Issues Upgrading from 10.2.0 HF2 to 11.1 HF1

Please let us know what support says. Also , have you run into this ? : Solution 3345

when you upgraded, did you check the provisioning status of ASM? I've hit that with APM and GTM in test box upgrades, where the provisioning status is disabled, but as soon as I enable, the config is present.

 

 

Also, on rare occasion I just needed to go to the CLI and do a tmsh load /sys config

Jason,

 

Thanks for the thoughts, but apparently that is not the case. I checked and ASM was set to Dedicated in provisioning, and I had AVR licensed apparently so I set it to Minimal and ASM to Nominal, and tried again but no dice. Right after I did the upgrade I noticed it said that the config was not loaded, so I also tried to the tmsh load /sys config and it loaded fine but still no config. It looks like the bigip.conf file is pretty empty, but I still have all the network configs. Strange.... Have a call with the support guys today to look into this more.

 

 

Mike

can you try "/usr/libexec/bigpipe daol"? is there any error?

I tried running that and I get the following message

 

 

BIGpipe HTTP compression profile creation error:

 

01070293:3: The profile name (/Common/http-accountlink-compression) is already assigned to another profile.

 

 

Whats the difference between between commnads?

 

 

ALso when I look in /var/log/asm I see the following the line for each policy

 

 

warning ts_configsync.pl[10992]: Internal inconsistency found while installing ucs file. Could not restore active version of policy ej4web-annuity3p-7574_class_default_2. Activating current version of policy.

 

 

As well as these messages which don't look good.

 

 

Feb 23 08:51:46 nrcaappfw-3 notice (ts_proc)recovery_mngr.pl[8768]: (0)(INIT_OK): ASM initialization started

 

Feb 23 08:52:09 nrcaappfw-3 info perl[8768]: 01310028:6: ASM subsystem info (recovery_mngr.pl,main::handle_agent_msg): ASM initialized successfully

 

Feb 23 08:52:14 nrcaappfw-3 debug tsconfd[11433]: mcpd_bridge_thread started

 

Feb 23 08:52:15 nrcaappfw-3 notice dcc[11433]: 01310022:5: ASM started successfully

 

Feb 23 08:52:36 nrcaappfw-3 info add_profile.pl[11614]: ASM Add profile: '/Common/ftp' has been added successfully ...

 

Feb 23 08:52:42 nrcaappfw-3 info add_profile.pl[11622]: ASM Add profile: '/Common/http' has been added successfully ...

 

Feb 23 08:55:19 nrcaappfw-3 crit ControlPlane.CorrelationEngine[11459]: ControlPlane.CorrelationEngine|CRIT|Feb 23 08:55:19.285|11459| [read_message() at CorrelationEngine.cpp:1413] flwctl_read() - Failed to receive event from flow control socket. Program terminated.

 

Feb 23 08:55:19 nrcaappfw-3 crit perl[11442]: 01310027:2: ASM subsystem error (learning_manager.pl,F5::LrnPipe::read): flwctl_read failed, error 0

 

Feb 23 08:55:20 nrcaappfw-3 info perl[12263]: 01310028:6: ASM subsystem info (kill_ts.pl,main::add_ts_log_line): ASM stopped

 

Feb 23 08:55:43 nrcaappfw-3 info set_ibdata1_size.pl[14058]: Setting ibdata1 size finished successfully, a new size is: 8466M

 

Feb 23 08:56:51 nrcaappfw-3 notice (ts_proc)recovery_mngr.pl[13234]: (0)(INIT_OK): ASM initialization started

 

Feb 23 08:57:13 nrcaappfw-3 info perl[13234]: 01310028:6: ASM subsystem info (recovery_mngr.pl,main::handle_agent_msg): ASM initialized successfully

 

Feb 23 08:57:17 nrcaappfw-3 debug tsconfd[15250]: mcpd_bridge_thread started

 

Feb 23 08:57:18 nrcaappfw-3 notice dcc[15250]: 01310022:5: ASM started successfully

 

Ok I just got off the phone with support, the issue had to do with the fact that I had an HTTP Profile with compression turned on. In 11.x they have sepereated compression into its own profiles. So I went into the /config/bigpipe/bigip.conf file and commented out the profile that was causing the issue and ran a /usr/libexec/bigpipe daol verified the configuration returned and then did a tmsh save /sys config. I will just rebuild the one profile manually.

 

 

On a side note, once I got my configuration back I noticed none of my SSL applications were working in looking at the ltm log I saw messages saying that Connection Attempted to Insecure SSL server. After a little digging I noted that renegotiation was enable by default in both the clientssl and serverssl parent profiles and that the server side was set to Require Strict in the Secure Renegotiation field. In order to make this work I had to disable renegotiation.

cool!

 

 

Whats the difference between between commnads?bigpipe daol is used to convert v10 configuration which is in /config/bigpipe directory to v11 configuration which will be store in /config directory. installation script does this for us. anyway, if there is an issue in converting, v11 configuration may not be converted.

Mike, thanks for making your problem public. As a service provider we had the same error with a single client as their app wasn't 'up to scratch'.

 

 

This time round I built a new server ssl profile that didn't require 'strict renegotiation' and the application functions!

 

 

Got an amazon wish list?

JessB,

 

I am glad to see this could help out someone, just goes to show how much time it can save if people post their issues and resolution on these forums.

Hi Mike,

 

There is a very simple answer to your problem as I went through a terrible time .... :(..... the configuration file which is on ver. 10.x.x is not applicable on ver 11.x.x as the architecture is different. Now either you have to do the things again from scratch or F5 support should help you.

 

 

Regards,