SCC console gets stuck when Sentienl has been running for a while

Hi All

    My have some customer use sentinel 8.5.0.1 /8.6.1.0  . total have 6 sentinel appliances

Within a week after Sentienl starts the service, SCC can be started and used normally.
But when Sentinel has been running for more than a month... I try to open the SCC Console, and it often gets stuck in the SSO step without the user authentication window appearing.
The method I usually use is to restart the Sentinel service, and SCC can smoothly provide users with normal use.
However, restarting the Sentinel service will affect the log collection function (especially UDP Connector events may be lost). Is there any good way to solve this problem without restarting the Sentinel service?

I have confirmed that it is not a problem with the Jre installed on the client... Because as long as the Sentinel service is restarted, SCC can basically be used smoothly no matter when it is turned on within a week.

Tags:

  • 0  

    You mention both Sentinel 8.5 and 8.6, so does this same issue happen against both?

    I'd be looking at resources on the Sentinel server itself e.g. CPU memory etc.  Do you have sufficient hardware? Performance_snapshot logs are good for this as you can look at the snapshots (every 15min) leading up to the issue and see if something is increasing over time up to the point of failure.

    Also, at the point of failure, have a look a the ports used on the server.  The SCC uses port 8443 and 10013.  Are there lots of these in use on the server?  If so, what state are they in?

    Lastly, check the logs.  Do they report anything on the issue?

    Regards

    Crof

  • 0 in reply to   

    1. Yes....From 8.5.x to latest 8.6.1.0.....all have the same problem....even my lab (it only for testing...) has the same problem

    2. usually when problem  occur...the CPU use not high usage...about 30-50%, 

    3. about these senitnel appliance...they does not install other software...they usually only for log server purpose.

    4. no , I did not see any special error...at the same time...the log server function and web console still work fine.

    whether I have any method let 8443 & 10013 or SCC service could restart ? we know now senitnel use docker for each compoment....whether I could restart which docker container to let SCC work fine ?

    Wencheng

  • 0   in reply to 

    I don't know what version of Java you are running on the Workstations but I would install a version that is at least equal to that running on the Sentinel server itself.  Then ....

    https://portal.microfocus.com/s/article/KM000015837

    Note - This is mentioned in the documentation ..

    https://www.microfocus.com/documentation/sentinel/8.6/s86-install/t4mah3gmso03.html

  • 0 in reply to   

    I have 2 appliance 

    server A is fresh install 8.6.0 when upgrade to 8.6.1.0

    ServerB is senitnel 8.5.1.0

    I use a win2022 install Eclipse Temurin JRE with Hotspot 8u382-b05 (x64), it could launch these appliace's SCC

    Today, I upgade 8.5.1.0 to 8.6.1.0 , then re-download SCC to launch.

    but it stuck at single sign on step.

    I refer the KB to disable TLSv1.3

    https://portal.microfocus.com/s/article/KM000015837?language=en_US

    ==============================

    I had disable TLSv1.3 ==>restart sentinel servce , I got the same result

    another appliance (ServerA) aso been set disable TLSv1.3, it work fine.

    Which step I miss ??

    Thanks!!

    Wencheng

  • Verified Answer

    0   in reply to 

    You might want to test this issue against Sentinel 8.6.1.1.  It's not specifically called out in the release notes, but I believe it includes a fix for a Control Center listener (port 10013) issue.

  • 0 in reply to   

    HI  

        I had create a case...and engineer said this condition has common...

    but like me (sentinel run a period , when could not open SCC)....still not find the root cause.

    My lab senitnel (almost no EPS, has the same issue occur...so I take snapshot to let engineer testing)

    Wencheng

  • 0   in reply to 

    It's a condition seen by a few customers.  As I said, it should be resolved by the Sentinel 8.6.1.1 patch.