CPD issue on management server

[Adity12]

Hi All,

I have some issue with a management server appliance 405.

the issue is start, when user want access Smart Console but they realize it take more time process for login and after they can login to Smart Console they realize there is error "the connection with the server is was lost, any unsaved changes will be preserved.

when we checking service on management with cpwd_admin list: the output show that cpd service is terminate. and we try for restart but not successfully.

so i have already open case to TAC, and from TAC give me advice to use sk116861 ( on the last section )

according sk116861 when cpd is fail the cpwd is try to restart using port 8989/tcp? is this correct, if correct when we checked on that port, the port is using by another service, so we must kill those service and try restart again for cpd.

After that CPD status is E, but when we access Smart Console we get some issue like previous, and checking again with cpwd_admin list for see status cpd still running.

when using command cpstat os, we got error message failed to establish session connection with AMON Server 127.0.0.1:18192, what is mean of this? if my suspect is right, this is the root cause of the issue, since FWM will use CPD for collecting status Gateway from port 18192.

so for testing, we try to telnet from management to each of gateway using port 18192, the result is refused.

the result cpd_admin list is CPD is not running.

Does anyone have faced this issue?

Let me know if, there is something missing for checking this issue.

Thanks Regards

Dio Aditya P

[Chris_Atkinson]

Hi - Can you confirm which version & jumbo is installed on this system?

(Please note as of September only R80.40 and higher will be considered supported so if not already please start planning your upgrades.)

[Adity12]

Hi @Chris_Atkinson 

Currently we using R80.20 take 101 and there is no jumbo installed for this appliance.

If this is will solve but must upgrade, which your recommendation for upgrade or clean install? since for root only 35 GB.

As far i know if choose upgrade the size will 2 times big then clean install ( CMIIW ) because checkpoint will snapshoot the gaia os.

Thanks Regards

Dio Aditya Pradana

[Chris_Atkinson]

If it were me I'd atleast give myself and the TAC a fighting chance by taking restore points and installing the latest GA Jumbo for your version for the Management and whilst at it update to the latest build of SmartConsole.

[Adity12]

Hi @the_rock and @Chris_Atkinson,

Yesterday, we have already try cpstop ; cpstart, but this process seems stuck.

and we planned to reboot the appliance and after that the issue is resolved.

If my suspect is right,

When cpd is terminated, we see on cpd.elg that cpwd kill cpd process, maybe this is because cpwd can't start cpd after stop it, because port 8989/tcp is using by other service. and when those service kill and restart cpd again, cpd is up but i think this not fully working.

Or maybe there is problem on storage root, currently storage root only free 8.0 GB from 35 GB and we try to delete some file is unused, is that right?

Thanks Regards

Dio Aditya Pradana

[the_rock]

Well, @Adity12 , as you already probably know, debugging a working process would not make much sense, so unless cpd is broken, then debugging it may help. Otherwise, I agree with you about root partition. Please delete any files that might not be needed, as long its safe to do so.

What I do is first take a backup, then from expert mode, run something like this on mgmt server:

find / -size +600000000c (this will look for files size 600 MB and higher, you can replace the value with anything else)

Andy

[the_rock]

I assume if you do cpstop and cpstart or reboot the server, issue is still there?

[Adity12]

Hi @the_rock 

Yes, we have planning to reboot this server on tomorrow, and i will update the result in here.

[the_rock]

Sounds good, keep us posted!

[the_rock]

You can also try below if you had not already:

- Stop CPD with command
cpwd_admin stop -name CPD -path "$CPDIR/bin/cpd_admin" -command "cpd_admin stop"

- Start CPD without unlimited retries
cpwd_admin start -name CPD -path "$CPDIR/bin/cpd" -command "cpd" -retry_limit u

[Adity12]

Hi @the_rock 

I have already replicate this issue on my lab, and currently CPD is terminated.

but there is some condition is very different with customer site, on my lab even CPD still terminated, i still can access smart console without issue and i just get error on object SMS, on customer site when they want login to smart console it take long process until they be able to login and get error message "the connection with the server is was lost, any unsaved changes will be preserved".

and another issue is when i try access web UI, it's working without any issue and can login like normal on my lab.

On customer site, web UI only be able just showing login page, after user login the page just loading without any error.

I am curious about this issue, for CPM and FWM is running.

Thanks Regards

Dio Aditya P