Log in to ask questions, share your expertise, or stay connected to content you value. Don’t have a login? Learn how to become a member.
If you get a message like this one in your 128T conductor node, when logging into conductor PCLI
persistentDataManager:Session <user:admin, id:419> exceeds the netconf session limit of 4 for the 128T product. System performance is likely to be degraded.
This warning indicates that there are too many users logged into conductor. The default warning threshold is set to 4 logins. This is tracked to prevent performance degradation. The warning is adjustable, but I wouldn’t set it much higher, especially in versions of conductor <4.2.X.To analyze the existing logins, first list out all the logins with the following command (with output below):
[root@SDWAN-Conductor1 ~]# ps -eaf | grep pdmTransportAgentadmin 2092 2079 0 2019 ? 00:17:00 /usr/bin/pdmTransportAgentadmin 10786 10758 0 Feb18 ? 00:00:45 /usr/bin/pdmTransportAgentadmin 12171 12158 0 2019 ? 00:16:21 /usr/bin/pdmTransportAgentroot 14917 13318 0 13:48 pts/7 00:00:00 grep --color=auto pdmTransportAgentadmin 25800 25785 0 2019 ? 00:32:24 /usr/bin/pdmTransportAgentadmin 26213 26212 0 Feb19 ? 00:00:30 /usr/bin/pdmTransportAgent --startup-log-level Infoadmin 30898 30870 0 2019 ? 00:15:10 /usr/bin/pdmTransportAgentadmin 31519 31506 0 Jan03 ? 00:12:23 /usr/bin/pdmTransportAgent
[root@SDWAN-Conductor1 ~]# ps -eaf | grep pdmTransportAgent
admin 2092 2079 0 2019 ? 00:17:00 /usr/bin/pdmTransportAgent
admin 10786 10758 0 Feb18 ? 00:00:45 /usr/bin/pdmTransportAgent
admin 12171 12158 0 2019 ? 00:16:21 /usr/bin/pdmTransportAgent
root 14917 13318 0 13:48 pts/7 00:00:00 grep --color=auto pdmTransportAgent
admin 25800 25785 0 2019 ? 00:32:24 /usr/bin/pdmTransportAgent
admin 26213 26212 0 Feb19 ? 00:00:30 /usr/bin/pdmTransportAgent --startup-log-level Info
admin 30898 30870 0 2019 ? 00:15:10 /usr/bin/pdmTransportAgent
admin 31519 31506 0 Jan03 ? 00:12:23 /usr/bin/pdmTransportAgent
Notice that a lot of these are very old, even going back to 2019. I would occasionally clean this up by using the “kill -9 <pid>” command (no need to kill the process that was launched with the "--startup-log-level info" option. For example, after running a few kill commands on this particular system a few times, the output of the ps command looks a lot better:
[root@SDWAN-Conductor1 ~]# ps -eaf | grep pdmTransportAgentroot 17422 13318 0 13:56 pts/7 00:00:00 grep --color=auto pdmTransportAgentadmin 26213 26212 0 Feb19 ? 00:00:30 /usr/bin/pdmTransportAgent --startup-log-level Info
root 17422 13318 0 13:56 pts/7 00:00:00 grep --color=auto pdmTransportAgent
If you issue the same command with an additional parm as such:
ps -eaf | grep pdmTransportAgent | wc -l
you’ll get the total process count without the actual process records.