Mark Frasa wrote:
> Hello again,
>
> On Fri, 2007-04-13 at 07:10 -0600, Alan Robertson wrote:
>> Mark Frasa wrote:
>>> On Wed, 2007-04-11 at 07:36 -0600, Alan Robertson wrote:
>>>
>>> Hello,
>>>
>>>> Mark Frasa wrote:
>>>>> Hello,
>>>>>
>>>>> thanks for the tip, however cl_status rscstatus doesn't output a thing:
>>>>>
>>>>> box:~# cl_status rcstatus
>>>>> box:~#
>>>>>
>>>>> The only thing my syslog shows:
>>>>>
>>>>> Apr 11 08:56:11 dns01 cl_status[3505]: info: start:optind: 1
>>>>> argv[optind+1]: (null)
>>>>>
>>>>> And my messages:
>>>>>
>>>>> Apr 11 08:57:02 dns01 cl_status[3507]: info: start:optind: 1
>>>>> argv[optind+1]: (null)
>>>>> Apr 11 08:57:02 dns01 cl_status[3507]: info: End: optind: 1 argv[optind
>>>>> +1]: (null)
>>>>>
>>>>> Is there any reason why this appears? Hearthbeat is running on the
>>>>> localhost, but not on the remote host when i do this.
>>>> The logs are the author never removing his debug info.
>>>>
>>>> Just to double-check... This is an R1-style (haresources) installation
>>>> not one with the cib.xml file (R2/CRM-style)?
>>>>
>>>>
>>> Yep, its the simple setup :-)
>> Here's the code in question:
>> tatic int
>> rscstatus(ll_cluster_t *hb, int argc, char ** argv, const char * optstr)
>> {
>> const char * rstatus;
>>
>> if ( general_simple_opt_deal(argc, argv, optstr) < 0 ) {
>> /* There are option errors */
>> return PARAMETER_ERROR;
>> };
>>
>> rstatus = hb->llc_ops->get_resources(hb);
>> if ( rstatus == NULL ) {
>> cl_log(LOG_ERR
>> , "Cannot get client %s's resource status"
>> , argv[optind+2]);
>> cl_log(LOG_ERR, "REASON: %s", hb->llc_ops->errmsg(hb));
>> return UNKNOWN_ERROR;
>> }
>> if (FOR_HUMAN_READ == TRUE) {
>> printf("This node is holding %s resources.\n", rstatus);
>> } else {
>> printf("%s\n", rstatus);
>> }
>> return 0;
>> }
>>
>> general_simple_opt_deal prints an error if it thinks there's an error
>> that's occurred. (But, reading the code and how it's used, its attempt
>> to return an error will be ignored because it returns a positive value.)
>>
>> All the other cases print something as well...
>>
>> You might try invoking it with rscstatus -m and see if that prints "This
>> node is holding"...
>>
>> I just reread the code in the client side and server side for this
>> request. It's not at all obvious how an empty string can be returned.
>> And, if an error occurs, it's not obvious how it doesn't get printed
>> somewhere...
>>
>> 1) Check the error exit code
>> 2) try it with -m and see if something comes out
>> 3) raise debug and see if the messages tell you something
>> kill -SIGUSR1 $hbpid; kill -SIGUSR1 $hbpid;
>> cl_status rscstatus -m; echo $?
>> kill -SIGUSR2 $hbpid; kill -SIGUSR2 $hbpid;
>>
>
> When i enable the debugging i get the message:
>
> # cl_status rscstatus
> # Apr 13 15:44:27 dns01 cl_status[4001]: info: start:optind: 1
> argv[optind+1]: (null)
> Apr 13 15:44:27 dns01 cl_status[4001]: ERROR: Cannot get client HZ=100's
> status
> Apr 13 15:44:27 dns01 cl_status[4001]: ERROR: REASON: Can't send message
> to IPC Channel: Resource temporarily unavailable
> Apr 13 15:44:27 dns01 cl_status[4001]: info: End: optind: 1 argv[optind
> +1]: (null)
> Apr 13 15:44:27 dns01 cl_status[4001]: ERROR: Cannot sign off from
> heartbeat.
> Apr 13 15:44:27 dns01 cl_status[4001]: ERROR: REASON: Can't send message
> to IPC Channel: Resource temporarily unavailable can't send message to
> IPC: Resource temporarily unavailable
> Apr 13 15:44:27 dns01 cl_status[4001]: ERROR: return value:13
>
> This is only when i do cl_status rscstatus -m
>
> Though i am unsure what this error means, can you clarify it?
It means the code you're running doesn't look much like the code I have
in source control... Nor any version that I've seen...
Your code appears to be calling the clientstatus sub-command, not the
rscstatus sub-command -- at least if those messages are exactly correct
and nothing has been omitted. But, it looks like maybe a word has been
omitted from the message, or the old versions of this code were
different (which they don't seem to be).
What version is this, and where did you get the binaries for it?
--
Alan Robertson <[EMAIL PROTECTED]>
"Openness is the foundation and preservative of friendship... Let me
claim from you at all times your undisguised opinions." - William
Wilberforce
_______________________________________________
Linux-HA mailing list
[EMAIL PROTECTED]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
