[ClusterLabs] Corosync main process was not scheduled

[Mr.R]

Hi all，
   
In the HA cluster built by corosync+pacemaker, the following log appears on host01:



Jan 03 03:38:41 [2095] host01 corosync warning [MAIN  ] Corosync main process was not scheduled (@1704224321984) for 6552.6865 ms (threshold is 4800.0000 ms). Consider token timeout increase.
 
Jan 03 03:38:41 [2095] host01 corosync notice  [TOTEM ] Token has not been received in 0 ms
 
Jan 03 03:38:41 [2095] host01 corosync notice  [TOTEM ] A processor failed, forming new configuration.
 
Jan 03 03:38:41 [2095] host01 corosync notice  [TOTEM ] A new membership (1.9740) was formed. Members
 
Jan 03 03:38:41 [2095] host01 corosync notice  [QUORUM] Members[2]: 1 2
 
Jan 03 03:38:41 [2095] host01 corosync notice  [MAIN  ] Completed service synchronization, ready to provide service.
 
Jan 03 03:39:48 [2095] host01 corosync warning [MAIN  ] Corosync main process was not scheduled (@1704224388160) for 7891.4028 ms (threshold is 4800.0000 ms). Consider token timeout increase.
 
Jan 03 03:39:48 [2095] host01 corosync notice  [TOTEM ] Token has not been received in 0 ms
 
Jan 03 03:39:48 [2095] host01 corosync notice  [TOTEM ] A new membership (1.9744) was formed. Members
 
Jan 03 03:39:48 [2095] host01 corosync notice  [QUORUM] Members[2]: 1 2
 
Jan 03 03:39:48 [2095] host01 corosync notice  [MAIN  ] Completed service synchronization, ready to provide service.
 
Jan 03 03:40:41 [2095] host01 corosync warning [MAIN  ] Corosync main process was not scheduled (@1704224441975) for 6544.5117 ms (threshold is 4800.0000 ms). Consider token timeout increase.
 
Jan 03 03:40:41 [2095] host01 corosync notice  [TOTEM ] Token has not been received in 0 ms
 
Jan 03 03:40:41 [2095] host01 corosync notice  [TOTEM ] A new membership (1.9748) was formed. Members
 
Jan 03 03:40:41 [2095] host01 corosync notice  [QUORUM] Members[2]: 1 2
 
Jan 03 03:40:41 [2095] host01 corosync notice  [MAIN  ] Completed service synchronization, ready to provide service.
 
Jan 03 03:43:41 [2095] host01 corosync info    [KNET  ] link: host: 2 link: 0 is down
 
Jan 03 03:43:41 [2095] host01 corosync info    [KNET  ] link: host: 2 link: 1 is down
 
Jan 03 03:43:41 [2095] host01 corosync info    [KNET  ] host: host: 2 (passive) best link: 0 (pri: 1)
 
Jan 03 03:43:41 [2095] host01 corosync warning [KNET  ] host: host: 2 has no active links
 
Jan 03 03:43:41 [2095] host01 corosync info    [KNET  ] host: host: 2 (passive) best link: 0 (pri: 1)
 
Jan 03 03:43:41 [2095] host01 corosync warning [KNET  ] host: host: 2 has no active links
 
Jan 03 03:43:45 [2095] host01 corosync info    [KNET  ] rx: host: 2 link: 0 is up
 
Jan 03 03:43:45 [2095] host01 corosync info    [KNET  ] rx: host: 2 link: 1 is up
 
Jan 03 03:43:45 [2095] host01 corosync info    [KNET  ] host: host: 2 (passive) best link: 0 (pri: 1)
 
Jan 03 03:43:45 [2095] host01 corosync info    [KNET  ] host: host: 2 (passive) best link: 0 (pri: 1)
 
Jan 03 03:47:46 [2095] host01 corosync notice  [TOTEM ] Token has not been received in 216 ms
 
Jan 03 03:47:48 [2095] host01 corosync notice  [TOTEM ] A new membership (1.974c) was formed. Members



1. According to the corosync log of host01 node, a large number of unscheduled logs occur. 
What is the cause of unscheduled logs?


2. In addition, link down and up appear successively. The doubt is related to the high system load.
 The log after the restart indicates that the corosync process is normal after the restart. 
 Have you ever been in that situation?


3. According to the corosync log of host01 node, a large number of unscheduled logs occur. 
What is the cause of unscheduled logs?


4. How do I check whether corosync is not scheduled because the system load is too high or 
because resources are preempted? Any good suggestions?


thanks,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/users/attachments/20240120/32e6ef51/attachment.htm>