Merge pull request #1919 from kgaillot/backports

Release 1.1.22-rc1

ChangeLog

@@ -1,3 +1,41 @@
+* Tue Oct 22 2019 Ken Gaillot <kgaillot@redhat.com> Pacemaker-1.1.22
+- Changesets: 198
+- Diff:       95 files changed, 3047 insertions(+), 1245 deletions(-)
+
+- Features added since Pacemaker-1.1.21
+  + crmd: new 'fence-reaction' cluster option specifies whether local node
+          should 'stop' or 'panic' if notified of own fencing
+  + Pacemaker Remote: allow file for environment variables when used in bundle
+  + Pacemaker Remote: allow configurable listen address and TLS priorities
+  + tools: crm_simulate --repeat option to repeat profiling tests
+  + tools: new pcmk_simtimes tool to compare crm_simulate profiling output
+
+- Changes since Pacemaker-1.1.21
+  + fencing: do not block concurrent fencing actions on a device
+             (regression since 1.1.21)
+  + crmd: set timeout on scheduler responses to avoid infinite wait
+  + crmd: confirm cancel of failed monitors, to avoid transition timeout
+  + lrmd: let controller cancel monitors, to avoid transition timeout
+  + lrmd: return error for stonith probes if stonith connection was lost
+  + fencing: ensure concurrent fencing commands always get triggered to execute
+  + fencing: fail pending actions and re-sync history after crash and restart
+  + fencing: don't let command with long delay block other pending commands
+  + fencing: allow functioning even if CIB updates arrive unceasingly
+  + pengine: calculate secure digests for unfencing, for replaying saved CIBs
+  + pengine: properly detect dangling migrations, to avoid restart loop
+  + pengine: avoid delay in recovery of failed remote connections
+  + pengine: avoid scheduling actions on remote node that is shutting down
+  + pengine: wait for probe actions to complete to prevent unnecessary
+             restart/re-promote of dependent resources
+  + libcrmcommon: avoid possible use-of-NULL when applying XML diffs
+  + libcrmcommon: correctly apply XML diffs with multiple move/create changes
+  + libcrmcommon: return error when applying XML diffs containing unknown operations
+  + tools: fail if tar is not available when running crm_report
+  + tools: correct crm_report argument parsing
+  + tools: crm_report: don't ignore log if unrelated file is too large
+  + agents: calculate #health_disk correctly in SysInfo
+  + agents: handle run-as-user properly in ClusterMon
+
 * Thu Jun 6 2019 Ken Gaillot <kgaillot@redhat.com> Pacemaker-1.1.21
 - Changesets: 64
 - Diff:       39 files changed, 1985 insertions(+), 843 deletions(-)

cts/CTStests.py

@@ -516,7 +516,7 @@ def errorstoignore(self):
             self.templates["Pat:Fencing_start"] % ".*",
             self.templates["Pat:Fencing_ok"] % ".*",
             r"error.*: Resource .*stonith::.* is active on 2 nodes attempting recovery",
-            r"error.*: Operation reboot of .*by .* for stonith_admin.*: Timer expired",
+            r"error.*: Operation 'reboot' targeting .* on .* for stonith_admin.*: Timer expired",
         ]
 
     def is_applicable(self):

cts/patterns.py

@@ -59,8 +59,8 @@ def __init__(self, name):
             "Pat:They_dead"     : "node %s.*: is dead",
             "Pat:TransitionComplete" : "Transition status: Complete: complete",
 
-            "Pat:Fencing_start"   : r"(Initiating remote operation|Requesting peer fencing ).* (for|of) %s",
-            "Pat:Fencing_ok"      : r"stonith.*:\s*Operation .* of %s by .* for .*@.*: OK",
+            "Pat:Fencing_start"   : r"Requesting peer fencing .* targeting %s",
+            "Pat:Fencing_ok"      : r"stonith.*:\s*Operation .* targeting %s on .* for .*@.*: OK",
             "Pat:Fencing_recover" : r"pengine.*: Recover %s",
             "Pat:Fencing_active"  : r"pengine.*: Resource %s is active on .* nodes",
             "Pat:Fencing_probe"   : r"crmd.*: Result of probe operation for %s on .*: Error",

fencing/regression.py.in

@@ -5,7 +5,7 @@
 # Pacemaker targets compatibility with Python 2.6+ and 3.2+
 from __future__ import print_function, unicode_literals, absolute_import, division
 
-__copyright__ = "Copyright (C) 2012-2016 Andrew Beekhof <andrew@beekhof.net>"
+__copyright__ = "Copyright 2012-2019 the Pacemaker project contributors"
 __license__ = "GNU General Public License version 2 or later (GPLv2+) WITHOUT ANY WARRANTY"
 
 import io
@@ -485,8 +485,8 @@ class Tests(object):
         ### one merger will happen
         test.add_stonith_log_pattern("Merging stonith action 'off' targeting node3 originating from client")
         ### the pattern below signifies that both the original and duplicate operation completed
-        test.add_stonith_log_pattern("Operation off of node3 by")
-        test.add_stonith_log_pattern("Operation off of node3 by")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
 
         ### Test that multiple mergers occur
         test = self.new_test("cpg_custom_merge_multiple",
@@ -506,11 +506,11 @@ class Tests(object):
         test.add_stonith_log_pattern("Merging stonith action 'off' targeting node3 originating from client")
         test.add_stonith_log_pattern("Merging stonith action 'off' targeting node3 originating from client")
         ### the pattern below signifies that both the original and duplicate operation completed
-        test.add_stonith_log_pattern("Operation off of node3 by")
-        test.add_stonith_log_pattern("Operation off of node3 by")
-        test.add_stonith_log_pattern("Operation off of node3 by")
-        test.add_stonith_log_pattern("Operation off of node3 by")
-        test.add_stonith_log_pattern("Operation off of node3 by")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
 
         ### Test that multiple mergers occur with topologies used
         test = self.new_test("cpg_custom_merge_with_topology",
@@ -533,11 +533,11 @@ class Tests(object):
         test.add_stonith_log_pattern("Merging stonith action 'off' targeting node3 originating from client")
         test.add_stonith_log_pattern("Merging stonith action 'off' targeting node3 originating from client")
         ### the pattern below signifies that both the original and duplicate operation completed
-        test.add_stonith_log_pattern("Operation off of node3 by")
-        test.add_stonith_log_pattern("Operation off of node3 by")
-        test.add_stonith_log_pattern("Operation off of node3 by")
-        test.add_stonith_log_pattern("Operation off of node3 by")
-        test.add_stonith_log_pattern("Operation off of node3 by")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
+        test.add_stonith_log_pattern("Operation 'off' targeting node3 on")
 
     def build_fence_no_merge_tests(self):
         """ Register tests to verify when fence operations should not be merged """
@@ -1130,16 +1130,16 @@ class Tests(object):
                      """-o "pcmk_off_timeout=2" -o "pcmk_reboot_timeout=20" """)
         test.add_cmd("stonith_admin", "-r node_fake -i 1 -v true1 -v true2")
         test.add_cmd("stonith_admin", "-B node_fake -t 5")
-        test.add_stonith_log_pattern("Remapping multiple-device reboot of node_fake")
+        test.add_stonith_log_pattern("Remapping multiple-device reboot targeting node_fake")
         # timeout should be sum of off timeouts (1+2=3), not reboot timeouts (10+20=30)
-        test.add_stonith_log_pattern("Total timeout set to 3 for peer's fencing of node_fake")
-        test.add_stonith_log_pattern("perform op 'node_fake off' with 'true1'")
-        test.add_stonith_log_pattern("perform op 'node_fake off' with 'true2'")
-        test.add_stonith_log_pattern("Remapped off of node_fake complete, remapping to on")
+        test.add_stonith_log_pattern("Total timeout set to 3 for peer's fencing targeting node_fake")
+        test.add_stonith_log_pattern("perform 'off' action targeting node_fake using 'true1'")
+        test.add_stonith_log_pattern("perform 'off' action targeting node_fake using 'true2'")
+        test.add_stonith_log_pattern("Remapped 'off' targeting node_fake complete, remapping to 'on'")
         # fence_dummy sets "on" as an on_target action
-        test.add_stonith_log_pattern("Ignoring true1 'on' failure (no capable peers) for node_fake")
-        test.add_stonith_log_pattern("Ignoring true2 'on' failure (no capable peers) for node_fake")
-        test.add_stonith_log_pattern("Undoing remap of reboot of node_fake")
+        test.add_stonith_log_pattern("Ignoring true1 'on' failure (no capable peers) targeting node_fake")
+        test.add_stonith_log_pattern("Ignoring true2 'on' failure (no capable peers) targeting node_fake")
+        test.add_stonith_log_pattern("Undoing remap of reboot targeting node_fake")
 
         test = self.new_test("cpg_remap_automatic",
                              "Verify remapped topology reboot skips automatic 'on'", 1)
@@ -1151,12 +1151,12 @@ class Tests(object):
                      """-o "mode=pass" -o "pcmk_host_list=node_fake" """)
         test.add_cmd("stonith_admin", "-r node_fake -i 1 -v true1 -v true2")
         test.add_cmd("stonith_admin", "-B node_fake -t 5")
-        test.add_stonith_log_pattern("Remapping multiple-device reboot of node_fake")
-        test.add_stonith_log_pattern("perform op 'node_fake off' with 'true1'")
-        test.add_stonith_log_pattern("perform op 'node_fake off' with 'true2'")
-        test.add_stonith_log_pattern("Remapped off of node_fake complete, remapping to on")
-        test.add_stonith_log_pattern("Undoing remap of reboot of node_fake")
-        test.add_stonith_neg_log_pattern("perform op 'node_fake on' with")
+        test.add_stonith_log_pattern("Remapping multiple-device reboot targeting node_fake")
+        test.add_stonith_log_pattern("perform 'off' action targeting node_fake using 'true1'")
+        test.add_stonith_log_pattern("perform 'off' action targeting node_fake using 'true2'")
+        test.add_stonith_log_pattern("Remapped 'off' targeting node_fake complete, remapping to 'on'")
+        test.add_stonith_log_pattern("Undoing remap of reboot targeting node_fake")
+        test.add_stonith_neg_log_pattern("perform 'on' action targeting node_fake using")
         test.add_stonith_neg_log_pattern("'on' failure")
 
         test = self.new_test("cpg_remap_complex_1",
@@ -1168,14 +1168,14 @@ class Tests(object):
         test.add_cmd("stonith_admin", "-r node_fake -i 1 -v false1")
         test.add_cmd("stonith_admin", "-r node_fake -i 2 -v true1 -v true2")
         test.add_cmd("stonith_admin", "-B node_fake -t 5")
-        test.add_stonith_log_pattern("perform op 'node_fake reboot' with 'false1'")
-        test.add_stonith_log_pattern("Remapping multiple-device reboot of node_fake")
-        test.add_stonith_log_pattern("perform op 'node_fake off' with 'true1'")
-        test.add_stonith_log_pattern("perform op 'node_fake off' with 'true2'")
-        test.add_stonith_log_pattern("Remapped off of node_fake complete, remapping to on")
-        test.add_stonith_log_pattern("Ignoring true1 'on' failure (no capable peers) for node_fake")
-        test.add_stonith_log_pattern("Ignoring true2 'on' failure (no capable peers) for node_fake")
-        test.add_stonith_log_pattern("Undoing remap of reboot of node_fake")
+        test.add_stonith_log_pattern("perform 'reboot' action targeting node_fake using 'false1'")
+        test.add_stonith_log_pattern("Remapping multiple-device reboot targeting node_fake")
+        test.add_stonith_log_pattern("perform 'off' action targeting node_fake using 'true1'")
+        test.add_stonith_log_pattern("perform 'off' action targeting node_fake using 'true2'")
+        test.add_stonith_log_pattern("Remapped 'off' targeting node_fake complete, remapping to 'on'")
+        test.add_stonith_log_pattern("Ignoring true1 'on' failure (no capable peers) targeting node_fake")
+        test.add_stonith_log_pattern("Ignoring true2 'on' failure (no capable peers) targeting node_fake")
+        test.add_stonith_log_pattern("Undoing remap of reboot targeting node_fake")
 
         test = self.new_test("cpg_remap_complex_2",
                              "Verify remapped topology reboot failure in second level proceeds to third level",
@@ -1189,13 +1189,13 @@ class Tests(object):
         test.add_cmd("stonith_admin", "-r node_fake -i 2 -v true1 -v false2 -v true3")
         test.add_cmd("stonith_admin", "-r node_fake -i 3 -v true2")
         test.add_cmd("stonith_admin", "-B node_fake -t 5")
-        test.add_stonith_log_pattern("perform op 'node_fake reboot' with 'false1'")
-        test.add_stonith_log_pattern("Remapping multiple-device reboot of node_fake")
-        test.add_stonith_log_pattern("perform op 'node_fake off' with 'true1'")
-        test.add_stonith_log_pattern("perform op 'node_fake off' with 'false2'")
+        test.add_stonith_log_pattern("perform 'reboot' action targeting node_fake using 'false1'")
+        test.add_stonith_log_pattern("Remapping multiple-device reboot targeting node_fake")
+        test.add_stonith_log_pattern("perform 'off' action targeting node_fake using 'true1'")
+        test.add_stonith_log_pattern("perform 'off' action targeting node_fake using 'false2'")
         test.add_stonith_log_pattern("Attempted to execute agent fence_dummy (off) the maximum number of times")
-        test.add_stonith_log_pattern("Undoing remap of reboot of node_fake")
-        test.add_stonith_log_pattern("perform op 'node_fake reboot' with 'true2'")
+        test.add_stonith_log_pattern("Undoing remap of reboot targeting node_fake")
+        test.add_stonith_log_pattern("perform 'reboot' action targeting node_fake using 'true2'")
         test.add_stonith_neg_log_pattern("node_fake with true3")
 
     def setup_environment(self, use_corosync):