Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hi All,
Having issues with drbd proxy initial setup. My config is as follows:
cat drbd.conf
# See /usr/share/doc/drbd-8.0.2/drbd.conf or fully annotated file.
global { usage-count no; }
common {
protocol A;
handlers {
pri-on-incon-degr "/usr/bin/logger -p local2.emerg -t DRBD
pri-on-incon-degr tripped";
pri-lost-after-sb "/usr/bin/logger -p local2.emerg -t DRBD
pri-lost-after-sb tripped";
local-io-error "/usr/bin/logger -p local2.emerg -t DRBD local-io-error
tripped";
#outdate-peer "/usr/sbin/drbd-peer-outdater"; # needs more setup to use
}
syncer {
#rate 10M; # 100baseT
#rate 100M; # Full Gig-E
rate 12M; # don't saturate disk
}
startup {
wfc-timeout 30; # 30 seconds
degr-wfc-timeout 120; # 2 minutes.
}
net {
# increase timeout and maybe ping-int in net{}, if you see
# problems with "connection lost/connection established"
# timeout 60; # 6 seconds (unit = 0.1 seconds)
# connect-int 10; # 10 seconds (unit = 1 second)
# ping-int 10; # 10 seconds (unit = 1 second)
# ping-timeout 5; # 500 ms (unit = 0.1 seconds)
# With this option set you might make both nodes primary. You only should
use
# this options if you use a shared storage file system on top of DRBD. At
the
# time of writing the only ones are: OCFS2 and GFS. If you use this option
with
# any other filesystem you are goint to crash your nodes and to corrupt your
data!
# allow-two-primaries;
cram-hmac-alg "sha1";
shared-secret "nyb";
}
}
resource r0 {
proxy {
compression on;
memlimit 100M;
}
on ip-10-251-193-191 {
device /dev/drbd0;
disk /dev/VolGroup00/backupvg;
address 127.0.0.1:7789;
flexible-meta-disk internal;
proxy on ip-10-251-193-191 {
inside 127.0.0.1:7788;
outside 10.251.193.191:7788;
}
}
on fx-5 {
device /dev/drbd0;
disk /dev/loop0;
address 127.0.0.1:7789;
flexible-meta-disk internal;
proxy on fx-5 {
inside 127.0.0.1:7788;
outside 38.104.nyb.nyb:7788;
}
}
}
I only have two nodes. One node is on a regular server and the other is in
the amazon cloud. The cloud log file says:
Aug 28 14:20:46 ip-10-251-193-191 kernel: [9313062.114113] block drbd0:
conn( Unconnected -> WFConnection )
Aug 28 14:21:18 ip-10-251-193-191 kernel: [9313093.773054] block drbd0:
Handshake successful: Agreed network protocol version 90
Aug 28 14:21:18 ip-10-251-193-191 kernel: [9313093.773111] block drbd0:
sock_sendmsg returned -32
Aug 28 14:21:18 ip-10-251-193-191 kernel: [9313093.773116] block drbd0:
conn( WFConnection -> BrokenPipe )
Aug 28 14:21:18 ip-10-251-193-191 kernel: [9313093.773121] block drbd0:
Authentication of peer failed
Aug 28 14:21:18 ip-10-251-193-191 kernel: [9313093.773123] block drbd0:
Discarding network configuration.
Aug 28 14:21:18 ip-10-251-193-191 kernel: [9313093.773125] block drbd0:
conn( BrokenPipe -> Disconnecting )
Aug 28 14:21:18 ip-10-251-193-191 kernel: [9313093.773348] block drbd0:
Connection closed
Aug 28 14:21:18 ip-10-251-193-191 kernel: [9313093.773358] block drbd0:
conn( Disconnecting -> StandAlone )
Aug 28 14:21:18 ip-10-251-193-191 kernel: [9313093.773409] block drbd0:
receiver terminated
Aug 28 14:21:18 ip-10-251-193-191 kernel: [9313093.773411] block drbd0:
Terminating receiver thread
netstat -an|grep 7788 on the cloud shows:
[root at ip-10-251-193-191:/etc] netstat -an|grep 7788
tcp 0 0 10.251.193.191:7788 0.0.0.0:*
LISTEN
tcp 0 0 127.0.0.1:7788 0.0.0.0:*
LISTEN
netstat -an|grep 7789
[root at ip-10-251-193-191:/etc]
I think the reason is because of the Authentication of peer failed message
above. How do I fix that?
On the non-cloud side it shows 7788 and 7789 listening but has this in the
logs constantly:
Aug 28 11:27:38 fx-5 kernel: block drbd0: Connection closed
Aug 28 11:27:38 fx-5 kernel: block drbd0: conn( BrokenPipe -> Unconnected )
Aug 28 11:27:39 fx-5 kernel: block drbd0: conn( Unconnected -> WFConnection
)
Aug 28 11:27:53 fx-5 kernel: block drbd0: sock_recvmsg returned -11
Aug 28 11:27:53 fx-5 kernel: block drbd0: conn( WFConnection -> BrokenPipe
)
Aug 28 11:27:53 fx-5 kernel: block drbd0: short read expecting header on
sock: r=-11
Aug 28 11:27:53 fx-5 kernel: block drbd0: Connection closed
Aug 28 11:27:53 fx-5 kernel: block drbd0: conn( BrokenPipe -> Unconnected )
Aug 28 11:27:54 fx-5 kernel: block drbd0: conn( Unconnected -> WFConnection
)
Aug 28 11:28:08 fx-5 kernel: block drbd0: sock_recvmsg returned -11
Aug 28 11:28:08 fx-5 kernel: block drbd0: conn( WFConnection -> BrokenPipe
)
Aug 28 11:28:08 fx-5 kernel: block drbd0: short read expecting header on
sock: r=-11
Aug 28 11:28:08 fx-5 kernel: block drbd0: Connection closed
-Tim
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20090828/df6f7ca8/attachment.htm>