Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Greetings! I am setting up a DRBD test system and having trouble with
the speed of the initial sync. I have searched the mailing list
archives and found numerous other users with similar problems, but
their problems either seemed unresolved or due to hardware/network
configuration issues.
My system consists of two Openfiler 2.2 installations (DRBD 8.0.0)
running on PCs directly connected by a 100baseT crossover cable and
also connected to my local network with a 100Mb interface.
Each PC has a single 40GB PATA drive which contains the OS as well as
a 30GB partition for the DRBD resource. FOr compatibility with the
Openfiler management interface, which wants everything to be LVM, the
30GB DRBD resource on each machine is configured as the sole physical
volume in an LVM volume group consisting of one LVM volume which
takes up the entire 30GB and is formatted as ext3.
Once I got everything created, I started the sync of my resource (r0)
from PC1 to PC2. 19 hours later, the sync is still progressing at
about 320 KB/s.
I do not believe the problem to be a network or hardware issue for the
following reasons:
a. I was able to scp a 2.5GB file from a machine on the local
network to the (still syncing) DRBD backed volume of the Primary
machine at a rate of 10MB/s.
b. I was able to scp that same 2.5GB file from the DRBD backed
volume to the root partition of the Secondary machine, over the DRBD
interface, at a rate of 5MB/s. That root partition is on the same
disk as the DRBD partition.
My drbd.conf file contains nothing but the default values:
global {
usage-count yes;
}
common {
syncer { rate 10M; }
}
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";
outdate-peer "/usr/sbin/drbd-peer-outdater";
}
startup {
degr-wfc-timeout 120; # 2 minutes.
}
disk {
on-io-error detach;
}
net {
after-sb-0pri disconnect;
after-sb-1pri disconnect;
after-sb-2pri disconnect;
rr-conflict disconnect;
}
syncer {
rate 10M;
}
on openfiler1.wcjb.com {
device /dev/drbd0;
disk /dev/hda4;
address 10.6.6.6:7788;
meta-disk internal;
}
on openfiler2.wcjb.com {
device /dev/drbd0;
disk /dev/hda4;
address 10.6.6.7:7788;
meta-disk internal;
}
}
Any help or advise on how to speed up the initial sync to expected
levels would be greatly appreciated.
-chris
--
Chris Ribe
TV/IT Engineer
WCJB-DT Gainesville, FL
(352) 416 0648
cribe at wcjb.com