Bigip.conf manual reconfiguration issue

v11 really doesn't like having the bigip.conf edited. I'm not sure I've managed to get this to be anywhere near reliable.

Unfortunately the only way I can see this working (Reliably and easy) is a fairly complex script that would pull the conifg, workout which pools a node is a member of... Then remove all the poolmembers from the pools... Delete the node, re-add the node with the required name, and re-add the poolmembers...

Painful... very painful... I too would like to see the ability to simply rename objects in the config...

H

Just had to do something very similar to move from v9 to v11 with the change from IP address to name as the key. Used pycontrol to get the data out of v9 and reformatted it into a set of .tcl files. Deleted all the pools and nodes on v11 and ran a 'tmsh load sys config file (filename) merge. Yes, it was painful to write as I had to do it for everything to move the whole box to v11 (nodes, monitors, virtauls, selfIP's etc. etc.) but can build the whole config onto the viprion in about 10 mins.

I'm sure I'm way too late to help you out, but this may help out another. I just ran into this exact issue editing the bigip.conf the same as you, in order to rename all of my nodes and pool members to a name rather than IP. I used a python script (2.4.3) and using fileinput() to rewrite the file inplace on a mac osX (not sure if these params have anything to do with the issue). The fileinput method re-writes every line of the file in place, and I changed what I need with a regex. The new file loaded just fine, but one could not disable/enable any pool members.

The fix for me was to delete the trailing space on line 1323 (which was also present in the original migrated bigip.conf), save the file, and reload it. Upon reloading, it also reformatted bigip.conf to still contain the space, yet my cli/UI error was gone. My error was at a node object on a line containing text "monitor /Common/icmp" and "monitor none" on two distinctly different F5s and on the exact same line of 1323. These two LTMs had completely different config. Odd to me that your problem was on 1324, and both of mine were on 1323. Seems odd this happening on two exact same lines for me, and the very next line for you. They have to be connected.

My error: modify ltm pool testpool-80 members modify {testnode:80 {session user-enabled state user-up }} 01070712:3: Caught configuration exception (0), Invalid monitor rule instance identifier: 5076 - ltm/validation/MonitorRule.cpp, line 1323.

Hi all,

I've face the same difficulties and found an easier fix, which is just to load twice the configuration with tmsh command : tmsh load sys config - After the first run, the configuration is loaded but some error may occur (such as invalid monitor instance identifier) - After the second run, the error is no longer there

Hope this helps !

Hi all,

I've face the same difficulties and found an easier fix, which is just to load twice the configuration with tmsh command : tmsh load sys config - After the first run, the configuration is loaded but some error may occur (such as invalid monitor instance identifier) - After the second run, the error is no longer there

Hope this helps !

Line numbers in the messages

... - ltm/validation/monitorRule.cpp,line 1324

... - ltm/validation/MonitorRule.cpp, line 1323

refer to lines in the TMOS source code (the messages show a source file name and line number which caused the message). The numbers may be different in each user's case as the software versions (builds) might differ. Source line numbers are useful only to F5 engineering, when the eventual support case might get to their attention. Unless you have the TMOS source code :P

The source code line numbers have nothing to do with the configuration file entries.

Those numbers should refer to the bigip.conf lines:

Invalid monitor rule instance identifier:6258

Invalid monitor rule instance identifier: 5076

Cheers

-tarmo-

Line numbers in the messages

... - ltm/validation/monitorRule.cpp,line 1324

... - ltm/validation/MonitorRule.cpp, line 1323

refer to lines in the TMOS source code (the messages show a source file name and line number which caused the message). The numbers may be different in each user's case as the software versions (builds) might differ. Source line numbers are useful only to F5 engineering, when the eventual support case might get to their attention. Unless you have the TMOS source code :P

The source code line numbers have nothing to do with the configuration file entries.

Those numbers should refer to the bigip.conf lines:

Invalid monitor rule instance identifier:6258

Invalid monitor rule instance identifier: 5076

Cheers

-tarmo-