Unexpected rolling over indices in RHOCP 4

Issue

• New empty indices (audit, app, infra) are created without fulfilling the index management cronjob roll-over conditions.

  $ oc exec -n openshift-logging -c elasticsearch $es_pod -- es_util --query=_cat/indices?v
  health status index        uuid                   pri rep docs.count docs.deleted store.size pri.store.size
  green  open   app-000001   5oA5aoOzQuuTZod6AyP6cw   1   0        119            0    216.3kb        216.3kb
  green  open   .security    PP8U8rGTTV-H4IJUsltsmg   1   0          6            0     33.2kb         33.2kb
  green  open   infra-000001 mdiyyrYCQUmCuUWF4f_MCA   1   0          0            0       261b           261b
  green  open   .kibana_1    m1LJ3pWFQxeW56bhdCCzGQ   1   0          0            0       261b           261b
  green  open   app-000002   r1qc1skYSPGlP-noNiPjqA   1   0          0            0       261b           261b  -->new empty index
  green  open   audit-000002 EbUGAFv5TMKqRhu2GY8C-Q   1   0          0            0       261b           261b  -->new empty index 
  green  open   infra-000002 JBvjHyTkSse0C9alXEsmgA   1   0          0            0       261b           261b  -->new empty index 
  green  open   audit-000001 TENxomASTPWZT3-UCbhsTg   1   0          0            0       261b           261b