I deeply confused. I have active standby pair of F5. there are about 40 pools and 20 nodes. All the nodes are up in active and all the node are up in passive accept one. I don't understand why this one is down. All the nodes are VM,s in exa-logic. If all the vm,s are accessible from standby unit then why one VM is not accessible?

I am getting a ping response from the node after 90 failures. what does "after 90 failures" mean? If I am not getting ping response then I can not get TCP half open (monitor) response as well. if bigip does not get response, can you check if server does send response?

you may try to troubleshoot health monitor. Troubleshooting Ltm Monitors https://devcentral.f5.com/s/articles/ltm-external-monitors-troubleshooting

getting these logs, it coming up and down. but its continuously up in active F5 Nov 23 16:50:38 www notice mcpd[6575]: 01070728:5: Node /Common/swbs04 address 10.50.169.29 monitor status up. [ /Common/ICMP: up ] [ was down for 0hr:1min:20sec ] Nov 23 16:50:58 www notice mcpd[6575]: 01070640:5: Node /Common/swbs04 address 1 0.50.169.29 monitor status down. [ /Common/ICMP: down ] [ was up for 0hr:0min:20sec ] Nov 23 16:52:57 www notice mcpd[6575]: 01070728:5: Node /Common/swbs04 address 10.50.169.29 monitor status up. [ /Common/ICMP: up ] [ was down for 0hr:1min:59sec ] Nov 23 16:53:18 www notice mcpd[6575]: 01070640:5: Node /Common/swbs04 address 10.50.169.29 monitor status down. [ /Common/ICMP: down ] [ was up for 0hr:0min:2

Nitass your link was very informative but I am getting a ping response from the node after 90 failures. If I am not getting ping response then I can not get TCP half open (monitor) response as well. but no such problem on active F5. All the VM,s are in same vlan and all other vm,s are reachable from from both F5 except this one from standby unit

I am getting a ping response from the node after 90 failures. what does "after 90 failures" mean? If I am not getting ping response then I can not get TCP half open (monitor) response as well. if bigip does not get response, can you check if server does send response?

90 failures means after drop 90 pings one ping succeeded in getting response. The mac entry learned for that node by standby unit is different then mac entry of active the rest nodes mac are same. When I looked for that mac address in switch it was pointing towards active F5 unit, What is this. I tried to clear dynamic entries but of no good. That entry is still there pointing towards active F5 for that node. Which means standby unit is searching for that node through switch to active unit. ahhh

One node is down in standby unit

19 Replies

nitass
Employee
Nov 23, 2014
you may try to troubleshoot health monitor. Troubleshooting Ltm Monitors https://devcentral.f5.com/s/articles/ltm-external-monitors-troubleshooting
Muhammad_Irfan1
Cirrus
Nov 23, 2014
getting these logs, it coming up and down. but its continuously up in active F5 Nov 23 16:50:38 www notice mcpd[6575]: 01070728:5: Node /Common/swbs04 address 10.50.169.29 monitor status up. [ /Common/ICMP: up ] [ was down for 0hr:1min:20sec ] Nov 23 16:50:58 www notice mcpd[6575]: 01070640:5: Node /Common/swbs04 address 1 0.50.169.29 monitor status down. [ /Common/ICMP: down ] [ was up for 0hr:0min:20sec ] Nov 23 16:52:57 www notice mcpd[6575]: 01070728:5: Node /Common/swbs04 address 10.50.169.29 monitor status up. [ /Common/ICMP: up ] [ was down for 0hr:1min:59sec ] Nov 23 16:53:18 www notice mcpd[6575]: 01070640:5: Node /Common/swbs04 address 10.50.169.29 monitor status down. [ /Common/ICMP: down ] [ was up for 0hr:0min:2
Muhammad_Irfan1
Cirrus
Nov 23, 2014
Nitass your link was very informative but I am getting a ping response from the node after 90 failures. If I am not getting ping response then I can not get TCP half open (monitor) response as well. but no such problem on active F5. All the VM,s are in same vlan and all other vm,s are reachable from from both F5 except this one from standby unit
nitass_89166
Noctilucent
Nov 23, 2014
I am getting a ping response from the node after 90 failures.

what does "after 90 failures" mean?

If I am not getting ping response then I can not get TCP half open (monitor) response as well.

if bigip does not get response, can you check if server does send response?
- Muhammad_Irfan1
  Cirrus
  Nov 23, 2014
  90 failures means after drop 90 pings one ping succeeded in getting response. The mac entry learned for that node by standby unit is different then mac entry of active the rest nodes mac are same. When I looked for that mac address in switch it was pointing towards active F5 unit, What is this. I tried to clear dynamic entries but of no good. That entry is still there pointing towards active F5 for that node. Which means standby unit is searching for that node through switch to active unit. ahhh
- nitass_89166
  Noctilucent
  Nov 23, 2014
  are you using vlangroup?
- Muhammad_Irfan1
  Cirrus
  Nov 23, 2014
  yesssss. is this the problem? here is my network diagram http://imgur.com/gallery/ItMTFum/new
nitass
Employee
Nov 23, 2014
I am getting a ping response from the node after 90 failures.

what does "after 90 failures" mean?

If I am not getting ping response then I can not get TCP half open (monitor) response as well.

if bigip does not get response, can you check if server does send response?
- Muhammad_Irfan1
  Cirrus
  Nov 23, 2014
  90 failures means after drop 90 pings one ping succeeded in getting response. The mac entry learned for that node by standby unit is different then mac entry of active the rest nodes mac are same. When I looked for that mac address in switch it was pointing towards active F5 unit, What is this. I tried to clear dynamic entries but of no good. That entry is still there pointing towards active F5 for that node. Which means standby unit is searching for that node through switch to active unit. ahhh
- nitass
  Employee
  Nov 23, 2014
  are you using vlangroup?
- Muhammad_Irfan1
  Cirrus
  Nov 23, 2014
  yesssss. is this the problem? here is my network diagram http://imgur.com/gallery/ItMTFum/new