I have two nodes (AMD K7 Athlon 1.333 Mhz 256 Mbyte DDR, 40 Gigabyte
HD IBM 7200 rpm, motherboard asus a7m-266) with
Debian Linux 2.2 RC4 , kernel 2.4.14 (also tested with 2.4.13) and drbd
version 0.6.1-pre7.
All worked fine until I shutdown all the machines:
1) restarting only one PC
2) boots corretly
3) it asks if it should become primary
4) Automatically it become primary after the inittimeout (working
corretly!)
5) it runs other services (included heartbeat)
7) login prompt
8) I wait for 10 -20 seconds... (second node off)
9) Kernel panic:
Unable to handle kernel NULL pointer dereference at virtual address
00000004
printing eip:
d0848997
Oops: 0000
CPU: 0
EIP: 0010:[<d0848997>] Not tainted
EFLAGS: 00010296
eax: 00000000 ebx: cfde6000 ecx: 00000003 edx: 00000000
esi: 00000000 edi: 00000000 ebp: c1427eb4 esp: c1427ea0
ds: 0018 es: 0018 ss: 0018
Process kupdated (pid: 6, stackpage=c1427000)
Stack: cfde6000 cf91cb40 00000000 00000001 00000029 c1427ee8 d084cb5b
00000000
00000000 0000000c 00000001 00002b00 cd9263c0 00000029 00000000
cf91cb48
00000004 00001000 00000000 c01932ec c03247cc 00000001 cd9263c0
00000008
Call Trace: [<d084cb5b>] [<c01932ec>] [<c0193340>] [<c01352f4>]
[<c0135388>]
[<c0137ef7>] [<c01381da>] [<c0105628>]
Code: 8b 40 08 89 45 fc bf 0c 00 00 00 2b 7d 10 89 f3 89 f9 d3 eb
10) It not mount /dev/nb0, and I can only reboot
11) when I reboot, another kernel panic:
Unable to handle kernel NULL pointer dereference at virtual address
00000004
printing eip:
d0848997
Oops: 0000
CPU: 0
EIP: 0010:[<d0848997>] Not tainted
EFLAGS: 00010296
eax: 00000000 ebx: cfde6000 ecx: 00000003 edx: 00b80020
esi: 00170004 edi: 00170004 ebp: cc79be64 esp: cc79be50
ds: 0018 es: 0018 ss: 0018
Process shutdown (pid: 613, stackpage=cc79b000)
Stack: cfde6000 cf91c8c0 00170004 00000001 00000014 cc79be98 d084cb5b
00000000
00170004 0000000c 00000001 00002b00 cd7edf40 00000014 00170004
cf91c8c8
00000004 00001000 00000000 c01932ec c03247cc 00000001 cd7edf40
00000008
Call Trace: [<d084cb5b>] [<c01932ec>] [<c0193340>] [<c01352f4>]
[<c0135388>]
[<c01353e9>] [<c01354f4>] [<c013561f>] [<c01356a7>] [<c0106eb3>]
Code: 8b 40 08 89 45 fc bf 0c 00 00 00 2b 7d 10 89 f3 89 f9 d3 eb
This is my drbd.conf:
resource drbd0 {
protocol=C
fsckcmd=fsck.ext2 -p -y
inittimeout=30
disk {
do-panic
disk-size=0
}
net {
sync-rate=2712
# skip-sync
tl-size=256
timeout=15
connect-int=10
ping-int=10
}
on jack {
device=/dev/nb0
disk=/dev/hda14
address=10.0.0.1
port=7788
}
on daniel {
device=/dev/nb0
disk=/dev/hda14
address=10.0.0.2
port=7788
}
}
Shouldn't it start in stand alone mode?
With this behavior of DRBD, there MUST be no reboot of a node if the
other is crashed and not quickly rebooted .
Sorry for my english....
Thanks for help.
Good job!
Ivano Picco, Kinetikon.com
ivano@example.com
HD IBM 7200 rpm, motherboard asus a7m-266) with
Debian Linux 2.2 RC4 , kernel 2.4.14 (also tested with 2.4.13) and drbd
version 0.6.1-pre7.
All worked fine until I shutdown all the machines:
1) restarting only one PC
2) boots corretly
3) it asks if it should become primary
4) Automatically it become primary after the inittimeout (working
corretly!)
5) it runs other services (included heartbeat)
7) login prompt
8) I wait for 10 -20 seconds... (second node off)
9) Kernel panic:
Unable to handle kernel NULL pointer dereference at virtual address
00000004
printing eip:
d0848997
Oops: 0000
CPU: 0
EIP: 0010:[<d0848997>] Not tainted
EFLAGS: 00010296
eax: 00000000 ebx: cfde6000 ecx: 00000003 edx: 00000000
esi: 00000000 edi: 00000000 ebp: c1427eb4 esp: c1427ea0
ds: 0018 es: 0018 ss: 0018
Process kupdated (pid: 6, stackpage=c1427000)
Stack: cfde6000 cf91cb40 00000000 00000001 00000029 c1427ee8 d084cb5b
00000000
00000000 0000000c 00000001 00002b00 cd9263c0 00000029 00000000
cf91cb48
00000004 00001000 00000000 c01932ec c03247cc 00000001 cd9263c0
00000008
Call Trace: [<d084cb5b>] [<c01932ec>] [<c0193340>] [<c01352f4>]
[<c0135388>]
[<c0137ef7>] [<c01381da>] [<c0105628>]
Code: 8b 40 08 89 45 fc bf 0c 00 00 00 2b 7d 10 89 f3 89 f9 d3 eb
10) It not mount /dev/nb0, and I can only reboot
11) when I reboot, another kernel panic:
Unable to handle kernel NULL pointer dereference at virtual address
00000004
printing eip:
d0848997
Oops: 0000
CPU: 0
EIP: 0010:[<d0848997>] Not tainted
EFLAGS: 00010296
eax: 00000000 ebx: cfde6000 ecx: 00000003 edx: 00b80020
esi: 00170004 edi: 00170004 ebp: cc79be64 esp: cc79be50
ds: 0018 es: 0018 ss: 0018
Process shutdown (pid: 613, stackpage=cc79b000)
Stack: cfde6000 cf91c8c0 00170004 00000001 00000014 cc79be98 d084cb5b
00000000
00170004 0000000c 00000001 00002b00 cd7edf40 00000014 00170004
cf91c8c8
00000004 00001000 00000000 c01932ec c03247cc 00000001 cd7edf40
00000008
Call Trace: [<d084cb5b>] [<c01932ec>] [<c0193340>] [<c01352f4>]
[<c0135388>]
[<c01353e9>] [<c01354f4>] [<c013561f>] [<c01356a7>] [<c0106eb3>]
Code: 8b 40 08 89 45 fc bf 0c 00 00 00 2b 7d 10 89 f3 89 f9 d3 eb
This is my drbd.conf:
resource drbd0 {
protocol=C
fsckcmd=fsck.ext2 -p -y
inittimeout=30
disk {
do-panic
disk-size=0
}
net {
sync-rate=2712
# skip-sync
tl-size=256
timeout=15
connect-int=10
ping-int=10
}
on jack {
device=/dev/nb0
disk=/dev/hda14
address=10.0.0.1
port=7788
}
on daniel {
device=/dev/nb0
disk=/dev/hda14
address=10.0.0.2
port=7788
}
}
Shouldn't it start in stand alone mode?
With this behavior of DRBD, there MUST be no reboot of a node if the
other is crashed and not quickly rebooted .
Sorry for my english....
Thanks for help.
Good job!
Ivano Picco, Kinetikon.com
ivano@example.com