Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Is there any other tweaking I can try to get more speed out of the sync?
I am currently limited to around 16M network speed on my 100M private
network.
### mii-tool:
eth0: negotiated 100baseTx-FD flow-control, link ok
(Verified on both nodes)
### ping:
PING bilbo2 (10.0.0.2) 56(84) bytes of data.
64 bytes from bilbo2 (10.0.0.2): icmp_seq=1 ttl=64 time=0.056 ms
64 bytes from bilbo2 (10.0.0.2): icmp_seq=2 ttl=64 time=0.050 ms
64 bytes from bilbo2 (10.0.0.2): icmp_seq=3 ttl=64 time=0.050 ms
64 bytes from bilbo2 (10.0.0.2): icmp_seq=4 ttl=64 time=0.052 ms
64 bytes from bilbo2 (10.0.0.2): icmp_seq=5 ttl=64 time=0.050 ms
64 bytes from bilbo2 (10.0.0.2): icmp_seq=6 ttl=64 time=0.051 ms
--- bilbo2 ping statistics ---
6 packets transmitted, 6 received, 0% packet loss, time 4995ms
rtt min/avg/max/mdev = 0.050/0.051/0.056/0.007 ms
### /proc/drbd:
version: 0.6.10+cvs (api:64/proto:62)
0: cs:SyncingAll st:Primary/Secondary ns:5332700 nr:0 dw:12 dr:5332741
pe:88 ua:0
[=>..................] sync'ed: 9.3% (51301/56509)M
finish: 0:54:57h speed: 16,224 (15,918) K/sec
### /etc/drbd.conf:
resource drbd0 {
protocol = C
fsckcmd = /bin/true
disk {
do-panic
disk-size = 57865316
}
net {
sndbuf-size = 65534 # Smaller still??
sync-nice = -10 # Would -20 help??
sync-min = 50M # Nice try..? :)
sync-max = 90M # left 10MB for heartbeat...?
tl-size = 5000
timeout = 60
connect-int = 10
ping-int = 10
ko-count = 4
}
on bilbo1 {
device = /dev/nb0
disk = /dev/cciss/c0d0p3
address = 10.0.0.1 # Could it be the 10.255.255.255 netowrk class??
port = 7788
}
on bilbo2 {
device = /dev/nb0
disk = /dev/cciss/c0d0p3
address = 10.0.0.2
port = 7788
}
}
Thank you,
Michael Vera