Calendar

[jquimi]

Hello.

I'll appreciate if someone can give me a hand with the following case I'm facing.

For some reason the ldirectord detects the server down and sets weights to 0. And after few seconds detects the server is up and sets weights to 1. The follow lines are taken in less than 5 seconds.

[[email protected] conf]# ipvsadm -S
-A -t ams-vip:8817 -s wlc
-a -t ams-vip:8817 -r p13-ams01:8817 -g -w 0
-a -t ams-vip:8817 -r p13-ams02:8817 -g -w 1
-A -t ams-vip:8818 -s wlc
-a -t ams-vip:8818 -r p13-ams01:8818 -g -w 0
-a -t ams-vip:8818 -r p13-ams02:8818 -g -w 1
-A -t ams-vip:8808 -s wlc
-a -t ams-vip:8808 -r p13-ams01:8808 -g -w 0
-a -t ams-vip:8808 -r p13-ams02:8808 -g -w 1
-A -t ams-vip:8807 -s wlc
-a -t ams-vip:8807 -r p13-ams01:8807 -g -w 0
-a -t ams-vip:8807 -r p13-ams02:8807 -g -w 1
[[email protected] conf]#
[[email protected] conf]# ipvsadm -S
-A -t ams-vip:8817 -s wlc
-a -t ams-vip:8817 -r p13-ams01:8817 -g -w 1
-a -t ams-vip:8817 -r p13-ams02:8817 -g -w 1
-A -t ams-vip:8818 -s wlc
-a -t ams-vip:8818 -r p13-ams01:8818 -g -w 1
-a -t ams-vip:8818 -r p13-ams02:8818 -g -w 1
-A -t ams-vip:8808 -s wlc
-a -t ams-vip:8808 -r p13-ams01:8808 -g -w 1
-a -t ams-vip:8808 -r p13-ams02:8808 -g -w 1
-A -t ams-vip:8807 -s wlc
-a -t ams-vip:8807 -r p13-ams01:8807 -g -w 1
-a -t ams-vip:8807 -r p13-ams02:8807 -g -w 1
[[email protected] conf]#

In ldirectord log I found:

[Fri Apr 26 11:11:48 2019|ldirectord.cf|28588] Quiescent real server: 10.124.145.134:8818 (10.124.145.152:8818) (Weight set to 0)
[Fri Apr 26 11:11:48 2019|ldirectord.cf|19984] Quiescent real server: 10.124.145.134:8807 (10.124.145.152:8807) (Weight set to 0)
[Fri Apr 26 11:11:48 2019|ldirectord.cf|9716] Quiescent real server: 10.124.145.134:8808 (10.124.145.152:8808) (Weight set to 0)
[Fri Apr 26 11:11:48 2019|ldirectord.cf|9714] Quiescent real server: 10.124.145.134:8817 (10.124.145.152:8817) (Weight set to 0)
[Fri Apr 26 11:11:52 2019|ldirectord.cf|13470] Invoking ldirectord invoked as: /usr/sbin/ldirectord /etc/ha.d/conf/ldirectord.cf status
[Fri Apr 26 11:11:52 2019|ldirectord.cf|13470] ldirectord for /etc/ha.d/conf/ldirectord.cf is running with pid: 29947
[Fri Apr 26 11:11:52 2019|ldirectord.cf|13470] Exiting from ldirectord status
[Fri Apr 26 11:11:54 2019|ldirectord.cf|28588] Restored real server: 10.124.145.134:8818 (10.124.145.152:8818) (Weight set to 1)
[Fri Apr 26 11:11:54 2019|ldirectord.cf|19984] Restored real server: 10.124.145.134:8807 (10.124.145.152:8807) (Weight set to 1)
[Fri Apr 26 11:11:54 2019|ldirectord.cf|9716] Restored real server: 10.124.145.134:8808 (10.124.145.152:8808) (Weight set to 1)
[Fri Apr 26 11:11:54 2019|ldirectord.cf|9714] Restored real server: 10.124.145.134:8817 (10.124.145.152:8817) (Weight set to 1)

I have a cluster with two servers and services distributed in both of them (first log). However the server never goes down, I mean, all the time is operative. I didn't find a reason for this behaviour of ldirectord.

Somebody can help me with the reason for this behaviour? I've not restarted the system yet.

Thanks in advance for any help.

Kimi