qa/suites/rados/mgr: ignorelist MGR_INFLUX_DB_LIST_FAILED in module_selftest

module_selftest fails on a cluster log warning that is not ignorelisted:

  cluster [WRN] Health check failed: Failed to list/create InfluxDB database (MGR_INFLUX_DB_LIST_FAILED)

module_selftest is the only job that enables the influx module, and no suite
provisions an InfluxDB server, so once enabled the module cannot reach a backend
and raises a health warning: MGR_INFLUX_NO_SERVER when no hostname is set,
MGR_INFLUX_DB_LIST_FAILED from serve() once one is.  test_influx sets a hostname,
so it hits the latter.

This surfaced only after the test image started shipping the influxdb python
module.  Until then can_run() returned false, the module never enabled, and the
sole log line was "influxdb python module not found", which is already
ignorelisted.  self_test() cannot run without enabling the module, so ignorelist
the warning it raises.

Signed-off-by: Kefu Chai <k.chai@proxmox.com>

diff --git a/qa/suites/rados/mgr/tasks/4-units/module_selftest.yaml b/qa/suites/rados/mgr/tasks/4-units/module_selftest.yaml
index e2a2ca03cc905f4cf6dda371d1b92748638f0f71..a40755129442329e57f42d516f23b15979b50699 100644 (file)
--- a/qa/suites/rados/mgr/tasks/4-units/module_selftest.yaml
+++ b/qa/suites/rados/mgr/tasks/4-units/module_selftest.yaml
@@ -6,6 +6,7 @@ overrides:
       - objects misplaced
       - Synthetic exception in serve
       - influxdb python module not found
+      - Failed to list/create InfluxDB database \(MGR_INFLUX_DB_LIST_FAILED\)
       - foo bar
       - Failed to open Telegraf
       - evicting unresponsive client