No traffic on network card in virtual Big-IP
Hi, We have two virtual deployments of F5 version 17.1.2.2 in a VMware environment, Each deployment has 3 network cards: MGMT external (for VSs) internal (for backend pool) Lately we started to experience a very strange behavior, where randomly, the external NIC stops receiving traffic. To resolve the issue we can restart the Big-IP, or disable/enable the NIC on the VMside. When taking pcaps on the BIG-IP there's just nothing coming in, this is why the F5 support told us they can't troubleshoot if there's no traffic, and pointed us to VMware support. We took pcaps and logs on the vmware side and it looks like traffic is not passing from the hypervisor to the VM but the support doesn't see any errors. Because the issue is random we can't simulate it. We tried to correlate the issue with vm migration, or external backups but couldn't pin point anything. We have a large virtual environment it this issue only occurs on two F5 machine (out of 400+ other VMs). I'm not entirely sure, but I think the issue started after upgrading from version 16.1.4.1 to 17.1.2.2. I was wondering if anyone stumbled upon something similar? ThanksEmpty reply from a server
Can anyone explain what does this "empty reply from a server" means See an example: [root@F5:Active:In Sync] config curl -i http://3.3.3.3:2037/docserver/inrotation.htmlcurl: (52) Empty reply from server ltm pool aaa_2037 { members { aaa_116:applus { address 3.3.3.3 priority-group 1 session monitor-enabled state up } min-active-members 1 monitor aaa_bbb_2037 and ccc_bbb_2037 ltm monitor tcp aaa_bbb_2037 { adaptive disabled defaults-from tcp destination *:applus interval 5 ip-dscp 0 recv none recv-disable none send none time-until-up 0 timeout 16 } ltm monitor http ccc_bbb_2037 { adaptive disabled defaults-from http destination *:webcache interval 5 ip-dscp 0 recv in_rotation recv-disable none send "GET /docserver/inrotation.html HTTP/1.1\r\nHost: ccc-bbb.com \r\nConnection: Close\r\n" time-until-up 0 timeout 16 } As you can see, it is still up. So i am not sure if "empty reply from server" is bad?
Best way to trace/troubleshoot traffic between Remedy web server and app server.
I was wondering, if we have a remedy web server pool and vip and a remedy app server pool and vip, what's the best way to trace/monitor/troubleshoot traffic from the web server pool to the app server pool? This is BMC remedy 8. Thanks.APM trouble shooting portal access
im wondering if there is a good way to troubleshoot portal access slowness? customer is saying they experience more slowness when going through the webtop. (a little might be expected, of course the trip to the APM takes time, but the say it is in the order of 10 seconds). of course i can pull out tcpdump and start capturing, but with an on-a-stick configuration, https and lots of production traffic that is going to mean a lot of data and a lot of work to match sessions. i checked the web application trace option, but that doesnt seem to log timestamps (shame btw, would be really useful). writing an irule to log requests and responses might be an option, but again quite some work. am i overlooking another option? or will it come down to one of the above (tcpdump / irule)? anyone ever experienced f5 support troubleshooting this, what do they do?
