[Linux-HA] sticky resource status "transition" with hb 2.0.2

Guochun Shi gshi at ncsa.uiuc.edu
Sat Oct 29 15:39:32 MDT 2005


I fixed it in CVS. here is the patch to apply  for 2.0.2

--- heartbeat.c.orig    2005-10-29 16:37:02.000000000 -0500
+++ heartbeat.c 2005-10-29 16:37:08.000000000 -0500
@@ -3372,7 +3372,7 @@
                 * It will then get written to the cluster properly.
                 */
 
-               if ((smsg = msg2wirefmt(msg, &len)) == NULL) {
+               if ((smsg = msg2wirefmt_noac(msg, &len)) == NULL) {
                        cl_log(LOG_ERR
                        ,       "send_cluster_msg: cannot convert"
                        " message to wire format (pid %d)", (int)getpid());


thanks
-Guochun


Guochun Shi wrote:

> Joachim Banzhaf wrote:
>
>> Am Freitag, 28. Oktober 2005 21:24 schrieb Guochun Shi:
>>  
>>
>>> I meant to say steps: did you do any operation like
>>> hb_standby/hb_takeover/shutdown one node/...
>>> before it gets into that "transition" state? Or as soon as you started
>>> heartbeat in both nodes, it got in there?
>>>   
>>
>>
>> I have only one node so far.
>> The only thing needed to reach that state is to start heartbeat on 
>> that node.
>>
>>  
>>
>
> OK, I am able to reproduce the problem. Thanks
>
> -Guochun
> _______________________________________________
> Linux-HA mailing list
> Linux-HA at lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
>
>




More information about the Linux-HA mailing list