11-11-2009
VCS will not let you stop an agent if any resources owned by that agent (that is, resources of that Type) are online in the cluster.
Now, if you manage to stop the agent outside VCS, then VCS will kick out a dead-agent trigger (if configured), and I think it will automatically restart the agent (not really sure from memory)
If your desire is to stop the agent but still keep the monitored application (Oracle in your example) up, then "simply" delete all resources of that type.
After that, your application-service stays up and is no longer monitored, nor managed in any way via VCS until you re-establish the resources.
BTW: once you re-created a resource of that type, as soon as you enable it, the companion agent starts -- so be aware of that (this is generally what you want to have happen).
10 More Discussions You Might Find Interesting
1. Solaris
Hi All,
Can anyone help me to analyse this VCS log. One of my applications failover suddenly and I need to find out the reason behind this.
2009/04/09 10:58:57 VCS ERROR V-16-2-13067 (CS49PAPS2) Agent is calling clean for resource(Web-ebill2app) because the resource became OFFLINE... (4 Replies)
Discussion started by: ronny_nch
4 Replies
2. High Performance Computing
Weird question I know, but I'm intrigued.
Say for instance you have an Application in an SG with the usual start/stop/monitor. To successfully probe does it just check for existence of the script/file ?
Thanks... (1 Reply)
Discussion started by: itsupplies
1 Replies
3. High Performance Computing
Hi,
is anybody here experienced in running VCS on AIX and happy to give me some advice regarding configuration questions?
I expect very high network IO on my cluster - so a single gig-e network will not be sufficient - and though I would like to configure more than one network card per node and... (0 Replies)
Discussion started by: zxmaus
0 Replies
4. Solaris
we have a vcs cluster set up and noticed that they were showing one of the heart beat link qfe3 as DOWN. Both qfe2 and qfe3 were fine all along, btw if I were to push in/re-set the hb cable, do you know whether it will panic or cause anything to the system,? Its a DB critical prod nodes (3 Replies)
Discussion started by: incredible
3 Replies
5. Solaris
:)Hi,
can someone please explain VCS clustering and where do we need VCS clusters ..?
:o:)Thanks in advance :o:) (1 Reply)
Discussion started by: amitbisht9
1 Replies
6. Solaris
hi, dear all
I get a problem "VCS ERROR V-16-2-13077 " on VCS 4.1 for Solaris 10.
I can not offline the host2 when the raid is bad.
I don't know the reason and how to offline host2 and switch to host1.
please help me, thank you!
the message of engine_A.log is :
... (2 Replies)
Discussion started by: ForgetChen
2 Replies
7. Solaris
Hi admins, I have a question on shared diskgroup in vcs. Let's say i have two node cluster node A and B. I am having the failover service group (including the required resources such as disk grp, IP, mount..etc)
For an instance, the service grp is running on node A and suddenly if one of the... (0 Replies)
Discussion started by: snchaudhari2
0 Replies
8. Solaris
i am installed solaris 10 latest release i want to install Veritas cluster in solaris and i already installed VXVM in it and it works fine but when i install cluser on solaris i display an error like this what does it mean ? explain in detailed.
2 Failed to call set_value for... (1 Reply)
Discussion started by: samiulla
1 Replies
9. UNIX for Advanced & Expert Users
Do we have common VCS error codes for all platforms.
eg. 10195 Agent(s) for group %s failed on system %s
for all Linux,Solaris and windows ? (1 Reply)
Discussion started by: NIMISH AGARWAL
1 Replies
10. Solaris
Dear All,
My current system run Solaris 9 SPARC 64 bit , and VXVM , VCS 4.1. I check all the node information .
ls -l /etc/vx/*.exclude
/etc/vx/*.exclude: No such file or directory
root@devuardbs01 # vxdctl license
All features are available:
Mirroring
Root Mirroring
Concatenation... (0 Replies)
Discussion started by: linux_user
0 Replies
LEARN ABOUT DEBIAN
ocf_heartbeat_symlink
OCF_HEARTBEAT_SYMLIN(7) OCF resource agents OCF_HEARTBEAT_SYMLIN(7)
NAME
ocf_heartbeat_symlink - Manages a symbolic link
SYNOPSIS
symlink [start | stop | monitor | meta-data | validate-all]
DESCRIPTION
This resource agent that manages a symbolic link (symlink).
It is primarily intended to manage configuration files which should be enabled or disabled based on where the resource is running, such as
cron job definitions and the like.
SUPPORTED PARAMETERS
link
Full path of the symbolic link to be managed. This must obviously be in a filesystem that supports symbolic links. (required, string,
no default)
target
Full path to the link target (the file or directory which the symlink points to). (required, string, no default)
backup_suffix
A suffix to append to any files that the resource agent moves out of the way because they clash with "link". .sp If this is unset (the
default), then the resource agent will simply refuse to create a symlink if it clashes with an existing file. (optional, string, no
default)
SUPPORTED ACTIONS
This resource agent supports the following actions (operations):
start
Starts the resource. Suggested minimum timeout: 15.
stop
Stops the resource. Suggested minimum timeout: 15.
monitor
Performs a detailed status check. Suggested minimum timeout: 15. Suggested interval: 60.
meta-data
Retrieves resource agent metadata (internal use only). Suggested minimum timeout: 5.
validate-all
Performs a validation of the resource configuration. Suggested minimum timeout: 10.
EXAMPLE
The following is an example configuration for a symlink resource using the crm(8) shell:
primitive p_symlink ocf:heartbeat:symlink
params
link=string
target=string
op monitor depth="0" timeout="15" interval="60"
SEE ALSO
http://www.linux-ha.org/wiki/symlink_(resource_agent)
AUTHOR
Linux-HA contributors (see the resource agent source for information about individual authors)
resource-agents UNKNOWN 03/09/2014 OCF_HEARTBEAT_SYMLIN(7)