Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Thu, Jan 22, 2004 at 08:45:51AM +0100, Francis SOUYRI wrote:
> >- Kernel change from 2.4.23 to 2.4.24 due to the security issue
> Vanilla kernel ? Distrib kernel ?
Vanilla patched with super-freeswan, lm-sensors and drbd.
> >- New CPU+Mainboards (Intel P4 & AMD-Athlon, both about 2,6GHz)
> >- New internal NICs, both Intel Gigabit with e1000.o driver
> Could you create a ram filesystems on each node and start a rsync.
> Do you have the possibility to test other NICs ?
Not possible while it's working in production...
> >Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn
> >dev8-0 58.00 6840.00 8.00 6840 8
> >
> > procs memory swap io system cpu
> >r b w swpd free buff cache si so bi bo in cs us sy id
> >...
> >0 4 7 0 592748 62792 177188 0 0 3780 188 762 935 0 2 98
> >
> >Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn
> >dev8-0 336.00 0.00 6496.00 0 6496
> >
> > procs memory swap io system cpu
> >r b w swpd free buff cache si so bi bo in cs us sy id
> >...
> >0 0 0 0 983316 4876 18260 0 0 0 3668 929 1332 0 5 95
> Could you give the config of each server (cpu, memory, disk adapter,
> nic...) the output on each node of:
What is worth mentioning is that the servers have 3ware 7500 RAID
controllers which issued "raid inititialization started" message after
the syncall was completet and the 3dmd daemon was started. I'm not sure
if this init process wasn't started prior starting the 3ware controller
daemon.
But the fact that I got much better values with dd and rsync *while*
the syncall was in progress does not speak for a great influence.
> Start only one synchro:
> /sbin/drbdsetup /dev/nbX replicate
Hm, not a good idea while they are running. But I posted the
iostat and vmstat values before.
> cat /proc/version
Linux version 2.4.24-mail3-3-intel (root at mail3a) (gcc version 2.95.4
20011002 (Debian prerelease)) #1 Tue Jan 20 17:41:49 CET 2004
(running Debian 3.0 (stable aka "woody"))
> cat /proc/drbd
I posted a snapshot of the syncall output before, this is normal:
version: 0.6.10 (api:64/proto:62)
0: cs:Connected st:Primary/Secondary ns:8390388 nr:0 dw:3342424 dr:5769481 pe:0 ua:0
1: cs:Connected st:Primary/Secondary ns:5405144 nr:0 dw:290200 dr:5146173 pe:0 ua:0
> cat /etc/drbd.conf
> cat /var/lib/drbd/drbd.conf.parsed
> /sbin/drbdsetup /dev/nb0 show
> /sbin/drbdsetup /dev/nb1 show
I attached these.
I did a throughput test with
dd if=/dev/zero of=/mnt/nb0/... bs=1M count=100
on the master and got the following iostat values on the primary:
Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn
dev8-0 300.00 54312.00 696.00 54312 696
which is about 10x as fast as durinc syncall although it is replicated,
too.
thanks,
-christian-
--
Christian Hammers WESTEND GmbH | Internet-Business-Provider
Technik CISCO Systems Partner - Authorized Reseller
Lütticher Straße 10 Tel 0241/701333-11
ch at westend.com D-52064 Aachen Fax 0241/911879
-------------- next part --------------
Lower device: 08:07 (/dev/sda7)
Disk options:
disk-size = 5124703 KB
do-panic
Local address: 10.10.10.1:7788
Remote address: 10.10.10.2:7788
Wire protocol: C
Net options:
timeout = 6.0 sec
tl-size = 256
connect-int = 10 sec
ping-int = 10 sec
sndbuf-size = 131070
sync-min = 100000 KB/sec
sync-max = 122070 KB/sec
sync-nice = -10
Lower device: 08:08 (/dev/sda8)
Disk options:
disk-size = 5124703 KB
do-panic
Local address: 10.10.10.1:7789
Remote address: 10.10.10.2:7789
Wire protocol: C
Net options:
timeout = 6.0 sec
tl-size = 256
connect-int = 10 sec
ping-int = 10 sec
sndbuf-size = 131070
sync-min = 100000 KB/sec
sync-max = 122070 KB/sec
sync-nice = -10
-------------- next part --------------
SEEN_RESOURCES=' drbd0 drbd1'
MINOR_COUNT=''
MODULE_OPTIONS=''
CONF=(\
__hash__68 \
'drbd0:protocol' 'C' \
'drbd0:fsckcmd' 'fsck -p -y' \
'drbd0:disk' ' --do-panic --disk-size=5124703' \
'drbd0:net' ' --sync-nice=-10 --sync-min=100000 --sync-max=122070 --timeout=60 --connect-int=10 --ping-int=10' \
'drbd0:on mail3a:device' '/dev/nb0' \
'drbd0:on mail3a:disk' '/dev/sda7' \
'drbd0:on mail3a:address' '10.10.10.1' \
'drbd0:on mail3a:port' '7788' \
'drbd0:HOSTS' ' mail3a mail3b ' \
'drbd0:on mail3b:device' '/dev/nb0' \
'drbd0:on mail3b:disk' '/dev/sda7' \
'drbd0:on mail3b:address' '10.10.10.2' \
'drbd0:on mail3b:port' '7788' \
'drbd0:skip-wait' 'no' \
'drbd0:load-only' 'no' \
'drbd0:inittimeout' '0' \
'drbd0:incon-degr-cmd' '0' \
'drbd1:protocol' 'C' \
'drbd1:fsckcmd' 'fsck -p -y' \
'drbd1:disk' ' --do-panic --disk-size=5124703' \
'drbd1:net' ' --sync-nice=-10 --sync-min=100000 --sync-max=122070 --timeout=60 --connect-int=10 --ping-int=10' \
'drbd1:on mail3a:device' '/dev/nb1' \
'drbd1:on mail3a:disk' '/dev/sda8' \
'drbd1:on mail3a:address' '10.10.10.1' \
'drbd1:on mail3a:port' '7789' \
'drbd1:HOSTS' ' mail3a mail3b ' \
'drbd1:on mail3b:device' '/dev/nb1' \
'drbd1:on mail3b:disk' '/dev/sda8' \
'drbd1:on mail3b:address' '10.10.10.2' \
'drbd1:on mail3b:port' '7789' \
'drbd1:skip-wait' 'no' \
'drbd1:load-only' 'no' \
'drbd1:inittimeout' '0' \
'drbd1:incon-degr-cmd' '0' \
)
MD5SUM='2fd4f1cb239e020f4bad87edc9f6cb89'
-------------- next part --------------
resource drbd0 {
protocol=C
fsckcmd=fsck -p -y
# inittimeout=60
# skip-wait
disk {
do-panic
disk-size=5124703
}
net {
sync-nice=-10
# KB/s
sync-min=100000
sync-max=122070
#tl-size=5000
timeout=60
connect-int=10
ping-int=10
}
on mail3a {
device=/dev/nb0
disk=/dev/sda7
address=10.10.10.1
port=7788
}
on mail3b {
device=/dev/nb0
disk=/dev/sda7
address=10.10.10.2
port=7788
}
}
resource drbd1 {
protocol=C
fsckcmd=fsck -p -y
# inittimeout=60
# skip-wait
disk {
do-panic
disk-size=5124703
}
net {
sync-nice=-10
# KB/s
sync-min=100000
sync-max=122070
#tl-size=5000
timeout=60
connect-int=10
ping-int=10
}
on mail3a {
device=/dev/nb1
disk=/dev/sda8
address=10.10.10.1
port=7789
}
on mail3b {
device=/dev/nb1
disk=/dev/sda8
address=10.10.10.2
port=7789
}
}
# vim: ts=2 syntax=awk