[DRBD-user] Fwd: DRBD+Heartbeat: Secondary node Refusing to be Primary

Maros Timko timkom at gmail.com
Tue Jul 20 20:35:34 CEST 2010

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


> as I have already explained my problem in my previous mail which is
> attached too, my logs shows that secondary node is demanding for at
> least one UptoDate disk.
> *the O/P of cat /proc/drbd * on my secondary system shows that;
>
> version: 8.3.2 (api:88/proto:86-90)
> GIT-hash: dd7985327f146f33b86d4bff5ca8c94234ce840e build by
> gmk at build64.infiscale.com, 2009-11-17 14:56:22
>
>  1: cs:SyncTarget ro:Secondary/Primary ds:*Inconsistent/UpToDate* C r----
>    ns:8 nr:2156892 dw:2156900 dr:313 al:0 bm:132 lo:0 pe:0 ua:1 ap:0
> ep:1 wo:b oos:112368444
>    [>....................] sync'ed:  1.9% (109732/111840)M
>    finish: 52:01:20 speed: 448 (316) K/sec

So the initial sync is in progress. You have to wait 52 hours to
finish or try to increase the communication speed.

>
> and *when primary node goes down *then it shows:
>
> [root at sec ~]# cat /proc/drbd
> version: 8.3.2 (api:88/proto:86-90)
> GIT-hash: dd7985327f146f33b86d4bff5ca8c94234ce840e build by
> gmk at build64.infiscale.com, 2009-11-17 14:56:22
>
>  1: cs:WFConnection ro:Secondary/Unknown ds:*Inconsistent/DUnknown* C r----
>    ns:8 nr:2190140 dw:2190148 dr:313 al:0 bm:134 lo:0 pe:0 ua:0 ap:0
> ep:1 wo:b oos:112335164
>
>
> As I expect drives must be in* UpToDate* /*UpToDate*  state for
> Secondary/Primary so that drbd could find atleast one UpToDate drive to
> make data consistent and start Secondary node as master.

You have to wait for initial sync to finish. Otherwise you don't have
a consistent data on the standby meaning you cannot use them.

Tino



More information about the drbd-user mailing list