[DRBD-user] Slowdowns during resync and normal operations

Jean-Francois Chevrette jfchevrette at iweb.com
Thu Nov 19 15:15:19 CET 2009


Hello,

Here I have a two nodes setup:
- CentOS 5.4 x64
- DRBD 8.3.2 (CentOS RPM)
- kernel 2.6.18-128.7.1.el5xen
- Dell PE1950
- Quad-Core Xeons
- 32GB of memory
- PCI NIC Intel 82571EB (e1000)
- Connected directly with a crossover cable.

Each server is a Xen host and only has 512MB of memory reserved for dom0.

DRBD resources appear to slow down or hang during a resync and 
apparently during normal operations too. During a rsync, we can clearly 
see the sync speed drop to a crawl for 1-2 seconds both in /proc/drbd 
and when using a network monitoring tool such as iptraf.

These slowdowns appear to be related to the kernel panics we are having 
on this cluster.

Any ideas what might cause this?
Anything you would suggest tweaking in our config?
Is 512MB RAM enough for DRBD to function properly?

We don't mind loosing a bit of performance for more stability :)


# drbdsetup /dev/drbd0 show
disk {
         size                    0s _is_default; # bytes
         on-io-error             detach;
         fencing                 dont-care _is_default;
         no-disk-barrier ;
         no-disk-flushes ;
         no-md-flushes   ;
         max-bio-bvecs           0 _is_default;
}
net {
         timeout                 60 _is_default; # 1/10 seconds
         max-epoch-size          8192;
         max-buffers             8192;
         unplug-watermark        128 _is_default;
         connect-int             10 _is_default; # seconds
         ping-int                10 _is_default; # seconds
         sndbuf-size             131070 _is_default; # bytes
         rcvbuf-size             131070 _is_default; # bytes
         ko-count                0 _is_default;
         cram-hmac-alg           "sha1";
         after-sb-0pri           disconnect _is_default;
         after-sb-1pri           disconnect _is_default;
         after-sb-2pri           disconnect _is_default;
         rr-conflict             disconnect _is_default;
         ping-timeout            5 _is_default; # 1/10 seconds
}
syncer {
         rate                    33792k; # bytes/second
         after                   -1 _is_default;
         al-extents              1801;
         verify-alg              "crc32c";
}
protocol C;
_this_host {
         device                  minor 0;
         disk                    "/dev/sda6";
         meta-disk               internal;
         address                 ipv4 10.11.1.1:7788;
}
_remote_host {
         address                 ipv4 10.11.1.2:7788;
}


Regards,
-- 
Jean-François Chevrette [iWeb]




More information about the drbd-user mailing list