[DRBD-user] write performance issues

Mark Dokter dokter at icg.tugraz.at
Mon Jul 11 16:23:09 CEST 2011

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.

On 07/11/2011 10:11 AM, Felix Frank wrote:
> On 07/08/2011 03:00 PM, Mark Dokter wrote:
>> I already did quite some research on that issue (that's how I came up
>> with that DRBD config), but didn't succeed in fixing it. Maybe someone
>> on this list has suggestions?
> It's possible that your Secondary is slowing you down.
> Could there be any weirdness going on with its RAID (there was recently
> a poster who had nearly-undetectable BBU issues)?

no BBU here...

> What IO schedulers are the nodes using? If CFQ, change it.

cat /sys/block/sda/queue/scheduler
noop [deadline]

> There can be other mean performance surprises, such as barrier
> synchronization being incredibly slow on certain platforms for no
> conceivable reason.
> I also notice that your syncer rate is quite high. As this is a Gigabit
> uplink, you may want to go as low as 30M, as 110 can bog down your link.
> Why is the DRBD link switched?

I had the syncer rate at 30M before and set it to that high value out of
desperation. There's nothing much going on on the syncer network, so
this shouldn't be a problem.
The two servers both have a dedicated nics which are connected to a gbit
Some time ago, those two servers were right next to each other
(physically) and I tried a direct connection of those two nics, which
did not really make a difference, so I plugged them to the switch again.
Now those two servers are physically about 1km appart in two different
server facilities and connected via VLAN. That shouldn't really matter,
as bandwith and latency between those two servers is quite acceptable
(ping measures ~0.2 ms).

> Regards,
> Felix


More information about the drbd-user mailing list