Puppet should generate /etc/wdqs/vars.yaml which on equiad would (temporary) be:
--- data_dir:
and on codfw ones it would be:
--- data_dir: /srv/wdqs/
When/if eqiad servers get repartitioned to have the same partition scheme as codfw ones, the codfw config should be applied to them too. The puppet code should be made flexible so we could transfer servers to new scheme one by one (i.e., we should be able to tell server use one scheme or another).
| Subject | Repo | Branch | Lines +/- | |
|---|---|---|---|---|
| wdqs - move data to /srv | operations/puppet | production | +19 -10 |
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | Gehel | T124627 Adjust balance of WDQS nodes to allow continued operation if eqiad went offline. | |||
| Resolved | Gehel | T124862 Deploy WDQS nodes on codfw | |||
| Resolved | Gehel | T144380 Install and configure new WDQS nodes on codfw | |||
| Resolved | Gehel | T144536 Move data storage to /srv/wdqs/ on codfw WDQS nodes | |||
| Resolved | Gehel | T144537 Make puppet generate path config for WDQS nodes |