Commit Graph

676 Commits

Author SHA1 Message Date
Jakub Sokołowski 6488efd4e9
nimbus-db-purge: deploy to two mainnet hosts
https://github.com/status-im/infra-nimbus/issues/86

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-28 17:32:01 +01:00
Jakub Sokołowski a8b48242b0
nimbus-db-purge: add role to periodicall purge DB
https://github.com/status-im/infra-nimbus/issues/86

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-28 17:31:31 +01:00
Jakub Sokołowski 01e129694b
dummy-module: don't set become_user for inventory
Causes `become_user` specified in tasks to be ignored.

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-28 16:51:16 +01:00
Jakub Sokołowski c4436f71f4
drop windows-01.gc-us-central1-a.nimbus.prater
Replaced by `windows-01.he-eu-hel1.nimbus.prater` due to disk space
constraints and high cost of Google Cloud host.

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-28 13:14:04 +01:00
Jakub Sokołowski 9b365b02a7
add new windows-01.he-eu-hel1.nimbus.prater host
To replace `windows-01.gc-us-central1-a.nimbus.prater` which is running
out of disk space, and is also far too expensive.

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-25 15:02:27 +01:00
Jakub Sokołowski 489878038e
requirements: bump bootstrap roles
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-24 01:40:03 +01:00
Jakub Sokołowski c0ab15c34f
mainnet.openeth: deploy OpenEthereum node to the host
We need some variety and redundancy in how we sync Eth1:
https://github.com/status-im/infra-nimbus/issues/80

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-21 17:26:33 +01:00
Jakub Sokołowski 3acbf34f70
add new mainnet-01.he-eu-hel1.nimbus.openeth host
https://github.com/status-im/infra-nimbus/issues/80

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-21 13:54:20 +01:00
Jakub Sokołowski 54c847b9db
nimbus.geth: bump Goerli data volume to 150 GB
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-18 19:37:25 +01:00
Jakub Sokołowski aebe67a9d8
nimbus.prater: add 100 GB data volume for AWS hosts
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-17 21:19:04 +01:00
Jakub Sokołowski 9f52f8701f
nimbus.mainnet: bump size of AWS data volumes
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-17 20:47:57 +01:00
Jakub Sokołowski 6940d0219c
drop libp2p node from macos host to save space
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-17 19:53:50 +01:00
Jakub Sokołowski 65304b6aaa new validator layout for prater fleet
The `0/1/10/30/60/899/4000` layout will require 3 more hosts.

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-17 19:41:07 +01:00
Jakub Sokołowski 3d4679f0d3
nimbus.prater: add nim_compiler overrides
For some reason `nim-1.6` branch doesn't actually use Nim 1.6 compiler.

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-17 14:24:37 +01:00
Jakub Sokołowski 01c6a06117
nimbus.prater: override some branches to nim-1.6
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-17 13:46:56 +01:00
Jakub Sokołowski d6414fd956
deploy nodes without validators to 3 new prater hosts
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-16 17:19:23 +01:00
Jakub Sokołowski 612eb284c6
add 3 new nimbus.prater hosts
Necessary for new layout from:
https://github.com/status-im/infra-nimbus/pull/85

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-16 17:09:51 +01:00
Jakub Sokołowski 84d2773a98
deploy nodes without validators to new prater hosts
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-15 21:09:51 +01:00
Jakub Sokołowski ec65bac02a
nimbus.pyrmont: drop two nodes and move validators
Two nodes will be reused in the Prater fleet.

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-15 21:08:44 +01:00
Jakub Sokołowski 18eb46836e
logs.nimbus: use docker compose setup for elasticsearch
https://github.com/status-im/infra-role-elasticsearch/commit/567f32ac

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-11 02:08:47 +01:00
Jakub Sokołowski b12a350cb9
bitwarden: include fixes for session invalidation
For more details see:
- https://github.com/status-im/infra-misc/issues/90
- https://github.com/c0sco/ansible-modules-bitwarden/issues/23

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-10 15:34:18 +01:00
Jakub Sokołowski 4e7625beee
requirements: bump wireguard role
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-10 14:50:05 +01:00
Jakub Sokołowski 392d4124c5
dash.nimbus: bump limit of log indices back to 20
Currently daily indices take up 10 GB per host, so it' fine.

https://github.com/status-im/infra-nimbus/issues/81

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-10 12:31:03 +01:00
Jakub Sokołowski 0398c31ad9
logs.nimbus: drop ES replicas down to 0 to save space
https://github.com/status-im/infra-nimbus/issues/81

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-09 18:43:23 +01:00
Jakub Sokołowski fbb1ee69d9
consul: update token paths in BitWarden
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-09 00:24:07 +01:00
Jakub Sokołowski d74ff169ef
wireguard: bump role and add Consul ACL token
Part of effort to lock down Consul ACLs:
https://github.com/status-im/infra-hq/issues/70

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-08 20:51:45 +01:00
Jakub Sokołowski cdf63cba43
consul: add mandatory agent ACL token
https://github.com/status-im/infra-hq/issues/70

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-08 11:58:55 +01:00
Jakub Sokołowski fb7bde2377
dash.nimbus: lower indices kept to 10
We have bigger log volume now that we fixed the logging setup in:
https://github.com/status-im/infra-nimbus/issues/81

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-07 15:07:55 +01:00
Jakub Sokołowski f665fcd8ce
rename logs-dash groups_vars to dash.nimbus
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-07 15:03:12 +01:00
Jakub Sokołowski ba2caedb64
logs.nimbus: bump JVM memory limits to 80%
We started getting errors like:
```
Caused by: org.elasticsearch.common.breaker.CircuitBreakingException: [parent] Data too large, data for [internal:cluster/nodes/indices/shard/store[n]] would be [36997638608/34.4gb], which is larger than the limit of [36795685273/34.2gb], real usage: [36997638456/34.4gb], new bytes reserved: [152/152b], usages [request=0/0b, fielddata=0/0b, in_flight_requests=152/152b, model_inference=0/0b, eql_sequence=0/0b, accounting=3913236/3.7mb]
```

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-07 11:49:44 +01:00
Jakub Sokołowski ade0a50a53
logs.nimbus: migrate to Hetzner metal hosts
For better performance, lower prices, and better network utilization.

https://github.com/status-im/infra-nimbus/issues/84

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-03 21:48:24 +01:00
Jakub Sokołowski 761b0d7e2b
terraform: inherit aws provider version from module
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-03 20:32:29 +01:00
Jakub Sokołowski 36a838e6ae
logs: rename tf resources before adding new hosts
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-03 20:32:28 +01:00
Jakub Sokołowski aed600506d
requirements: deploy consul listen address fix
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-03 19:06:52 +01:00
Jakub Sokołowski b3da28aeb8
requirements: bump bootstrap to fix journald config
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-02-01 12:58:13 +01:00
Jakub Sokołowski a6d673366a
all: lowe logrotate count to 48, bump bootstrap
Also remove `delaycompress` from logrotate config.

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-31 17:09:28 +01:00
Jakub Sokołowski 4983b11282
requirements: bump bootstrap and beacon node roles
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-28 13:41:02 +01:00
Jakub Sokołowski 0dc470aae2
requirements: add support for ansible toggle script
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-26 11:00:20 +01:00
Jakub Sokołowski 2962ebbee0
disable use of HTTP Eth1 node endpoints
Should fix issues with leaking file descriptors causing peer issues and
metrics endpoint timeouts.

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-26 01:22:16 +01:00
Jakub Sokołowski 0369818fa0
replace macos-01.ms-eu-dublin.nimbus.prater host
The old one had only 256 GB of SSD storage which is not enough for
three nodes syncing Prater testnet.

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-24 15:44:38 +01:00
Jakub Sokołowski 9b54c84d96
enable validator monitoring flags for unstable nodes
Once the next release is out we can drop the `unstable` conditionals.

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-21 15:36:53 +01:00
Jakub Sokołowski 4ff2422fe2
nimbnus.prater: use the AWS NVMe drive for data
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-19 11:26:16 +01:00
Jakub Sokołowski e00443d6e2
nimbus.mainnet: drop libp2p-02 nodes to save space
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-13 14:16:41 +01:00
Jakub Sokołowski 79f55366f2
requirements: upgrade Geth to 1.10.15
https://github.com/ethereum/go-ethereum/releases/tag/v1.10.15

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-07 13:05:58 +01:00
Jakub Sokołowski 667fd3184e
builds: add libp2p metrics flags
This is done instead of this PR:
https://github.com/status-im/infra-role-beacon-node-linux/pull/3/files

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-05 19:38:23 +01:00
Jakub Sokołowski f106c9e5b8
requirements: parametrize nim build flags
https://github.com/status-im/infra-role-beacon-node-linux/commit/d993d06d
https://github.com/status-im/infra-role-beacon-node-macos/commit/ed0453ef
https://github.com/status-im/infra-role-beacon-node-windows/commit/f6dd9dd6

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-05 19:38:23 +01:00
Jakub Sokołowski 237da365e7
bootstrap: add missing admin pass for macos hosts
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-04 20:55:08 +01:00
Jakub Sokołowski 9c8b0f6185
users: give tanguy sudo permissions
Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-04 15:35:30 +01:00
Jakub Sokołowski d99b195bc1
consul: upgrade from 1.10.1 to 1.11.1
https://github.com/hashicorp/consul/releases/tag/v1.11.1

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2022-01-03 17:53:34 +01:00
Jakub Sokołowski 462f373827
mainnet.geth: use snap sync method
We had issues with mainnet node not fully syncing.

Signed-off-by: Jakub Sokołowski <jakub@status.im>
2021-12-18 18:06:59 +01:00