Forum Discussion
rightfax scaling issue
I'm currently attemping to loabalance rightfax and for some reason, going through the F5, the transaction never scales and as a result, is very slow. going directly to the server, I noticed it was very fast, but the source kept sending a very small window size, as opposed to sending a large one via the F5.
going through F5:
flags [DF], proto: TCP (6), length: 40) 10.10.10.234.64731 > 192.168.11.11.10520: ., cksum 0xf11f (correct), 70196:70196(0) ack 51049 win 64660 20:20:58.839247 IP (tos 0x0, ttl 255, id 17389, offset 0, flags [DF], proto: TCP (6), length: 1400) 10.10.10.234.64731 > 192.168.11.11.10520: P 70196:71556(1360) ack 51049 win 64660 20:20:58.840105 IP (tos 0x0, ttl 255, id 17391, offset 0, flags [DF], proto: TCP (6), length: 744) 10.10.10.234.64731 > 192.168.11.11.10520: P 71556:72260(704) ack 51049 win 64660 20:20:58.840408 IP (tos 0x0, ttl 255, id 17399, offset 0, flags [DF], proto: TCP (6), length: 40) 10.10.10.234.64731 > 192.168.11.11.10520: ., cksum 0xddba (correct), 72260:72260(0) ack 53075 win 65535 20:20:58.968870 IP (tos 0x0, ttl 255, id 17486, offset 0, flags [DF], proto: TCP (6), length: 1400) 10.10.10.234.64731 > 192.168.11.11.10520: P 72260:73620(1360) ack 53075 win 65535 20:20:58.968875 IP (tos 0x0, ttl 255, id 17488, offset 0, flags [DF], proto: TCP (6), length: 744) 10.10.10.234.64731 > 192.168.11.11.10520: P 73620:74324(704) ack 53075 win 65535 20:20:58.969442 IP (tos 0x0, ttl 255, id 17493, offset 0, flags [DF], proto: TCP (6), length: 40) 10.10.10.234.64731 > 192.168.11.11.10520: ., cksum 0xcdc0 (correct), 74324:74324(0) ack 55101 win 65535 20:20:59.097646 IP (tos 0x0, ttl 255, id 17603, offset 0, flags [DF], proto: TCP (6), length: 1400) 10.10.10.234.64731 > 192.168.11.11.10520: P 74324:75684(1360) ack 55101 win 65535 20:20:59.098578 IP (tos 0x0, ttl 255, id 17605, offset 0, flags [DF], proto: TCP (6), length: 744) 10.10.10.234.64731 > 192.168.11.11.10520: P 75684:76388(704) ack 55101 win 65535 20:20:59.098880 IP (tos 0x0, ttl 255, id 17610, offset 0, flags [DF], proto: TCP (6), length: 40) 10.10.10.234.64731 > 192.168.11.11.10520: ., cksum 0xbdc6 (correct), 76388:76388(0) ack 57127 win 65535 20:20:59.228072 IP (tos 0x0, ttl 255, id 17791, offset 0, flags [DF], proto: TCP (6), length: 1400) 10.10.10.234.64731 > 192.168.11.11.10520: P 76388:77748(1360) ack 57127 win 65535 20:20:59.228076 IP (tos 0x0, ttl 255, id 17793, offset 0, flags [DF], proto: TCP (6), length: 744) 10.10.10.234.64731 > 192.168.11.11.10520: P 77748:78452(704) ack 57127 win 65535 20:20:59.228662 IP (tos 0x0, ttl 255, id 17798, offset 0, flags [DF], proto: TCP (6), length: 40) 10.10.10.234.64731 > 192.168.11.11.10520: ., cksum 0xadcc (correct), 78452:78452(0) ack 59153 win 65535 20:20:59.354195 IP (tos 0x0, ttl 255, id 17906, offset 0, flags [DF], proto: TCP (6), length: 1400) 10.10.10.234.64731 > 192.168.11.11.10520: P 78452:79812(1360) ack 59153 win 65535 20:20:59.354200 IP (tos 0x0, ttl 255, id 17908, offset 0, flags [DF], proto: TCP (6), length: 744) 10.10.10.234.64731 > 192.168.11.11.10520: P 79812:80516(704) ack 59153 win 65535 20:20:59.354780 IP (tos 0x0, ttl 255, id 17915, offset 0, flags [DF], proto: TCP (6), length: 40) 10.10.10.234.64731 > 192.168.11.11.10520: ., cksum 0x9dd2 (correct), 80516:80516(0) ack 61179 win 65535 20:20:59.480357 IP (tos 0x0, ttl 255, id 17988, offset 0, flags [DF], proto: TCP (6), length: 1400) 10.10.10.234.64731 > 192.168.11.11.10520: P 80516:81876(1360) ack 61179 win 65535
- VFBCirrus
I was able to narrow the issue down to the upstream checkpoint fragmenting the bits. After enabling "do not fragment", things started working as expected.
Recent Discussions
Related Content
* Getting Started on DevCentral
* Community Guidelines
* Community Terms of Use / EULA
* Community Ranking Explained
* Community Resources
* Contact the DevCentral Team
* Update MFA on account.f5.com