search cancel

Service seosload.socket inactive

book

Article ID: 230504

calendar_today

Updated On:

Products

CA Privileged Identity Management Endpoint (PIM)

Issue/Introduction

The rseload utility requires service seosload.socket to be loaded and active at the remote endpoint. 

However in some cases the service may not be active and checking its status or trying to load it results in error:

ROOT.xe26298.mymachine./usr/seos>systemctl status seosload.socket
● seosload.socket - seosload remote seload service (rseloadd)
   Loaded: loaded (/etc/systemd/system/seosload.socket; enabled; vendor preset: disabled)
   Active: failed (Result: resources)
   Listen: 0.0.0.0:8892 (Stream)
 Accepted: 0; Connected: 0

Oct 31 10:43:27 mymachine systemd[1]: seosload.socket failed to listen on sockets: Address family not supported by protocol
Oct 31 10:43:27 mymachine systemd[1]: Failed to listen on seosload remote seload service (rseloadd).
Oct 31 10:43:27 mymachine systemd[1]: Unit seosload.socket entered failed state.

This prevents remote management of ControlMinder at that endpoint

Environment

CA ControlMinder, CA PIM and CA PAM SC UNIX all versions about 12.8 running on Linux systems using systemd

Cause

The problem is a failure of a native facility.   The  PAM SC/PIM  listener is not involved at all until one actually attempts to connect to  port 8892 which is where the service should be listening and send on it a recognized sequence of arguments.   When  systemctl status or systemctl enable are run,  a service is assigned in the native facility.  The problem is if systemd is in an inconsistent state rather than in the rseosload service,

It is possible to try to get systemd out of its erroneous state by stopping it (disregarding that it is not running) and then starting it again as will be shown below.

 

Resolution

A proposed sample of commands to execute to try to get systemd out of its erroneous state is the following:

/tmp> netstat -tln | grep 8892
tcp6       0      0 :::8892                 :::*                    LISTEN     
/tmp> systemctl stop seosload.socket
/tmp> netstat -tln | grep 8892  <----  This should not show anything.  Wait about a minute is that socket in in closing state ...
/tmp> systemctl start  seosload.socket
/tmp> systemctl  status seosload.socket
● seosload.socket - CA PAMSC seosload remote seload service (rseloadd)
   Loaded: loaded (/etc/systemd/system/seosload.socket; enabled; vendor preset: disabled)
   Active: active (listening) since Sun 2021-11-21 09:12:00 EST; 5s ago
   Listen: [::]:8892 (Stream)
 Accepted: 0; Connected: 0

Nov 21 09:12:00 lvndev010596.bpc.broadcom.net systemd[1]: Listening on CA PAMSC seosload re....

If unit files were updated or situation is unknown, prior to the sequence shown above please go through the whole installation service sequence starting with double-checking unit files:

cat /etc/systemd/system/seosload.socket

cat /etc/systemd/system/[email protected]

systemctl disable seosload.socket

systemctl enable seosload.socket

systemctl daemon-reload

This should get the service to behave once again normally