Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Bart, > I have to agree it's an ugly deal when iperf gives good > results. Puts it's validity as a good performance measurement > tool into perspective. The problem was not with iperf, but with my assumption that the iperf tests I ran a week or month before the problem were still valid. If I had re-run the iperf tests WHILE the problem was occurring, I think the source would have become evident instantly. Totally my own fault. > Another pointer might have been the error counters in > ifconfig for that interface. Yeah, that became clear to me as well. And yet another might have been the 'pe' counter in /proc/drbd on the primary showing the number of pending requests to the slave. On a healthy system, it is probably always at or near 0. I bet if I had known what to look for, I would have noticed some astronomically high number. I guess the good news is that since I figured all this out the hard way, I won't soon forget it. Thank you once again, School of Hard Knocks. :-) --Eric Disclaimer - February 19, 2011 This email and any files transmitted with it are confidential and intended solely for Bart Coninckx,drbd-user at lists.linbit.com. If you are not the named addressee you should not disseminate, distribute, copy or alter this email. Any views or opinions presented in this email are solely those of the author and might not represent those of Physicians' Managed Care or Physician Select Management. Warning: Although Physicians' Managed Care or Physician Select Management has taken reasonable precautions to ensure no viruses are present in this email, the company cannot accept responsibility for any loss or damage arising from the use of this email or attachments. This disclaimer was added by Policy Patrol: http://www.policypatrol.com/