Hi Yan,
I referred to your patch.
I gave the following patch.
----------------------- patch
------------------------------------------------------------
+ mgmt_log(LOG_INFO, "yamauchi ----- migrate_cmd0:%s",argv[3]);
if (STRNCMP_CONST(argv[3], "True") == 0){
strncat(cmd, " -f", sizeof(cmd)-strlen(cmd)-1);
mgmt_log(LOG_INFO, "yamauchi ----- migrate_cmd1:%s",argv[3]);
}
+ mgmt_log(LOG_INFO, "yamauchi ----- migrate_cmd2:%s",cmd);
if (STRNCMP_CONST(argv[4], "") != 0){
if (regex_match(duration_regex, argv[4])) {
strncat(cmd, " -u \"", sizeof(cmd)-strlen(cmd)-1);
strncat(cmd, argv[4], sizeof(cmd)-strlen(cmd)-1);
strncat(cmd, "\"", sizeof(cmd)-strlen(cmd)-1);
}
else {
mgmt_log(LOG_ERR, "invalid duration specified: \"%s\"",
argv[1]);
return strdup(MSG_FAIL"\nInvalid duration.\nPlease
refer to "
"http://en.wikipedia.org/wiki/ISO_8601#Duration for examples
of valid durations");
}
}
strncat(cmd, " 2>&1", sizeof(cmd)-strlen(cmd)-1);
+ mgmt_log(LOG_INFO, "yamauchi ----- migrate_cmd3:%s",cmd);
------------------------------------------------------------------------------------
And the log when I chose True and False by operation to migrate was the
following thing.
[EMAIL PROTECTED] ~]# tail -f /var/log/messages | grep yamauchi
Nov 4 09:39:33 ais-1 mgmtd: [12837]: info: yamauchi ----- migrate_cmd0:\True
Nov 4 09:39:33 ais-1 mgmtd: [12837]: info: yamauchi -----
migrate_cmd2:crm_resource -M -r group -H
ais-2
Nov 4 09:39:33 ais-1 mgmtd: [12837]: info: yamauchi -----
migrate_cmd3:crm_resource -M -r group -H
ais-2 2>&1
Nov 4 09:39:53 ais-1 mgmtd: [12837]: info: yamauchi ----- migrate_cmd0:\False
Nov 4 09:39:53 ais-1 mgmtd: [12837]: info: yamauchi -----
migrate_cmd2:crm_resource -M -r group -H
ais-1
Nov 4 09:39:53 ais-1 mgmtd: [12837]: info: yamauchi -----
migrate_cmd3:crm_resource -M -r group -H
ais-1 2>&1
After all backslash seems to be in the top.
Possibly will not there be a problem at the time of the transfer of the
parameter with the daemon?
Best Regards,
Hideo Yamauchi.
--- Yan Gao <[EMAIL PROTECTED]> wrote:
> On Fri, 2008-10-31 at 20:17 +0900, HIDEO YAMAUCHI wrote:
> > Hi Yan,
> >
> > It confirmed it with the machine with a little different environment.
> > (CentOS5.2 on VirtualMachin, Python 2.4.3)
> >
> > > + print migrate
> > Result :
> > {'duration': '', 'force': 'True', 'to_node': 'cent51-vm2', 'rsc':
> > 'primitive'}
> >
> Strange, nothing is wrong. So could you pls add the following line in
> mgmt_crm.c, you get "\\True" or "True" 鏤・n the log?
>
> mgmt/daemon/mgmt_crm.c
> @@ -1124,6 +1124,7 @@
> }
> }
>
> + mgmt_log(LOG_WARNING, "migrate argument 3: %s", argv[3]);
> if (STRNCMP_CONST(argv[3], "True") == 0){
> strncat(cmd, " -f", sizeof(cmd)-strlen(cmd)-1);
> }
>
> > And, crm_resource that goes out to the log is the following.
> >
> > Oct 31 20:14:22 cent51-vm1 crm_resource: [15882]: info: Invoked:
> > crm_resource -M -r primitive -H cent51-vm2
> > Oct 31 20:14:22 cent51-vm1 crm_resource: [15882]: notice:
> > unpack_config: On
> > loss of CCM Quorum: Ignore
> >
> > Is the influence of the difference between my environment and your
> > environment?
> >
>
> Regards,
> --
> Yan Gao
> China R&D Software Engineer
> [EMAIL PROTECTED]
>
> Novell, Inc.
> SUSE速 Linux Enterprise 10
> Your Linux is ready
> http://www.novell.com/linux
>
>
> _______________________________________________
> Pacemaker mailing list
> [email protected]
> http://list.clusterlabs.org/mailman/listinfo/pacemaker
>
_______________________________________________
Pacemaker mailing list
[email protected]
http://list.clusterlabs.org/mailman/listinfo/pacemaker
