[DRBD-user] DRBD Slow/Laggy IO

Jeffery Soo js at realtechtalk.com
Tue Aug 31 15:16:37 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.

I just realized that during intensive disk operations such as copying 
many files or deleting results in not being able to list the contents of 
the mounted drbd partition for several seconds or even a few minutes.

I've confirmed on other machines with different kernels and DRBD (older) 
that this does not happen.

Does anyone have any idea?
The load goes up to like 16-19 but when you look at top there's not much 
evidence of what is causing the problem.

I can also confirm that I setup some Virtualbox test machines using the 
exact same kernel and modules and could not replicate this issue, so I 
am extremely stumped.

I can say the server I'm using DRBD on is faster than any of my test 
machines, and in general usage (not using the DRBD partition it is 
extremely fast).
Could it be the 64MB Cache & 2TB WD drives with lower RPMs causing this 
issue by some chance?  I know they're not superfast, but surely they can 
beat an old

If anyone has any suggestions or solutions I would appreciate it.


version: 8.3.4 (api:88/proto:86-91)
GIT-hash: 70a645ae080411c87b4482a135847d69dc90a6a2 build by 
xemul at ovzcore.sw.ru, 2009-10-12 19:29:01
  0: cs:Connected ro:Primary/Secondary ds:UpToDate/UpToDate C r----
     ns:5437192 nr:0 dw:254735772 dr:1099997 al:184126 bm:2121 lo:0 pe:0 
ua:0 ap:0 ep:1 wo:b oos:0

global {
   usage-count no;

resource r0 {
   protocol C;

   handlers {
#    pri-on-incon-degr "echo o > /proc/sysrq-trigger ; halt -f";
#    pri-lost-after-sb "echo o > /proc/sysrq-trigger ; halt -f";
     local-io-error "echo o > /proc/sysrq-trigger ; halt -f";

    net {

    after-sb-0pri discard-older-primary;
    after-sb-1pri call-pri-lost-after-sb;
    after-sb-2pri call-pri-lost-after-sb;


   startup {
     degr-wfc-timeout 120;

   disk {
     on-io-error   call-local-io-error;

   syncer {
     rate 80M;
     al-extents 257;

   on test2.ti {
     device     /dev/drbd0;
     disk       /dev/sda3;
     meta-disk  internal;

   on test1.ti {
     device     /dev/drbd0;
     disk       /dev/sda3;
     meta-disk  internal;


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20100831/51076a19/attachment.htm>

More information about the drbd-user mailing list