1. Log in to the CSP OM portal, choose Application Configuration > Service Management > csp > OMLB for Application Name, and obtain the pod information and project of the active instance OMLB. Reference
Obtain the following information:
podName:
status_omlb:M
KUBE_POD_NAMESPACE
2. Run the following command to obtain the floating IP address of the CSP on the PaaS:
kubectl get cm -n {Project name} omlb -oyaml|grep virtual
3. Run the following command on the PaaS background to obtain the IP address of the node where the active OMLB instance resides and run the SSH command to log in to the node.
kubectl get po -n {Project name} -owide|grep omlb;
4. Log in to the OMLB container using SSH, switch to user root, and run the following command to switch to the OMLB container:
docker exec -tiu root `docker ps|grep omlb|grep -v pause|awk '{print $1}'` bash
Check the floating IP address configuration. (The floating IP address is randomly bound to one OMLB. If the floating IP address does not exist, check the other OMLB.)
ifconfig
Viewing Routes
route -n
Check whether the OMLB can communicate with the LCM.
ping {vnfmip}
5. After switching to user root on the VM node, run the following command to copy the trace (tracepath or traceroute) command to the OMLB container:
docker cp /usr/bin/tracepath `docker ps|grep omlb|grep -v pause|awk '{print $1}'`:/usr/bin
docker cp /usr/bin/tracepath6 `docker ps|grep omlb|grep -v pause|awk '{print $1}'`:/usr/bin
6. Run the following command to switch to the OMLB container:
tracepath -n {lcmip}
tracepath6 -n {lcmip}
7. Contact datacom engineers to check the switch configuration based on the physical topology and trace path.
----------------Command summary. Run the following command on the node where the active OMLB node is located:
omlbid=`docker ps|grep omlb|grep -v pause|awk '{print $1}'`
docker exec -tiu root $omlbid ifconfig
docker exec -tiu root $omlbid route -n
docker exec -tiu root $omlbid ping $lcmip
docker cp /usr/bin/tracepath $omlbid:/usr/bin
docker exec -tiu root $omlbid tracepath -n $lcmip
----------------------------------------
1. Run the ping command.
Time to lie exceeded: A loop occurs on the physical switching device.
2. Run the trace command to continuously loop traffic between two IP addresses.
Check the datacom configurations of the two IP addresses.
3. After the trace command is executed, traffic times out at a hop and cannot be forwarded.
Check the switch configuration of the datacom device.
Attachment: Network Architecture
