[DRBD-user] [Q] What is causing drbd to be slow and cycle between a little fast and very slow?

Maurice Volaski mvolaski at aecom.yu.edu
Mon Feb 20 04:02:08 CET 2006

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


I'm seeing the "problem" again. This time secondary is doing a 
partial sync to the primary.

  1: cs:SyncTarget st:Secondary/Primary ld:Inconsistent
     ns:0 nr:558860 dw:558860 dr:0 al:0 bm:994 lo:1 pe:0 ua:0 ap:0
         [=================>..] sync'ed: 89.7% (68540/627400)K
         finish: 0:15:42 speed: 48 (2,972) K/sec

  2: cs:Connected st:Secondary/Primary ld:Consistent
     ns:0 nr:536140 dw:536140 dr:0 al:0 bm:814 lo:0 pe:0 ua:0 ap:0
  3: cs:SyncTarget st:Secondary/Primary ld:Inconsistent
     ns:0 nr:975720 dw:975720 dr:0 al:0 bm:494 lo:1 pe:0 ua:0 ap:0
         [============>.......] sync'ed: 61.4% (621876/1597596)K
         finish: 0:01:39 speed: 6,232 (5,188) K/sec
  4: cs:SyncTarget st:Secondary/Primary ld:Inconsistent
     ns:0 nr:970908 dw:970908 dr:0 al:0 bm:335 lo:1 pe:0 ua:0 ap:0
         [==============>.....] sync'ed: 70.9% (401564/1372472)K
         finish: 0:00:41 speed: 9,644 (5,164) K/sec
  5: cs:Connected st:Secondary/Primary ld:Consistent
     ns:0 nr:608476 dw:608476 dr:0 al:0 bm:687 lo:0 pe:0 ua:0 ap:0
  6: cs:SyncTarget st:Secondary/Primary ld:Inconsistent
     ns:0 nr:268732 dw:268732 dr:0 al:0 bm:1875 lo:1 pe:0 ua:0 ap:0
         [==========>.........] sync'ed: 50.4% (267664/536396)K
         finish: 0:00:39 speed: 6,724 (1,428) K/sec

  7: cs:SyncTarget st:Secondary/Primary ld:Inconsistent
     ns:0 nr:525212 dw:525212 dr:0 al:0 bm:1071 lo:19 pe:0 ua:19 ap:0
         [===================>] sync'ed:100.0% (3944/529080)K
         finish: 0:00:19 speed: 72 (2,792) K/sec

  8: cs:SyncTarget st:Secondary/Primary ld:Inconsistent
     ns:0 nr:750064 dw:750064 dr:0 al:0 bm:775 lo:805 pe:0 ua:804 ap:0
         [=====>..............] sync'ed: 27.8% (1969268/2716116)K
         finish: 0:03:06 speed: 10,536 (3,972) K/sec

Note the low values for 1 and 7. I'm not sure the actual throughput 
is this low. It appears to be just an artifact of how the values are 
being calculated.
-- 

Maurice Volaski, mvolaski at aecom.yu.edu
Computing Support, Rose F. Kennedy Center
Albert Einstein College of Medicine of Yeshiva University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20060219/9318a20e/attachment.htm>


More information about the drbd-user mailing list