Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hi, I'm trying to test drbd performance using https://www.drbd.org/en/doc/users-guide-84/ch-benchmark#s-measure-throughput for throughput and https://www.drbd.org/en/doc/users-guide-84/s-measure-latency for latency I have some questions: I'm using 2 vms for testing purposes. Measure latency and throughput using vms is a good idea? Should I expect valid results? Actually when I try to use the script for throughput I have wrong values: 20MB/s when I try to write on /dev/drbd3 (obtained with drbdadm sh-dev <resource>) 2.0GB/s when I try to write on /dev/disk/by-id/ata-QEMU_HARDDISK_QM00011 (obtained with drbdadm sh-ll-dev <resource>) Command is: dd if=/dev/zero of=/dev/disk/by-id/ata-QEMU_HARDDISK_QM00011 bs=512M count=1 oflag=direct Is there a way to do a valid test in my environment?? Furthermore, a test of this type should destroy metadata on the resource. Is this true? How can I check if metadata are ok? it seems that drbdadm down <resource> and drbdad up <resource> working well.... but I think this is not correct Thank you Marco -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20170124/cd0c837e/attachment.htm>