Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hi folks!
I have set up 2 almost identical machines with drbd 8.0.4 and the following
configuration:
---
global { usage-count yes; }
common {
protocol C;
disk {}
net {
ko-count 10;
after-sb-0pri disconnect;
after-sb-1pri call-pri-lost-after-sb;
after-sb-2pri call-pri-lost-after-sb;
always-asbp;
rr-conflict call-pri-lost;
}
startup {
degr-wfc-timeout 30;
}
syncer {}
}
resource mail {
on mailsrv {
device /dev/drbd1;
disk /dev/sda6;
address 192.168.0.3:8899;
meta-disk internal;
}
on websrv {
device /dev/drbd1;
disk /dev/sda6;
address 192.168.0.2:8899;
meta-disk internal;
}
handlers {
local-io-error "/root/bin/drbd/local-io-error mail";
outdate-peer "/root/bin/drbd/outdate-peer mail";
pri-lost "/root/bin/drbd/pri-lost mail";
pri-lost-after-sb "/root/bin/drbd/pri-lost-after-sb mail";
pri-on-incon-degr "/root/bin/drbd/pri-on-incon-degr mail";
}
}
resource web {
on mailsrv {
device /dev/drbd0;
disk /dev/sda5;
address 192.168.0.3:7788;
meta-disk internal;
}
on websrv {
device /dev/drbd0;
disk /dev/sda5;
address 192.168.0.2:7788;
meta-disk internal;
}
handlers {
local-io-error "/root/bin/drbd/local-io-error web";
outdate-peer "/root/bin/drbd/outdate-peer web";
pri-lost "/root/bin/drbd/pri-lost web";
pri-lost-after-sb "/root/bin/drbd/pri-lost-after-sb web";
pri-on-incon-degr "/root/bin/drbd/pri-on-incon-degr web";
}
}
---
All of the files in /root/bin/drbd are symlinks to /root/bin/drbd-handler,
which sends an email and then evaluates $0 to perform the task.
However, I don't get any email because no handlers are run. How can I fix
that?
The second problem is that when I reboot mailsrv, drbd on websrv seems to
sync tht complete partition (which takes over 20 minutes), while when I
reboot websrv, the sync to mailsrv takes a second. Is this related to the
command 'drbdadm -- --discard-my-data connect mail' I had to use because of
problems connecting the mail partitions? (This is also why I've disable hmac
authentication, the nodes are connected together over an own private LAN)
I hope I was clear enough and looking for answers
Greetings,
Fabiano