Comment 6 for bug 1960036

Revision history for this message
Jason Grammenos (jason.grammenos.agility) wrote :

version tested: 2.0.19-2ubuntu0.2

tested keepalived ability to regain quorum after the tracked process (with high numbers pid) restarts.

Based on my testing the version tested fixes the reported issue. I do not know if it introduces new bugs. I only tested to see if it fixes the reported issue.
Note: after downgrading back to 2.0.19-2ubuntu0.1 the issue re appears.

testing output

```
pre patch

----

pp01
stop haproxy service
Feb 16 11:30:26 pp01 systemd[1]: Stopping HAProxy Load Balancer...
Feb 16 11:30:26 pp01 systemd[1]: haproxy.service: Succeeded.
Feb 16 11:30:26 pp01 systemd[1]: Stopped HAProxy Load Balancer.
Feb 16 11:30:27 pp01 Keepalived_vrrp[18854]: Quorum lost for tracked process track_haproxy
Feb 16 11:30:27 pp01 Keepalived_vrrp[18854]: (LB_VIP) Changing effective priority from 111 to 101
Feb 16 11:30:31 pp01 Keepalived_vrrp[18854]: (LB_VIP) Master received advert from 10.4.150.182 with higher priority 110, ours 101
Feb 16 11:30:31 pp01 Keepalived_vrrp[18854]: (LB_VIP) Entering BACKUP STATE
Feb 16 11:30:54 pp01 systemd[1]: Starting HAProxy Load Balancer...
Feb 16 11:30:54 pp01 systemd[1]: Started HAProxy Load Balancer.
start haproxy service
-- nothing
restart haproxy service
eb 16 11:32:15 pp01 systemd[1]: Stopping Keepalive Daemon (LVS and VRRP)...
Feb 16 11:32:16 pp01 Keepalived_vrrp[18854]: Released DBus
Feb 16 11:32:16 pp01 Keepalived_vrrp[18854]: Stopped
Feb 16 11:32:16 pp01 Keepalived[18853]: Stopped Keepalived v2.0.19 (10/19,2019)
Feb 16 11:32:16 pp01 systemd[1]: keepalived.service: Succeeded.
Feb 16 11:32:16 pp01 systemd[1]: Stopped Keepalive Daemon (LVS and VRRP).
Feb 16 11:32:16 pp01 systemd[1]: Started Keepalive Daemon (LVS and VRRP).
Feb 16 11:32:16 pp01 Keepalived[511543]: Starting Keepalived v2.0.19 (10/19,2019)
Feb 16 11:32:16 pp01 Keepalived[511543]: Running on Linux 5.11.0-1028-aws #31~20.04.1-Ubuntu SMP Fri Jan 14 14:37:50 UTC 2022 (built for Linux 5.4.151)
Feb 16 11:32:16 pp01 Keepalived[511543]: Command line: '/usr/sbin/keepalived' '--dont-fork'
Feb 16 11:32:16 pp01 Keepalived[511543]: Opening file '/etc/keepalived/keepalived.conf'.
Feb 16 11:32:16 pp01 Keepalived[511543]: Starting VRRP child process, pid=511544
Feb 16 11:32:16 pp01 Keepalived_vrrp[511544]: Registering Kernel netlink reflector
Feb 16 11:32:16 pp01 Keepalived_vrrp[511544]: Registering Kernel netlink command channel
Feb 16 11:32:16 pp01 Keepalived_vrrp[511544]: Opening file '/etc/keepalived/keepalived.conf'.
Feb 16 11:32:16 pp01 Keepalived_vrrp[511544]: (LB_VIP) Changing effective priority from 101 to 111
Feb 16 11:32:16 pp01 Keepalived_vrrp[511544]: Registering gratuitous ARP shared channel
Feb 16 11:32:16 pp01 Keepalived_vrrp[511544]: (LB_VIP) Entering BACKUP STATE (init)
Feb 16 11:32:16 pp01 Keepalived_vrrp[511544]: Acquired DBus bus org.keepalived.Vrrp1
Feb 16 11:32:16 pp01 Keepalived_vrrp[511544]: Acquired the name org.keepalived.Vrrp1 on the session bus
Feb 16 11:32:17 pp01 Keepalived_vrrp[511544]: (LB_VIP) received lower priority (110) advert from 10.4.150.182 - discarding
Feb 16 11:32:20 pp01 Keepalived_vrrp[511544]: message repeated 3 times: [ (LB_VIP) received lower priority (110) advert from 10.4.150.182 - discarding]
Feb 16 11:32:20 pp01 Keepalived_vrrp[511544]: (LB_VIP) Entering MASTER STATE

pp02
stop haproxy service
Feb 16 11:30:28 pp02 Keepalived_vrrp[399]: (LB_VIP) received lower priority (101) advert from 10.4.150.181 - discarding
Feb 16 11:30:30 pp02 Keepalived_vrrp[399]: message repeated 2 times: [ (LB_VIP) received lower priority (101) advert from 10.4.150.181 - discarding]
Feb 16 11:30:31 pp02 Keepalived_vrrp[399]: (LB_VIP) Entering MASTER STATE
start haproxy service
-- nothing
restart haproxy service
Feb 16 11:32:20 pp02 Keepalived_vrrp[399]: (LB_VIP) Master received advert from 10.4.150.181 with higher priority 111, ours 110
Feb 16 11:32:20 pp02 Keepalived_vrrp[399]: (LB_VIP) Entering BACKUP STATE

-----
post patch

pp01

stop haproxy service
eb 16 11:35:03 pp01 systemd[1]: Stopping HAProxy Load Balancer...
Feb 16 11:35:03 pp01 systemd[1]: haproxy.service: Succeeded.
Feb 16 11:35:03 pp01 systemd[1]: Stopped HAProxy Load Balancer.
Feb 16 11:35:04 pp01 Keepalived_vrrp[511912]: Quorum lost for tracked process track_haproxy
Feb 16 11:35:04 pp01 Keepalived_vrrp[511912]: (LB_VIP) Changing effective priority from 111 to 101
Feb 16 11:35:07 pp01 Keepalived_vrrp[511912]: (LB_VIP) Master received advert from 10.4.150.182 with higher priority 110, ours 101
Feb 16 11:35:07 pp01 Keepalived_vrrp[511912]: (LB_VIP) Entering BACKUP STATE
start haproxy service
Feb 16 11:35:45 pp01 systemd[1]: Starting HAProxy Load Balancer...
Feb 16 11:35:45 pp01 Keepalived_vrrp[511912]: Quorum gained for tracked process track_haproxy
Feb 16 11:35:45 pp01 Keepalived_vrrp[511912]: (LB_VIP) Changing effective priority from 101 to 111
Feb 16 11:35:46 pp01 systemd[1]: Started HAProxy Load Balancer.
Feb 16 11:35:46 pp01 Keepalived_vrrp[511912]: (LB_VIP) received lower priority (110) advert from 10.4.150.182 - discarding
Feb 16 11:35:48 pp01 Keepalived_vrrp[511912]: message repeated 2 times: [ (LB_VIP) received lower priority (110) advert from 10.4.150.182 - discarding]
Feb 16 11:35:49 pp01 Keepalived_vrrp[511912]: (LB_VIP) Entering MASTER STATE

pp02
stop haproxy service
Feb 16 11:35:05 pp02 Keepalived_vrrp[42938]: (LB_VIP) received lower priority (101) advert from 10.4.150.181 - discarding
Feb 16 11:35:07 pp02 Keepalived_vrrp[42938]: message repeated 2 times: [ (LB_VIP) received lower priority (101) advert from 10.4.150.181 - discarding]
Feb 16 11:35:07 pp02 Keepalived_vrrp[42938]: (LB_VIP) Entering MASTER STATE
start haproxy service
Feb 16 11:35:49 pp02 Keepalived_vrrp[42938]: (LB_VIP) Master received advert from 10.4.150.181 with higher priority 111, ours 110
Feb 16 11:35:49 pp02 Keepalived_vrrp[42938]: (LB_VIP) Entering BACKUP STATE
```