VSX R80.40 lab at eve-ng Failed to remove the following drivers from GW machine:

[KostasGR]

Hello community

I have setup an eve-ng lab (distributed) with r80.40  latest JHF take  102.

During the creation of a VSX VSLS cluster gateway with two members i get the below error.

Tue May 4 10:46:12 EEST 2021: Failed to remove the following drivers from GW machine: fw_2 44678357 0 fw_1 44678357 0 fw_0 44678357 2 simmod_0 18789024 0

Any ideas?

BR,

Kostas

[Jerom_van_den_1]

Looks like an issue in the JHF, I am actualy trying a reconfigure right now without an jhf and this does seem to work.

[Jerom_van_den_1]

Just to add another reply to this, also JHF Take 94 has this issue, so do not user this JHF as well.

[_Val_]

It seems to be an issue with JHF 102 that affect reconfigure of VSX. Please use older Jumbo when installing GWs. Once VSX is set up, you can then upgrade GWs to 102.

[cosmos]

I'm upgrading a customer right now from R77.30 to R80.40 and hit this issue on HFA 125 after the initial vsx_util reconfigure failed due to timeout. The process was restarted with the 'continue' option before this error and we've had to rebuild the target gateway (because reset_gw fails to load a line from cpprofile.sh).

My understanding was that hotfixes are intended to fix problems, not introduce them - is this not the case? Slowly chipping away at what was once a quality product...

[HristoGrigorov]

@cosmos Make a difference between hotfix and jumbo hotfix. While the first one solves one particular problem, later one is like a pack of many hotfixes and it is often so large that it can replace nearly entire system and not only solve problem but also introduce new features. In such case, pure statistically speaking it is no surprise problems happen here and there. Also complexity of the gateway software few years ago and now differs quite a lot. No excuses for CheckPoint in spoiling it here but just be more relaxed about doing such conclusions 😀

[Bob_Zimmerman]

I hit the same issue with jumbo 118 a month and change ago, and was given sk173683 and a hotfix. According to the SK, the issue is now solved in jumbo 126.

[cosmos]

Confirmed, we've just rolled HFA126 and it resolved the issue. We'd be rolling back if it weren't for someone else running into the problem first (for once).

Thanks Bob!

[jafara]

Hi KostasGR,

In most cases this message means that we have failed to remove Firewall-1 driver due to some process (or more) that currently using the driver.

to understand which process is it, please SSH to the Gateway and run the following command (from Expert mode):
fuser /dev/fw0

This command will output the PIDs of the process that currently using the driver.

you can run "ps aux | grep <the pid you got>" in order to understand which process is it.

Please let me know which process currently using the Firewall so we can assist you.

[KostasGR]

Hello jafara

Finally i followed the  _Val_ solution.

Created first the VSX cluster object and then installed the JHF 102.

BR,

Kostas

[_Val_]

@KostasGR 

Just to make sure, that was a workaround and not a solution. I am glad it worked for you