[DRBD-user] Centos5 DRBD 8.3 protocol C very slow

Brian R. Hellman brian at linbit.com
Tue Jul 28 16:53:05 CEST 2009

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


If you make the changes Robert suggested make sure you understand what
they do, you can find more information on the DRBD users guide
http://www.drbd.org/users-guide/re-drbdconf.html

In addition you may find this DRBD Performance Tuning webinar helpful
you can watch and download it for free at the following link.

http://www.linbit.com/en/education/on-demand-webinars/drbd-performance-tuning/


Hope that helps,
Brian

LINBIT - Your Way to High Availability
8152 SW Hall Blvd., #209 : Beaverton, OR 97008

http://www.linbit.com


Coert Waagmeester wrote:
> Hello all DRBDers,
> 
> Great product!
> 
> My DRBD setup is as follows:
> 
> 2 hosts with each:
> 2x 1TB seagate SATA software RAID1
> LVM on top of that
> crossover gigabit link between the two.
> 
> My drbd.conf:
> ________________________________________________________________________
> global {
>   usage-count yes;
> }
> common {
>   protocol C;
>   syncer { rate 50M; }
>   net {
>   startup { wfc-timeout 0 ; }
> }
> #####<I skip 6 drbd sections, similar setup as drbd7>
> resource testlv {
>   device    /dev/drbd7;
>   disk      /dev/vg0/testlv;
>   meta-disk internal;
> 
>   on baldur.mydomain.local {
>     address   10.99.99.1:7794;
>   }
>   on thor.mydomain.local {
>     address   10.99.99.2:7794;
>   }
> }
> ________________________________________________________________________
> 
> the problem I am experiencing is that the maximum transfer rate is too
> slow.
> 
> I get full through-put on the back-to-back gigabit ( 990Mb/sec )
> 
> as in the config file drbd7 mirrors a logical volume mounted on dom0
> when i run dd on that lv:
> # dd if=/dev/zero of=300meg.file bs=1M count=300 oflag=direct
> 300+0 records in
> 300+0 records out
> 314572800 bytes (315 MB) copied, 39.5295 seconds, 8.0 MB/s
> 
> I have tried no-disk-flush and no-md-flush.
> 
> When I use protocol A the maximum throughput is also only 8MB/sec,
> but as soon as I run 'drbdadm disconnect all' on the secondary host, and
> rerun the dd I get 32MB/sec!
> 
> Any tips would be greatly appreciated!
> 
> Kind regards,
> Coert
> 
> PS I sent a mail prior to this with an attachment, but that is moderated
> maybe?
> 
> _______________________________________________
> drbd-user mailing list
> drbd-user at lists.linbit.com
> http://lists.linbit.com/mailman/listinfo/drbd-user



More information about the drbd-user mailing list