[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
Sat Jan 7 02:08:36 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.


My drbd's syncing speed is cycling between a little fast and very 
slow spending most of its time being slow.

It is running on two near identical computers (IBM x340s).
Both are running Gentoo, which is completely up-to-date with kernel 
2.6.15 and drbd 0.7.14.
One has gigabit interface and the other 100 Mbit. Both show full 
duplex on their respective interfaces and on the switch. The 
underlying drives are SCSI via a ServeRAID adapter.
One computer has 1 GB real RAM; the other, only 256 MB.
There is no LVM. The filesystem is ext3.
The performance is not affected by whether iptables is running with 
any rules or not.

The computers are not doing much else during this time.

As far as I can tell, this is the only aspect of the machines that 
are not running at full speed.

Here is output of /proc/drbd showing the performance start out 
reasonable, then slow down dramatically, sometimes even stopping, 
only to jump back to full speed. It spends most of its time at 
suboptimal speed though.

version: 0.7.14 (api:77/proto:74)
SVN Revision: 1989 build by root at kennedy1, 2006-01-05 20:07:14
  0: cs:SyncSource st:Secondary/Secondary ld:Consistent
     ns:123136 nr:0 dw:0 dr:123136 al:0 bm:13 lo:0 pe:135 ua:0 ap:0
         [>...................] sync'ed:  4.8% (410680/426656)K
         finish: 0:02:51 speed: 2,240 (1,452) K/sec

  0: cs:SyncSource st:Secondary/Secondary ld:Consistent
     ns:123676 nr:0 dw:0 dr:123676 al:0 bm:13 lo:0 pe:270 ua:0 ap:0
         [>...................] sync'ed:  4.8% (410680/426656)K
         finish: 0:05:42 speed: 1,120 (1,328) K/sec

  0: cs:SyncSource st:Secondary/Secondary ld:Consistent
     ns:125308 nr:0 dw:0 dr:125308 al:0 bm:13 lo:0 pe:316 ua:0 ap:0
         [=>..................] sync'ed:  5.8% (409232/426656)K
         finish: 0:06:49 speed: 920 (1,244) K/sec

  0: cs:SyncSource st:Secondary/Secondary ld:Consistent
     ns:128020 nr:0 dw:0 dr:128020 al:0 bm:13 lo:0 pe:496 ua:0 ap:0
         [=>..................] sync'ed:  5.8% (407240/426656)K
         finish: 0:07:55 speed: 808 (1,140) K/sec

  0: cs:SyncSource st:Secondary/Secondary ld:Consistent
     ns:131336 nr:0 dw:0 dr:131460 al:0 bm:13 lo:0 pe:1326 ua:31 ap:0
         [=>..................] sync'ed:  5.8% (407240/426656)K
         finish: 0:12:26 speed: 516 (924) K/sec

  0: cs:SyncSource st:Secondary/Secondary ld:Consistent
     ns:133320 nr:0 dw:0 dr:133488 al:0 bm:13 lo:0 pe:1769 ua:42 ap:0
         [=>..................] sync'ed:  5.8% (407028/426656)K
         finish: 0:33:55 speed: 104 (816) K/sec

  0: cs:SyncSource st:Secondary/Secondary ld:Consistent
     ns:133488 nr:0 dw:0 dr:133488 al:0 bm:13 lo:0 pe:61 ua:0 ap:0
         [=>..................] sync'ed:  7.7% (400032/426656)K
         finish: 0:02:37 speed: 2,400 (1,108) K/sec

Here is the relevant drbd.conf, which should allow drbd to move up to 
4 MB per second.

resource database {
         protocol C;
         incon-degr-cmd "echo '!DRBD! pri on incon-degr' | wall ; 
sleep 60 ; halt -f";
         startup { wfc-timeout 0; degr-wfc-timeout 120; }
         disk    { on-io-error detach; }
         net     { timeout 60; connect-int 10; ping-int 10;
             max-buffers 2048; max-epoch-size 2048; }
         syncer  { rate 4M; group 1; } # sync when r0 and r1 are 
finished syncing.
         on kennedy1 {
                 address 192.168.1.13:7790;
                 disk /dev/sdb1;
                 device /dev/drbd0;
                 meta-disk "internal";
         }
         on kennedy2 {
                 address 192.168.1.16:7790;
                 disk /dev/sdb1;
                 device /dev/drbd0;
                 meta-disk "internal";
         }
}


Other keywords: poor performance, slow performance, poor syncing 
speed, slow syncing speed, poor sync speed, slow sync speed, poor 
throughput, slow throughput
-- 

Maurice Volaski, mvolaski at aecom.yu.edu
Computing Support, Rose F. Kennedy Center
Albert Einstein College of Medicine of Yeshiva University



More information about the drbd-user mailing list