[DRBD-user] Several cases of hangups found. I have some stack traces to send.

Jesus Climent djmooch at gmail.com
Fri Feb 8 17:12:29 CET 2013

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


Interesting, indeed. I performed "echo t > /proc/sysrq-trigger" when:

* there was a D state process in hostname 14 and 15
* there was a stalled process in hostname16 (trace.pri) with the
following conditions:

hostname17
 3: cs:SyncSource ro:Primary/Primary ds:UpToDate/Inconsistent C r-----
    ns:2504848 nr:0 dw:26648380 dr:17020682 al:7872 bm:152 lo:2
pe:13023 ua:0 ap:1 ep:1 wo:d oos:352
        [===================>] sync'ed:100.0% (352/952)K
        finish: 9:17:03 speed: 0 (0) K/sec (stalled)

hostname16
 3: cs:SyncTarget ro:Primary/Primary ds:Inconsistent/UpToDate C r-----
    ns:0 nr:2504848 dw:26648376 dr:158380 al:4886 bm:150 lo:0 pe:27
ua:13112 ap:0 ep:13113 wo:d oos:296
        [===================>] sync'ed:100.0% (296/952)K
        finish: 7:35:55 speed: 0 (0) want: 61,440 K/sec (stalled)

And the cluster was stalled.



More information about the drbd-user mailing list