search

CNA-Tech / PKS-Ninja

Other
133 stars 162 forks source link

Routing issues with OneCloud PKS-Ninja-T1-Baseline-0.1 template #704

Open sudhakarau1 opened 4 years ago

sudhakarau1 commented 4 years ago

While creating a guide for https://docs.pivotal.io/pks/1-6/console/prereqs-nsxt-automatednat.html at https://github.com/sudhakarau1/PKS-Ninja/blob/Pks1.6/LabGuides/PksInstallEpmc-PI9606/readme-autonat.md I used the PKS-Ninja-T1-Baseline-0.1. I found that the DNS server needed to be rebooted after updating the DNS entries after EPMC install succeeded. Even after rebooting ping appears to have some issues eg:

PING pks.corp.local (10.40.14.36) 56(84) bytes of data.
64 bytes from 10.40.14.36 (10.40.14.36): icmp_seq=1 ttl=61 time=2.09 ms
From 192.168.110.2 (192.168.110.2): icmp_seq=2 Redirect Host(New nexthop: 192.16   8.210.3 (192.168.210.3))
64 bytes from 10.40.14.36 (10.40.14.36): icmp_seq=2 ttl=61 time=7.34 ms
From 192.168.110.2 (192.168.110.2): icmp_seq=3 Redirect Host(New nexthop: 192.16   8.210.3 (192.168.210.3))
64 bytes from 10.40.14.36 (10.40.14.36): icmp_seq=3 ttl=61 time=2.19 ms
From 192.168.110.2 (192.168.110.2): icmp_seq=4 Redirect Host(New nexthop: 192.16   8.210.3 (192.168.210.3))
64 bytes from 10.40.14.36 (10.40.14.36): icmp_seq=4 ttl=61 time=1.90 ms
From 192.168.110.2 (192.168.110.2): icmp_seq=5 Redirect Host(New nexthop: 192.16   8.210.3 (192.168.210.3))
64 bytes from 10.40.14.36 (10.40.14.36): icmp_seq=5 ttl=61 time=2.11 ms
^C
--- pks.corp.local ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 4006ms
rtt min/avg/max/mdev = 1.907/3.129/7.341/2.108 ms

and

PING harbor.corp.local (10.40.14.37) 56(84) bytes of data.
64 bytes from 10.40.14.37 (10.40.14.37): icmp_seq=1 ttl=61 time=4.79 ms
From 192.168.110.2 (192.168.110.2): icmp_seq=2 Redirect Host(New nexthop: 192.16   8.210.3 (192.168.210.3))
64 bytes from 10.40.14.37 (10.40.14.37): icmp_seq=2 ttl=61 time=1.89 ms
From 192.168.110.2 (192.168.110.2): icmp_seq=3 Redirect Host(New nexthop: 192.16   8.210.3 (192.168.210.3))
64 bytes from 10.40.14.37 (10.40.14.37): icmp_seq=3 ttl=61 time=1.80 ms
From 192.168.110.2 (192.168.110.2): icmp_seq=4 Redirect Host(New nexthop: 192.16   8.210.3 (192.168.210.3))
64 bytes from 10.40.14.37 (10.40.14.37): icmp_seq=4 ttl=61 time=2.01 ms

--- harbor.corp.local ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 3004ms
rtt min/avg/max/mdev = 1.805/2.626/4.796/1.255 ms
afewellvmware commented 4 years ago

Hey Freddy so sorry I didnt see this sooner! I was creating an autonat template & guide at the same time you were, one difference is I massaged the ip pools that epmc uses to match pretty much the same ip addressing scheme we already used, I am not having any problems with my template and setup steps so perhaps the answer can be found by comparing the 2 guides. One thing to note is that ping in onecloud is often very unusual even when things are working due to the very unusual networking configuration we use in the nested underlay for the onecloud pods ... but in any case it should be working in the pks-ninja-t1-clusterReady-0.1 template so maybe you can reference the deltas to see if that highlights the issue.

Thanks!