Abandoning in case of server auth errors

Details: * When the HMC userid cannot log on to the HMC because the password is invalid or expired, or because the maximum number of sessions has been reached, the exporter no longer retries the logon but abandons. The previous retrying has lead to disabling the userid after some unsuccessful retries. (issue #493) Signed-off-by: Andreas Maier <[email protected]>
zhmcclient · Apr 2, 2024 · ad77792 · ad77792
1 parent 32b28d5
commit ad77792
Show file tree

Hide file tree

Showing 2 changed files with 16 additions and 7 deletions.
diff --git a/docs/changes.rst b/docs/changes.rst
@@ -139,6 +139,11 @@ will also work with the prior version of the file (but not vice versa).
 * The safety run for all dependencies now must succeed when the test workflow
   is run for a release (i.e. branch name 'release_...').
 
+* When the HMC userid cannot log on to the HMC because the password is invalid
+  or expired, or because the maximum number of sessions has been reached, the
+  exporter no longer retries the logon but abandons. The previous retrying has
+  lead to disabling the userid after some unsuccessful retries. (issue #493)
+
 **Cleanup:**
 
 * Increased versions of GitHub Actions plugins to increase node.js runtime

diff --git a/zhmc_prometheus_exporter/zhmc_prometheus_exporter.py b/zhmc_prometheus_exporter/zhmc_prometheus_exporter.py
@@ -1687,6 +1687,13 @@ def collect(self):
                 try:
                     metrics_object = retrieve_metrics(self.context)
                 except zhmcclient.HTTPError as exc:
+                    if exc.http_status == 400 and exc.reason in (13, 45):
+                        # 400.13: Logon: Max sessions reached for user
+                        # 400.45: Logon: Password expired
+                        logprint(logging.ERROR, PRINT_ALWAYS,
+                                 "Abandoning after HTTP status {}.{}: {}".
+                                 format(exc.http_status, exc.reason, exc))
+                        raise
                     if exc.http_status == 404 and exc.reason == 1:
                         logprint(logging.WARNING, PRINT_ALWAYS,
                                  "Recreating the metrics context after HTTP "
@@ -1708,13 +1715,10 @@ def collect(self):
                     time.sleep(RETRY_SLEEP_TIME)
                     continue
                 except zhmcclient.ServerAuthError as exc:
-                    http_exc = exc.details  # zhmcclient.HTTPError
-                    logprint(logging.WARNING, PRINT_ALWAYS,
-                             "Retrying after server authentication error with "
-                             "HTTP status {}.{}".
-                             format(http_exc.http_status, http_exc.reason))
-                    time.sleep(RETRY_SLEEP_TIME)
-                    continue
+                    logprint(logging.ERROR, PRINT_ALWAYS,
+                             "Abandoning after server authentication error: {}".
+                             format(exc))
+                    raise
                 except zhmcclient.ClientAuthError as exc:
                     logprint(logging.ERROR, PRINT_ALWAYS,
                              "Abandoning after client authentication error: {}".