Bug 743214 - Cluster problem: "hard" service dependencies
Summary: Cluster problem: "hard" service dependencies
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Enterprise Linux 5
Classification: Red Hat
Component: rgmanager
Version: 5.7
Hardware: x86_64
OS: Linux
high
high
Target Milestone: rc
: 5.8
Assignee: Lon Hohberger
QA Contact: Cluster QE
URL:
Whiteboard:
: 679753 (view as bug list)
Depends On:
Blocks: 743218 772956
TreeView+ depends on / blocked
 
Reported: 2011-10-04 09:24 UTC by Frank Danapfel
Modified: 2012-02-21 06:19 UTC (History)
6 users (show)

Fixed In Version: rgmanager-2.0.52-27.el5
Doc Type: Bug Fix
Doc Text:
Cause: A "stopped" event for a given service was being processed after the a service's dependent service had already been restarted. Consequence: The dependent service would erroneously restart. Fix: Ignore the "stopped" event if both services are running and the dependent service was started after the dependency. Result: The "stopped" event no longer causes the dependent service to restart.
Clone Of:
: 743218 (view as bug list)
Environment:
Last Closed: 2012-02-21 06:19:28 UTC
Target Upstream Version:
Embargoed:

Attachments (Terms of Use)
cluster.conf (5.91 KB, text/plain)
2011-10-04 09:27 UTC, Frank Danapfel 		no flags Details
syslog entries during test showing the duplicate start of dependent service (10.98 KB, text/plain)
2011-10-04 09:30 UTC, Frank Danapfel 		no flags Details
View All


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2012:0163 0 normal SHIPPED_LIVE rgmanager bug fix and enhancement update 2012-02-20 15:07:03 UTC

Description Frank Danapfel 2011-10-04 09:24:46 UTC 
Description of problem:
Two Cluster Services are defined. The first service "sapgt1ci" starts th
e database and the central SAP instance.
 The second cluster service "sapgt1ap1" starts another SAP application instance.
 "sapgt1ap1" has a hard dependency on sapgt1ci.



While executing the cluster tests / failover tests we encountered a bug:



Setup: Cluster Server lxcn1gt1 running no services and cluster server

lxcn2gt1 running both sapgt1ci and sapgt1ap1.



Cluster node lxcn2gt1 (the one running everything) gets switched off

(power off).
 Now both services get migrated to the remaining node. The start sequence
 is correct and matches the configured dependency. But after both service
s
 were successfully started sapgt1ap1 gets stopped again (!) while

rgmanager prints a message about "lost dependency" into the syslog. Afte
r

sapgt1ap1 was successfully stopped it gets started again. This time

everything stays up.



The re-start of sapgt1ap1 should not occur.


with the services swapped in /etc/cluster/cluster.conf (now sapgt1ap1 is

defined first and the sapgtci second) we see a slightly different behavi
or:



If both services run on the same cluster node (in our test case

lxcn2gt1-hb) and this node gets powered off....:



The remaining node (lxcn1gt1-hb) first tries to start sapgt1ap1

(completely ignoring the hard dependency on sapgt1ci!!) - this start

attempt fails without the central instance running.

Then it starts sapgt1ci (SAP DB and central instance) and then it starts sapgt1ap1 again (which works this time).



To recapture the initial problem with the service definition of sapgt1ci

first in /etc/cluster/cluster.conf and sapgtap1 second:



During the failover of both services the remaining node

- first starts sapgt1ci

- then starts sapgt1ap1

- then stops sapgt1ap1 (message about missing dependency)

- then starts sapgt1ap1 again



Both start sequences are flawed.....

Nevertheless all instances were started in both cases (with an extra ste
p

in between).



Version-Release number of selected component (if applicable):
rgmanager-2.0.52-21.el5.x86_64              

How reproducible:
always
Comment 1 Frank Danapfel 2011-10-04 09:27:42 UTC 
Created attachment 526209 [details]
cluster.conf

cluster.conf from customer
Comment 2 Frank Danapfel 2011-10-04 09:30:07 UTC 
Created attachment 526212 [details]
syslog entries during test showing the duplicate start of dependent service
Comment 4 Lon Hohberger 2012-01-05 05:17:26 UTC 
Here, b depends on a -- both were on member 1, and when it failed:

Jan  5 00:14:31 rhel5-2 clurgmgrd[3023]: <info> Marking service service:a on down member 1 as stopped
Jan  5 00:14:31 rhel5-2 clurgmgrd[3023]: <notice> Starting service:a on [ 2 ]
Jan  5 00:14:31 rhel5-2 clurgmgrd[3023]: <debug> Sent remote-start request to 2
Jan  5 00:14:31 rhel5-2 clurgmgrd[3023]: <notice> Starting stopped service service:a
Jan  5 00:14:32 rhel5-2 clurgmgrd[3023]: <notice> Service service:a is now running on member 2 
Jan  5 00:14:32 rhel5-2 clurgmgrd[3023]: <debug> Executing sortedservices node event handler for service: service:b 
Jan  5 00:14:32 rhel5-2 clurgmgrd[3023]: <debug> Evaluating service:b state=started owner=1 
Jan  5 00:14:32 rhel5-2 clurgmgrd[3023]: <info> Marking service service:b on down member 1 as stopped 
Jan  5 00:14:32 rhel5-2 clurgmgrd[3023]: <notice> Starting service:b on [ 2 ] 
Jan  5 00:14:32 rhel5-2 clurgmgrd[3023]: <debug> Sent remote-start request to 2 
Jan  5 00:14:32 rhel5-2 clurgmgrd[3023]: <notice> Starting stopped service service:b 
Jan  5 00:14:33 rhel5-2 clurgmgrd[3023]: <notice> Service service:b started 
Jan  5 00:14:33 rhel5-2 clurgmgrd[3023]: <notice> Service service:b is now running on member 2 
Jan  5 00:14:33 rhel5-2 clurgmgrd[3023]: <debug> Executing default service event handler 
Jan  5 00:14:33 rhel5-2 clurgmgrd[3023]: <debug> Skipping service:b; restart not needed 
Jan  5 00:14:33 rhel5-2 clurgmgrd[3023]: <debug> Executing default service event handler 
Jan  5 00:14:33 rhel5-2 last message repeated 2 times
Jan  5 00:14:33 rhel5-2 clurgmgrd[3023]: <debug> 5 events processed 

The important line is the 'Skipping service:b' debug information - prior to the patches here, this is where service:b would have been restarted.
Comment 5 Lon Hohberger 2012-01-05 05:26:48 UTC 
https://www.redhat.com/archives/cluster-devel/2012-January/msg00006.html
https://www.redhat.com/archives/cluster-devel/2012-January/msg00007.html
Comment 7 Lon Hohberger 2012-01-05 15:32:28 UTC 
Some bugs with the patches, so I am submitting new ones.

Test report (after uncommenting debug lines in default_event_script.sl) with fixed version for:

* config #1:

   <rm central_processing="1" log_level="7" ... >
     <service "a" />
     <service "b" depend="service:a" />
   </rm>

* failure case #1
  a and b on node 1, kill node 1.
  Expected results, move A over, move B over, B does not cycle:

Jan  5 10:21:24 rhel5-2 clurgmgrd[4759]: <debug> Membership Change Event 
Jan  5 10:21:24 rhel5-2 clurgmgrd[4759]: <info> State change: rhel5-1.lhh.pvt DOWN 
Jan  5 10:21:24 rhel5-2 clurgmgrd[4759]: <debug> sorted_service_list service:a, service:b 
Jan  5 10:21:24 rhel5-2 clurgmgrd[4759]: <debug> Executing sortedservices node event handler for service: service:a 
Jan  5 10:21:24 rhel5-2 clurgmgrd[4759]: <debug> Evaluating service:a state=started owner=1 
Jan  5 10:21:25 rhel5-2 clurgmgrd[4759]: <info> Marking service service:a on down member 1 as stopped 
Jan  5 10:21:25 rhel5-2 clurgmgrd[4759]: <notice> Starting service:a on [ 2 ] 
Jan  5 10:21:25 rhel5-2 clurgmgrd[4759]: <debug> Sent remote-start request to 2 
Jan  5 10:21:25 rhel5-2 clurgmgrd[4759]: <notice> Starting stopped service service:a 
Jan  5 10:21:25 rhel5-2 clurgmgrd[4759]: <notice> Service service:a started 
Jan  5 10:21:25 rhel5-2 clurgmgrd[4759]: <notice> Service service:a is now running on member 2 
Jan  5 10:21:26 rhel5-2 clurgmgrd[4759]: <debug> Executing sortedservices node event handler for service: service:b 
Jan  5 10:21:26 rhel5-2 clurgmgrd[4759]: <debug> Evaluating service:b state=started owner=1 
Jan  5 10:21:26 rhel5-2 clurgmgrd[4759]: <info> Marking service service:b on down member 1 as stopped 
Jan  5 10:21:26 rhel5-2 clurgmgrd[4759]: <notice> Starting service:b on [ 2 ] 
Jan  5 10:21:26 rhel5-2 clurgmgrd[4759]: <debug> Sent remote-start request to 2 
Jan  5 10:21:26 rhel5-2 clurgmgrd[4759]: <notice> Starting stopped service service:b 
Jan  5 10:21:26 rhel5-2 clurgmgrd[4759]: <notice> Service service:b started 
Jan  5 10:21:26 rhel5-2 clurgmgrd[4759]: <notice> Service service:b is now running on member 2 
Jan  5 10:21:26 rhel5-2 clurgmgrd[4759]: <debug> Executing default service event handler 
Jan  5 10:21:26 rhel5-2 clurgmgrd[4759]: <debug> S:service:a trans 1325776885 
Jan  5 10:21:26 rhel5-2 clurgmgrd[4759]: <debug> D:service:b trans 1325776886 
Jan  5 10:21:27 rhel5-2 clurgmgrd[4759]: <debug> Skipping service:b; restart not needed 
Jan  5 10:21:27 rhel5-2 clurgmgrd[4759]: <debug> Executing default service event handler 
Jan  5 10:21:27 rhel5-2 clurgmgrd[4759]: <debug> 5 events processed 

  Result: PASS


* failure case #2
  a on node 1, b on node 2, kill node 1.
  Expected results, move a over, b restarts:

Jan  5 10:28:27 rhel5-2 clurgmgrd[4759]: <debug> Membership Change Event 
Jan  5 10:28:27 rhel5-2 clurgmgrd[4759]: <info> State change: rhel5-1.lhh.pvt DOWN 
Jan  5 10:28:27 rhel5-2 clurgmgrd[4759]: <debug> sorted_service_list service:a, service:b 
Jan  5 10:28:27 rhel5-2 clurgmgrd[4759]: <debug> Executing sortedservices node event handler for service: service:a 
Jan  5 10:28:27 rhel5-2 clurgmgrd[4759]: <debug> Evaluating service:a state=started owner=1 
Jan  5 10:28:27 rhel5-2 clurgmgrd[4759]: <info> Marking service service:a on down member 1 as stopped 
Jan  5 10:28:28 rhel5-2 clurgmgrd[4759]: <notice> Starting service:a on [ 2 ] 
Jan  5 10:28:28 rhel5-2 clurgmgrd[4759]: <debug> Sent remote-start request to 2 
Jan  5 10:28:28 rhel5-2 clurgmgrd[4759]: <notice> Starting stopped service service:a 
Jan  5 10:28:28 rhel5-2 clurgmgrd[4759]: <notice> Service service:a started 
Jan  5 10:28:28 rhel5-2 clurgmgrd[4759]: <notice> Service service:a is now running on member 2 
Jan  5 10:28:28 rhel5-2 clurgmgrd[4759]: <debug> Executing sortedservices node event handler for service: service:b 
Jan  5 10:28:28 rhel5-2 clurgmgrd[4759]: <debug> Evaluating service:b state=started owner=2 
Jan  5 10:28:28 rhel5-2 clurgmgrd[4759]: <debug> service:b is already running on best node 
Jan  5 10:28:28 rhel5-2 clurgmgrd[4759]: <debug> Executing default service event handler 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <info> Dependency lost; stopping service:b 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <notice> Stopping service service:b 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <notice> Service service:b is stopped 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <debug> Executing default service event handler 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <info> Dependency met; starting service:b 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <debug> Evaluating service:b state=stopped owner=-1 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <notice> Starting service:b on [ 2 ] 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <debug> Sent remote-start request to 2 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <notice> Starting stopped service service:b 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <notice> Service service:b started 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <notice> Service service:b is now running on member 2 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <debug> Executing default service event handler 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <debug> Executing default service event handler 
Jan  5 10:28:29 rhel5-2 clurgmgrd[4759]: <debug> 5 events processed

  Result: PASS


* failure case #3
  b on node 1, a on node 2, kill node 1.
  Expected results: move b over

Jan  5 10:31:18 rhel5-2 clurgmgrd[4759]: <info> State change: rhel5-1.lhh.pvt DOWN 
Jan  5 10:31:18 rhel5-2 clurgmgrd[4759]: <debug> sorted_service_list service:a, service:b 
Jan  5 10:31:18 rhel5-2 clurgmgrd[4759]: <debug> Executing sortedservices node event handler for service: service:a 
Jan  5 10:31:19 rhel5-2 clurgmgrd[4759]: <debug> Evaluating service:a state=started owner=2 
Jan  5 10:31:19 rhel5-2 clurgmgrd[4759]: <debug> service:a is already running on best node 
Jan  5 10:31:19 rhel5-2 clurgmgrd[4759]: <debug> Executing sortedservices node event handler for service: service:b 
Jan  5 10:31:19 rhel5-2 clurgmgrd[4759]: <debug> Evaluating service:b state=started owner=1 
Jan  5 10:31:19 rhel5-2 clurgmgrd[4759]: <info> Marking service service:b on down member 1 as stopped 
Jan  5 10:31:19 rhel5-2 clurgmgrd[4759]: <notice> Starting service:b on [ 2 ] 
Jan  5 10:31:19 rhel5-2 clurgmgrd[4759]: <debug> Sent remote-start request to 2 
Jan  5 10:31:19 rhel5-2 clurgmgrd[4759]: <notice> Starting stopped service service:b 
Jan  5 10:31:20 rhel5-2 clurgmgrd[4759]: <notice> Service service:b started 
Jan  5 10:31:20 rhel5-2 clurgmgrd[4759]: <notice> Service service:b is now running on member 2 
Jan  5 10:31:20 rhel5-2 clurgmgrd[4759]: <debug> Executing default service event handler 
Jan  5 10:31:20 rhel5-2 clurgmgrd[4759]: <debug> Executing default service event handler 
Jan  5 10:31:20 rhel5-2 clurgmgrd[4759]: <debug> 3 events processed 

  Result: PASS
Comment 8 Lon Hohberger 2012-01-05 15:51:40 UTC 
* config #2 (similar to config #1 except services in opposite order)

   <rm central_processing="1" log_level="7" ... >
     <service "b" depend="service:a" />
     <service "a" />
   </rm>

* failure case #1 (same expected results, but logs will differ)

Jan  5 10:40:18 rhel5-2 clurgmgrd[6800]: <debug> Membership Change Event 
Jan  5 10:40:18 rhel5-2 clurgmgrd[6800]: <info> State change: rhel5-1.lhh.pvt DOWN 
Jan  5 10:40:18 rhel5-2 clurgmgrd[6800]: <debug> sorted_service_list service:b, service:a 
Jan  5 10:40:18 rhel5-2 clurgmgrd[6800]: <debug> Executing sortedservices node event handler for service: service:b 
Jan  5 10:40:19 rhel5-2 clurgmgrd[6800]: <debug> service:b is not runnable; dependency not met 
Jan  5 10:40:19 rhel5-2 clurgmgrd[6800]: <debug> Executing sortedservices node event handler for service: service:a 
Jan  5 10:40:19 rhel5-2 clurgmgrd[6800]: <debug> Evaluating service:a state=started owner=1 
Jan  5 10:40:19 rhel5-2 clurgmgrd[6800]: <info> Marking service service:a on down member 1 as stopped 
Jan  5 10:40:19 rhel5-2 clurgmgrd[6800]: <notice> Starting service:a on [ 2 ] 
Jan  5 10:40:19 rhel5-2 clurgmgrd[6800]: <debug> Sent remote-start request to 2 
Jan  5 10:40:19 rhel5-2 clurgmgrd[6800]: <notice> Starting stopped service service:a 
Jan  5 10:40:20 rhel5-2 clurgmgrd[6800]: <notice> Service service:a started 
Jan  5 10:40:20 rhel5-2 clurgmgrd[6800]: <notice> Service service:a is now running on member 2 
Jan  5 10:40:20 rhel5-2 clurgmgrd[6800]: <debug> Executing default service event handler 
Jan  5 10:40:20 rhel5-2 clurgmgrd[6800]: <debug> Executing default service event handler 
Jan  5 10:40:20 rhel5-2 clurgmgrd[6800]: <debug> Executing default service event handler 
Jan  5 10:40:20 rhel5-2 clurgmgrd[6800]: <info> Dependency met; starting service:b 
Jan  5 10:40:20 rhel5-2 clurgmgrd[6800]: <debug> Evaluating service:b state=stopped owner=-1 
Jan  5 10:40:20 rhel5-2 clurgmgrd[6800]: <notice> Starting service:b on [ 2 ] 
Jan  5 10:40:21 rhel5-2 clurgmgrd[6800]: <debug> Sent remote-start request to 2 
Jan  5 10:40:21 rhel5-2 clurgmgrd[6800]: <notice> Starting stopped service service:b 
Jan  5 10:40:21 rhel5-2 clurgmgrd[6800]: <notice> Service service:b started 
Jan  5 10:40:21 rhel5-2 clurgmgrd[6800]: <notice> Service service:b is now running on member 2 
Jan  5 10:40:21 rhel5-2 clurgmgrd[6800]: <debug> Executing default service event handler 
Jan  5 10:40:21 rhel5-2 clurgmgrd[6800]: <debug> 5 events processed 

  Result: PASS


* failure case 2, same expected results:

Jan  5 10:48:40 rhel5-2 clurgmgrd[6800]: <debug> Membership Change Event 
Jan  5 10:48:40 rhel5-2 clurgmgrd[6800]: <info> State change: rhel5-1.lhh.pvt DOWN 
Jan  5 10:48:40 rhel5-2 clurgmgrd[6800]: <debug> sorted_service_list service:b, service:a 
Jan  5 10:48:40 rhel5-2 clurgmgrd[6800]: <debug> Executing sortedservices node event handler for service: service:b 
Jan  5 10:48:40 rhel5-2 clurgmgrd[6800]: <debug> service:b is not runnable; dependency not met 
Jan  5 10:48:41 rhel5-2 clurgmgrd[6800]: <notice> Stopping service service:b 
Jan  5 10:48:41 rhel5-2 clurgmgrd[6800]: <notice> Service service:b is stopped 
Jan  5 10:48:41 rhel5-2 clurgmgrd[6800]: <debug> Executing sortedservices node event handler for service: service:a 
Jan  5 10:48:41 rhel5-2 clurgmgrd[6800]: <debug> Evaluating service:a state=started owner=1 
Jan  5 10:48:41 rhel5-2 clurgmgrd[6800]: <info> Marking service service:a on down member 1 as stopped 
Jan  5 10:48:41 rhel5-2 clurgmgrd[6800]: <notice> Starting service:a on [ 2 ] 
Jan  5 10:48:41 rhel5-2 clurgmgrd[6800]: <debug> Sent remote-start request to 2 
Jan  5 10:48:42 rhel5-2 clurgmgrd[6800]: <notice> Starting stopped service service:a 
Jan  5 10:48:42 rhel5-2 clurgmgrd[6800]: <notice> Service service:a started 
Jan  5 10:48:42 rhel5-2 clurgmgrd[6800]: <notice> Service service:a is now running on member 2 
Jan  5 10:48:42 rhel5-2 clurgmgrd[6800]: <debug> Executing default service event handler 
Jan  5 10:48:42 rhel5-2 clurgmgrd[6800]: <info> Dependency met; starting service:b 
Jan  5 10:48:42 rhel5-2 clurgmgrd[6800]: <debug> Evaluating service:b state=stopped owner=-1 
Jan  5 10:48:42 rhel5-2 clurgmgrd[6800]: <notice> Starting service:b on [ 2 ] 
Jan  5 10:48:43 rhel5-2 clurgmgrd[6800]: <debug> Sent remote-start request to 2 
Jan  5 10:48:43 rhel5-2 clurgmgrd[6800]: <notice> Starting stopped service service:b 
Jan  5 10:48:43 rhel5-2 clurgmgrd[6800]: <notice> Service service:b started 
Jan  5 10:48:43 rhel5-2 clurgmgrd[6800]: <notice> Service service:b is now running on member 2 
Jan  5 10:48:43 rhel5-2 clurgmgrd[6800]: <debug> Executing default service event handler 
Jan  5 10:48:43 rhel5-2 clurgmgrd[6800]: <debug> 5 events processed

  Result: PASS


* failure case 3, same expected results:

Jan  5 10:51:18 rhel5-2 clurgmgrd[6800]: <debug> Membership Change Event 
Jan  5 10:51:18 rhel5-2 clurgmgrd[6800]: <info> State change: rhel5-1.lhh.pvt DOWN 
Jan  5 10:51:18 rhel5-2 clurgmgrd[6800]: <debug> sorted_service_list service:b, service:a 
Jan  5 10:51:19 rhel5-2 clurgmgrd[6800]: <debug> Executing sortedservices node event handler for service: service:b 
Jan  5 10:51:19 rhel5-2 clurgmgrd[6800]: <debug> Evaluating service:b state=started owner=1 
Jan  5 10:51:19 rhel5-2 clurgmgrd[6800]: <info> Marking service service:b on down member 1 as stopped 
Jan  5 10:51:19 rhel5-2 clurgmgrd[6800]: <notice> Starting service:b on [ 2 ] 
Jan  5 10:51:19 rhel5-2 clurgmgrd[6800]: <debug> Sent remote-start request to 2 
Jan  5 10:51:19 rhel5-2 clurgmgrd[6800]: <notice> Starting stopped service service:b 
Jan  5 10:51:20 rhel5-2 clurgmgrd[6800]: <notice> Service service:b started 
Jan  5 10:51:20 rhel5-2 clurgmgrd[6800]: <notice> Service service:b is now running on member 2 
Jan  5 10:51:20 rhel5-2 clurgmgrd[6800]: <debug> Executing sortedservices node event handler for service: service:a 
Jan  5 10:51:20 rhel5-2 clurgmgrd[6800]: <debug> Evaluating service:a state=started owner=2 
Jan  5 10:51:20 rhel5-2 clurgmgrd[6800]: <debug> service:a is already running on best node 
Jan  5 10:51:20 rhel5-2 clurgmgrd[6800]: <debug> Executing default service event handler 
Jan  5 10:51:21 rhel5-2 clurgmgrd[6800]: <debug> Executing default service event handler 
Jan  5 10:51:21 rhel5-2 clurgmgrd[6800]: <debug> 3 events processed 

  Result: PASS
Comment 9 Lon Hohberger 2012-01-05 15:59:16 UTC 
https://www.redhat.com/archives/cluster-devel/2012-January/msg00033.html
https://www.redhat.com/archives/cluster-devel/2012-January/msg00034.html
https://www.redhat.com/archives/cluster-devel/2012-January/msg00035.html
https://www.redhat.com/archives/cluster-devel/2012-January/msg00036.html
https://www.redhat.com/archives/cluster-devel/2012-January/msg00037.html
Comment 21 Lon Hohberger 2012-01-19 21:06:58 UTC 
    Technical note added. If any revisions are required, please edit the "Technical Notes" field
    accordingly. All revisions will be proofread by the Engineering Content Services team.
    
    New Contents:
Cause: A "stopped" event for a given service was being processed after the a service's dependent service had already been restarted.

Consequence: The dependent service would erroneously restart.

Fix: Ignore the "stopped" event if both services are running and the dependent service was started after the dependency.

Result: The "stopped" event no longer causes the dependent service to restart.
Comment 22 Lon Hohberger 2012-02-10 17:29:19 UTC 
*** Bug 679753 has been marked as a duplicate of this bug. ***
Comment 23 errata-xmlrpc 2012-02-21 06:19:28 UTC 
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHBA-2012-0163.html
Note You need to log in before you can comment on or make changes to this bug.