Postgresql FATAL: could not create semaphores: No space left on device

转自：http://blog.163.com/dazuiba_008/blog/static/363349812016314739538/

  今天做恢复的时候，数据库做完恢复后，无法启动报错

FATAL:  could not create semaphores: No space left on device

DETAIL:  Failed system call was semget(xxxxxxxxxx).

HINT:  This error does *not* mean that you have run out of disk space.  It occurs when either the system limit for the maximum number of semaphore sets (SEMMNI), or the system wide maximum number of semaphores (SEMMNS), would be exceeded.  You need to raise the respective kernel parameter.  Alternatively, reduce PostgreSQL's consumption of semaphores by reducing its max_connections parameter.

看提示是空间不足，其实不是空间不足，是信号量不足

这里是共享内存段的限制，没有报错，简单介绍一下

如果数据库报FATAL: could not create shared memory segment:Cannot allocate memory 错误，可以考虑修改以下相关参数

#ipcs -lm

------ Shared Memory Limits --------

max number of segments = 4096

max seg size (kbytes) = 67108864

max total shared memory (kbytes) = 17179869184

min seg size (bytes) = 1

 #cat  /proc/sys/kernel/shmmax

68719476736

SHMMAX 单个共享内存段最大字节数

# cat  /proc/sys/kernel/shmmni

4096

SHMMNI 共享内存段最大个数 

# cat  /proc/sys/kernel/shmall

4294967296

SHMALL系统中共享内存也总数，至少为ceil(shmmax/PAGE_SIZE)

获取page_size值

# getconf PAGE_SIZE

4096

可以根据实际情况修改以上参数值，在/etc/sysctl.conf配置文件中

今天遇到的错误，主要解决以下参数的限制才能解决我们数据库启动的报错

# ipcs -ls

------ Semaphore Limits --------

max number of arrays = 1280

max semaphores per array = 50100

max semaphores system wide = 64128000

max ops per semop call = 50100

semaphore max value = 32767

# cat  /proc/sys/kernel/sem

SEMMSL   SEMMNS         SEMOPM  SEMMNI

50100   128256000       50100   2560

SEMMSL 每个信号量set中信号量最大个数

SEMMNS linux系统中信号量最大个数

SEMOPM semop系统调用允许的信号量最大个数设置，设置成和SEMMSL一样即可

SEMMNI  linux系统信号量set最大个数

所以SEMMNS=SEMMSL*SEMMNI

所以要么增大信号量，要么减少max_connect参数

这里我选择增大信号量

修改 vi /etc/sysctl.conf 的以下参数

kernel.sem = 50100 128256000 50100 2560

修改完成后，执行：

# sysctl -p 

使修改参数生效

ipcs -ls

------ Semaphore Limits --------

max number of arrays = 2560

max semaphores per array = 50100

max semaphores system wide = 128256000

max ops per semop call = 50100

semaphore max value = 32767

重新启动数据库并无报错

Name	Description	Reasonable values
SHMMAX	Maximum size of shared memory segment (bytes)	at least 1kB (more if running many copies of the server)
SHMMIN	Minimum size of shared memory segment (bytes)	1
SHMALL	Total amount of shared memory available (bytes or pages)	if bytes, same as SHMMAX; if pages, ceil(SHMMAX/PAGE_SIZE)
SHMSEG	Maximum number of shared memory segments per process	only 1 segment is needed, but the default is much higher
SHMMNI	Maximum number of shared memory segments system-wide	like SHMSEG plus room for other applications
SEMMNI	Maximum number of semaphore identifiers (i.e., sets)	at least ceil((max_connections + autovacuum_max_workers + max_worker_processes + 5) / 16)
SEMMNS	Maximum number of semaphores system-wide	ceil((max_connections + autovacuum_max_workers + max_worker_processes + 5) / 16) * 17 plus room for other applications
SEMMSL	Maximum number of semaphores per set	at least 17
SEMMAP	Number of entries in semaphore map	see text
SEMVMX	Maximum value of semaphore	at least 1000 (The default is often 32767; do not change unless necessary)