ConfD User Community

Reason for confd restart on cdb suspend/resume operation


#1

I am implementing suspend/resume for subscriptions.

And i see CONFD logs in confd.conf

16-Jul-2018::16:23:30.613 swdev-hng012-red1 confd[31080]: - Daemon SystemDataMgr died
16-Jul-2018::16:23:30.614 swdev-hng012-red1 confd[31080]: - Daemon uniOperdaemon died
16-Jul-2018::16:23:30.614 swdev-hng012-red1 confd[31080]: - Daemon unidaemon_confd died
16-Jul-2018::16:23:30.614 swdev-hng012-red1 confd[31080]: - Daemon unidaemonactions_confd died
16-Jul-2018::16:23:32.943 swdev-hng012-red1 confd[31080]: - Daemon SystemDataMgr died
16-Jul-2018::16:23:32.944 swdev-hng012-red1 confd[31080]: - Daemon uniOperdaemon died
16-Jul-2018::16:23:32.944 swdev-hng012-red1 confd[31080]: - Daemon unidaemon_confd died
16-Jul-2018::16:23:32.945 swdev-hng012-red1 confd[31080]: - Daemon unidaemonactions_confd died
16-Jul-2018::16:23:35.374 swdev-hng012-red1 confd[31080]: - Daemon SystemDataMgr died
16-Jul-2018::16:23:35.374 swdev-hng012-red1 confd[31080]: - Daemon uniOperdaemon died
16-Jul-2018::16:23:35.374 swdev-hng012-red1 confd[31080]: - Daemon unidaemon_confd died
16-Jul-2018::16:23:35.375 swdev-hng012-red1 confd[31080]: - Daemon unidaemonactions_confd died
16-Jul-2018::16:23:37.889 swdev-hng012-red1 confd[31080]: - Daemon SystemDataMgr died
16-Jul-2018::16:23:37.889 swdev-hng012-red1 confd[31080]: - Daemon uniOperdaemon died
16-Jul-2018::16:23:37.890 swdev-hng012-red1 confd[31080]: - Daemon unidaemon_confd died
16-Jul-2018::16:23:37.891 swdev-hng012-red1 confd[31080]: - Daemon unidaemonactions_confd died

At the CLI prompt

System message at 2018-07-16 16:23:30…
Subsystem stopped: SystemDataMgr

System message at 2018-07-16 16:23:30…
Subsystem stopped: uniOperdaemon
swdev#
System message at 2018-07-16 16:23:30…
Subsystem stopped: unidaemon_confd
swdev# swdev-hng012-red1#
System message at 2018-07-16 16:23:32…
Subsystem started: unidaemon_confd
swdev#
System message at 2018-07-16 16:23:32…
Subsystem started: SystemDataMgr
swdev#
System message at 2018-07-16 16:23:32…
Subsystem started: uniOperdaemon
swdev#
System message at 2018-07-16 16:23:32…
Subsystem stopped: SystemDataMgr
swdev#
System message at 2018-07-16 16:23:32…
Subsystem stopped: uniOperdaemon

The clientTimeout is set to PT120S and the resume is received much before 120 secs.

Not sure what is happening.


#2

Did the caller of cdb_diff_iterate() call cdb_diff_iterate_resume() to continue/finish the iteration after the iterator function returned ITER_SUSPEND?


#3

yes. the caller called cdb_diff_iterate_resume after iterator function returned ITER_SUSPEND?.


#4

Great, so the CDB subscriber seems to be doing fine. The logs entires you posted are for data providers that handle callpoints that disconnect / reconnect. Not for CDB subscribers.

What makes you think that you have a “ConfD” restart on suspend/resume operaton?