Closed curtkersey closed 3 years ago
I setup my lab and ran into this same issue. I was on f5-bigip-14-1-2-3-0-0-5-payg-good-1gbps-191218142235, uploaded the 14.1.2.3 ISO (yes...same version), upgraded to boot slot 2, rebooted. After reboot, device unreachable other than through console.
Behavior seems identical to past bug https://github.com/F5Networks/f5-google-gdm-templates/issues/33
Workaround (example)
Note: you must delete all mgmt routes first because they still exist. Somehow tmsh list sys management-route still shows the routes but netstat output does not. Therefore delete sys mgmt routes first, then add back in just like workaround above.
If you need to know the commands to use, you can do this...
P.S. This also happens on a 15.1.0.2 upgrading to 15.1.0.2. Seems to be related to version upgrade. A regular reboot causes no harm. But a version upgrade + reboot causes loss of mgmt routing.
Thanks for reaching out to us with this bug. A DB variable is used for the nic swap and it looks like this may be getting lost during the upgrade.
Internal ID VECLOUD-989 has been created for this.
@shyawnkarim, we've seen this issue when automating upgrade testing from 15.0.1 (template default) to 15.1.0.4 (latest code) as well. This was deployed with GDM template 3.6.0.
Closing.
This bug has been fixed and details of the the fix can be seen on the F5 Bug Tracker.
Do you already have an issue opened with F5 support?
No support case opened for this issue. A support case was opened on the original issue, which prompted the need to load engineering HF, bug ID852437.
Description
Deployed v3.4 of template, https://github.com/F5Networks/f5-google-gdm-templates/tree/master/supported/failover/same-net/via-lb/3nic/existing-stack/byol. imageName: f5-bigip-14-1-2-3-0-0-5-byol-all-modules-2boot-loc-19121814234.
On upgrade (to same version, 14.1.2.3), I was not able to connect to management IP any longer - not SSH or TMUI. I tried to connect via a machine local to the BIG-IP (same GCP VCP), and that did not work either.
Found issue with management routes being goofed up. I grabbed the commands from /var/log/cloud/google/install.log that set the management routes. I ran those commands, then it worked.
Routes after upgrade (broken): netstat -rn Kernel IP routing table Destination Gateway Genmask Flags MSS Window irtt Iface 0.0.0.0 10.10.2.1 0.0.0.0 UG 0 0 0 external 10.10.1.1 0.0.0.0 255.255.255.255 UH 0 0 0 mgmt 10.10.2.0 10.10.2.1 255.255.255.0 UG 0 0 0 external 10.10.2.1 0.0.0.0 255.255.255.255 UH 0 0 0 external 10.10.3.0 10.10.3.1 255.255.255.0 UG 0 0 0 internal 10.10.3.1 0.0.0.0 255.255.255.255 UH 0 0 0 internal 127.1.1.0 0.0.0.0 255.255.255.0 U 0 0 0 tmm 127.7.0.0 127.1.1.253 255.255.0.0 UG 0 0 0 tmm 127.20.0.0 0.0.0.0 255.255.0.0 U 0 0 0 tmm_bp
Routes after adding management routes back in:
netstat -rn
Kernel IP routing table Destination Gateway Genmask Flags MSS Window irtt Iface 0.0.0.0 10.10.2.1 0.0.0.0 UG 0 0 0 external 0.0.0.0 10.10.1.1 0.0.0.0 UG 0 0 0 mgmt 10.10.1.0 10.10.1.1 255.255.255.0 UG 0 0 0 mgmt 10.10.1.1 0.0.0.0 255.255.255.255 UH 0 0 0 mgmt 10.10.2.0 10.10.2.1 255.255.255.0 UG 0 0 0 external 10.10.2.1 0.0.0.0 255.255.255.255 UH 0 0 0 external 10.10.3.0 10.10.3.1 255.255.255.0 UG 0 0 0 internal 10.10.3.1 0.0.0.0 255.255.255.255 UH 0 0 0 internal 127.1.1.0 0.0.0.0 255.255.255.0 U 0 0 0 tmm 127.7.0.0 127.1.1.253 255.255.0.0 UG 0 0 0 tmm 127.20.0.0 0.0.0.0 255.255.0.0 U 0 0 0 tmm_bp
Severity Level
Severity 2 -- big issue for my customer that needs to load engineering HF.