We need to use Karapace for the DataHub MVP.
Hopefully we won't have to use it for long, because the dependencey might be removed by upstream.
However in the meantime we have decided to deploy a single VM in Ganeti to run this service.
We need to use Karapace for the DataHub MVP.
Hopefully we won't have to use it for long, because the dependencey might be removed by upstream.
However in the meantime we have decided to deploy a single VM in Ganeti to run this service.
Status | Subtype | Assigned | Task | ||
---|---|---|---|---|---|
Resolved | BTullis | T299910 Data Catalog MVP | |||
Resolved | • razzi | T301562 Set up karapace instance for datahub | |||
Resolved | akosiaris | T301563 eqiad: 1 VM requested for karapace | |||
Resolved | • razzi | T301565 Create debian package of karapace |
I have created the VM in T301563 - Now proceeding to boot the machine and make a suitable role.
Change 769468 had a related patch set uploaded (by Btullis; author: Btullis):
[operations/puppet@production] Add boot configuration for karapace1001
Change 769468 merged by Btullis:
[operations/puppet@production] Add boot configuration for karapace1001
Mentioned in SAL (#wikimedia-analytics) [2022-03-09T18:19:28Z] <btullis> btullis@ganeti1024:~$ sudo gnt-instance start karapace1001.eqiad.wmnet (T301562)
The karapace1001.eqiad.wmnet machine has now booted and is ready for karapace to be installed.
Change 771419 had a related patch set uploaded (by Razzi; author: Razzi):
[operations/puppet@production] karapace: add karapace role
Change 771419 merged by Razzi:
[operations/puppet@production] karapace: add karapace role
The server is running; hostname is karapace1001.eqiad.wmnet and port is 8081.
curl karapace1001.eqiad.wmnet:8081 {}
Looks like the monitoring is working as well:
journalctl -u karapace ... Mar 22 19:26:29 karapace1001 python[37435]: aiohttp.access MainThread INFO 0.001872s - "GET / HTTP/1.0" 200 "check_http/v2.2 (monitoring-plugins 2.2)" response=379b request_body=-b
Currently this is connected to kafka-test1006.eqiad.wmnet. It is my understanding we will use the "jumbo" kafka cluster. The following netcat times out; we'll need to open the firewall for traffic to kafka-jumbo.
nc -vz kafka-jumbo1001.eqiad.wmnet 9092 nc: connect to kafka-jumbo1001.eqiad.wmnet (10.64.0.175) port 9092 (tcp) timed out: Operation now in progress
Merged my latest patch to remove Type=notify (https://gerrit.wikimedia.org/r/c/operations/puppet/+/773387, I used the other karapace ticket so it didn't post here) and after manually restarting karapace, puppet competes without error.
Change 774538 had a related patch set uploaded (by Razzi; author: Razzi):
[operations/puppet@production] kafka: allow access to jumbo from karapace1001
Change 774538 merged by Razzi:
[operations/puppet@production] kafka: allow access to jumbo from karapace1001
Change 787112 had a related patch set uploaded (by Razzi; author: Razzi):
[operations/puppet@production] karapace: switch karapace to use kafka-jumbo1001
Change 787112 abandoned by Btullis:
[operations/puppet@production] karapace: switch karapace to use kafka-jumbo1001
Reason:
Superseded by: https://gerrit.wikimedia.org/r/c/operations/puppet/+/936753