Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Hello Chris! Ok, these numbers make more sense, your platform is sane. However this shows something very interesting about what's happening under the cover : The IO profile is different on both side, the destination issuing smaller IOs. In order to get something optimized, I'd rather expect the IO size information on primary side being transmitted to the secondary one in order to achieve "equal" things, unfortunately I don't have enough DRBD knowledge to tell if that's something one might act upon. If it turned to be impossible, the underliying disk technology and cache would then probably have a significant impact in terms of latencies, especially in C policy. Especially if one was to use it for virtualization which is known to be a bigger disk stresser than traditional contexts. I give hand to the DRBD experts now, just let's wait for what they think about that... Maybe there's some magic parameter that would allow us to make things go better. Anyhow, that's an interesting point, good to know! Best regards, Pascal. -----Message d'origine----- De : drbd-user-bounces at lists.linbit.com [mailto:drbd-user-bounces at lists.linbit.com] De la part de Chris.Gouveia at spirent.com Envoyé : samedi 11 juin 2011 01:50 À : drbd-user at lists.linbit.com Objet : [DRBD-user] Secondary server works harder than primary Thank you for your reply. Enclosed are more accurate and consistent results: Primary machine avg-cpu: %user %nice %system %iowait %steal %idle 0.16 0.00 1.19 15.70 0.00 82.94 Device: tps kB_read/s kB_wrtn/s kB_read kB_wrtn sda 37.33 24.00 985.33 72 2956 sdb 34.67 21.33 954.67 64 2864 sdd 40.33 16.00 950.67 48 2852 md0 729.33 61.33 2890.67 184 8672 Secondary machine avg-cpu: %user %nice %system %iowait %steal %idle 0.04 0.00 0.28 0.00 0.00 99.68 Device: tps kB_read/s kB_wrtn/s kB_read kB_wrtn sda 194.00 0.00 978.67 0 2936 sdb 199.67 0.00 990.67 0 2972 sdc 191.33 0.00 948.00 0 2844 md0 734.00 0.00 2916.00 0 8748 Would it be accurate to interpret that DRBD is issuing the same number of requests to both servers, but for whatever reason the software RAID on the secondary server is breaking up each request into smaller IO requests? If so, would anyone have any suggestions on how to debug the software RAID portion? Thanks for everyone's time, Chris Gouveia <DIV><FONT size="1"> E-mail confidentiality. -------------------------------- This e-mail contains confidential and / or privileged information belonging to Spirent Communications plc, its affiliates and / or subsidiaries. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution and / or the taking of any action based upon reliance on the contents of this transmission is strictly forbidden. If you have received this message in error please notify the sender by return e-mail and delete it from your system. Spirent Communications plc, Northwood Park, Gatwick Road, Crawley, West Sussex, RH10 9XN, United Kingdom. Tel No. +44 (0) 1293 767676 Fax No. +44 (0) 1293 767677 Registered in England Number 470893 Registered at Northwood Park, Gatwick Road, Crawley, West Sussex, RH10 9XN, United Kingdom. Or if within the US, Spirent Communications, 26750 Agoura Road, Calabasas, CA, 91302, USA. Tel No. 1-818-676- 2300 </FONT></DIV>