puppet@production] site: remove 13 old jobrunners from codfw rack C3
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw2150.codfw.wmnet
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw[2151-2155].codfw.wmnet
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw[2156-2159].codfw.wmnet
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw[2160-2162].codfw.wmnet
Change 598025 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/puppet@production] site: decom mw2163 through mw2169 appservers
Change 598025 merged by Dzahn:
[operations/puppet@production] site: decom mw2163 through mw2169 appservers
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw[2163-2166].codfw.wmnet
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw[2167-2169].codfw.wmnet
@Papaul 23 servers from rack C3 have been decom'ed. mw2150 through mw2172. (lower part of the rack)
You can:
Next week I am planning to remove even more so we might be able to empty out almost the entire C3.
Change 598036 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/puppet@production] site: decom mw2170 - mw2172
Change 598039 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/puppet@production] DHCP: remove recently decom'ed codfw appservers
Change 598040 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/puppet@production] mcrouter: replace proxy in codfw row C
Change 598040 merged by Dzahn:
[operations/puppet@production] mcrouter: replace proxy in codfw row C
Change 598036 merged by Dzahn:
[operations/puppet@production] site: decom mw2170 - mw2172
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw[2170-2172].codfw.wmnet
Change 598039 merged by Dzahn:
[operations/puppet@production] DHCP: remove recently decom'ed codfw appservers
@Dzahn Please to not resolve yet. I still have mgmt DNS and switch port to remove.
Thanks
Change 599603 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/puppet@production] site: decom mw2173 through mw2179
Change 599606 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/puppet@production] site: decom mw2180 through mw2186
Change 579073 abandoned by Dzahn:
decom 15 codfw appservers from rack C3
Reason:
duplicate, already done in other changes
Change 599610 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/dns@master] remove production IPs of mw2163 through mw2172
Change 599610 merged by Dzahn:
[operations/dns@master] remove production IPs of mw2163 through mw2172
Change 599614 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/dns@master] remove production IPs of mw2150 through mw2162
Change 599614 merged by Dzahn:
[operations/dns@master] remove production IPs of mw2150 through mw2162
@Papaul I also still had to remove production IPs. Done for mw2150 through mw2172.
Also uploaded new changes to decom mw2173 through mw2186 but they are not merged yet. Once they are this rack will be empty (of mw servers, minus the special cases that are not mw).
I see you started racking 4 of the new servers and made https://gerrit.wikimedia.org/r/c/operations/puppet/+/599749 for that (for now).
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw2173.codfw.wmnet
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw[2174-2177].codfw.wmnet
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw[2178-2179].codfw.wmnet
Change 599603 merged by Dzahn:
[operations/puppet@production] site: decom mw2173 through mw2179
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw[2180-2183].codfw.wmnet
cookbooks.sre.hosts.decommission executed by dzahn@cumin1001 for hosts: mw[2184-2186].codfw.wmnet
Change 599606 merged by Dzahn:
[operations/puppet@production] site: decom mw2180 through mw2186
Change 602012 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/dns@master] remove production IPs for mw2173 through mw2186
Change 602012 merged by Dzahn:
[operations/dns@master] remove production IPs for mw2173 through mw2186
Change 602013 had a related patch set uploaded (by Dzahn; owner: Dzahn):
[operations/dns@master] remove mgmt IPs for mw2150 through mw2186
@Papaul All remaining old mw servers in rack C3 (mw2154 through mw2186) are also decom'ed now. Removed from site, production IPs removed, only pending removing mgmt IPs (https://gerrit.wikimedia.org/r/c/operations/dns/+/602013).
You can feel free to remove and replace them with new servers and use C3 for new cabling (except the thumbor/db servers). Let me know when we are ready to merge the change above to remove mgmt IPs as well.
I am assigning to you for now because you said "don't resolve yet, i still have switch ports to do" etc. Please assign back to me once that is done.
@Papaul Thanks! Does that mean https://gerrit.wikimedia.org/r/c/operations/dns/+/602013 can be merged?
Change 602013 merged by Dzahn:
[operations/dns@master] remove mgmt IPs for mw2150 through mw2186