- Jan 16, 2024
-
-
Vincent Sellier authored
Related to swh/infra/sysadm-environment#5212
-
Vincent Sellier authored
The moma instance had around of 1G of memory allowed. The 256m in the kubernetes environment are probably not enough Related to swh/infra/sysadm-environment#5212
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5211
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5211
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5211
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5211
-
Nicolas Dandrimont authored
-
Nicolas Dandrimont authored
This allows using it for the cassandra webapp as well
-
Nicolas Dandrimont authored
-
Vincent Sellier authored
The nginx is not notified as soon a service endpoints are updated so it sends a couple of requests to a terminated pod. It's not a problem for idempotent request as nginx retry on another pod but it's not the case by default for POST requests. The option is activated by default for all the ingresses because POST requests are used everywhere Related to swh/infra/sysadm-environment#5210
-
- Jan 15, 2024
-
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5194
-
Antoine R. Dumont authored
This matches what used to be done in the previous deposit (moma) instance. Refs. swh/infra/sysadm-environment#5207
-
Antoine R. Dumont authored
It used the storageConfigurationRef key instead of the dedicated storageMetadataConfigurationRef. Refs. swh/infra/sysadm-environment#5207
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5205
-
- Jan 12, 2024
-
-
Nicolas Dandrimont authored
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5205
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5205
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5205
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5205
-
Antoine R. Dumont authored
This also fixes an existing wrong configuration regarding the cluster network ranges. Refs. swh/infra/sysadm-environment#5205
-
Antoine R. Dumont authored
Fix the typo for fun and profit. Refs. swh/infra/sysadm-environment#5205
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5205
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5205
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5194
-
Nicolas Dandrimont authored
-
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5194
-
Antoine R. Dumont authored
To solve the remaining keycloak authentication issue in the new dynamic webapp. Refs. swh/infra/sysadm-environment#5194 Refs. swh/infra/sysadm-environment#5095 Refs. c22ffb07
-
Antoine R. Dumont authored
The reverse proxy is the one in charge of this. Refs. swh/infra/sysadm-environment#5194
-
Antoine R. Dumont authored
It's one step towards migrating the main archive webapp to the dynamic infrastructure. Refs. swh/infra/sysadm-environment#5194
-
Antoine R. Dumont authored
It's a first step towards decomissionning the gunicorn swh service from moma. Refs. swh/infra/sysadm-environment#5194
-
Antoine R. Dumont authored
Probably broken during a previous iteration with a rebase which swapped things up (& silently unnoticed). Refs. swh/infra/sysadm-environment#5194
-
Antoine R. Dumont authored
This reverts commit 6e3416e8. Refs. swh/infra/sysadm-environment#5194
-
Nicolas Dandrimont authored
The client id is only useful for debugging, but let's make it meaningful.
-
Nicolas Dandrimont authored
The journal_writer config needs to be under the indexer_storage key, rather than at the toplevel.
-
Antoine R. Dumont authored
It's the reverse proxy which has one, the ingress does not need it. Internally, we only use http connection to communicate. Refs. swh/infra/sysadm-environment#5194
-
- Jan 11, 2024
-
-
Vincent Sellier authored
-
Antoine R. Dumont authored
Refs. swh/infra/sysadm-environment#5191
-
Antoine R. Dumont authored
As a first step towards migrating the dynamic instance as the main public one. Refs. swh/infra/sysadm-environment#5191
-
Vincent Sellier authored
-