No response after heartbeat startup

Will be heartbeat After applying to online , start-up service heartbeat start, There was no response after discovery , Check the log tail -f /var/log/ha-log as follows ：

heartbeat[30680]: 2015/01/27_18:04:29 info: Version 2 support: false heartbeat[30680]: 2015/01/27_18:04:29 WARN: Logging daemon is disabled --enabling logging daemon is recommended heartbeat[30680]: 2015/01/27_18:04:29 info: ************************** heartbeat[30680]: 2015/01/27_18:04:29 info: Configuration validated. Starting heartbeat 2.1.3 heartbeat[30681]: 2015/01/27_18:04:29 info: heartbeat: version 2.1.3 heartbeat[30681]: 2015/01/27_18:04:29 WARN: No Previous generation - starting at 1422353070 heartbeat[30681]: 2015/01/27_18:04:29 info: Heartbeat generation: 1422353070 heartbeat[30681]: 2015/01/27_18:04:29 info: No uuid found for current node - generating a new uuid. heartbeat[30681]: 2015/01/27_18:04:29 info: Creating FIFO /var/lib/heartbeat/fifo.

Always stay here ,heartbeat The managed resources are not started .

At first, there was no clue , Even if viewed debug Logs are useless , Where else will there be error messages ？

Finally thought of service hearbeat status, Finally, I have a hint ：

Cannot shmget for process status: Invalid argument

Along this line Baidu , It is found that it is due to kernel.shmmax, It is caused by kernel parameters . Because this machine is an old one , Version is 5.5, And the newly configured 6.5 The server There is no such problem .

My machine shmmax The data setting is too large , Thus, the creation of shared memory fails .

shmmax Kernel parameters define the maximum value of a single shared memory segment , It is generally recommended to use half of the physical memory , The unit is byte .

With 2G For example ：2048/2*1024*1024=1073741824

stay /etc/sysctl.conf Add kernel.shmmax=1073741824, It still takes effect after the system is restarted .

After a change ,service heartbeat start, Still not started , Again service hearbeat status Report errors

Cannot shmget for process status: No space left on device

This is because shmall The shared memory setting is too large, resulting in

shmall The kernel parameter defines the number of shared memory pages , It is generally recommended to set the physical memory size , The unit is page ,1 page =4k

With 2G For example ：2048*1024*1024/4000= 536870

stay /etc/sysctl.conf Add kernel.shmall=536870, It still takes effect after the system is restarted .

After all the above changes , Again service heartbeat start, Start normally and take over smoothly VIP.

ok, I hope to help you .