Kernel Resources
This section describes kernel resource parameters. Whether these parameters take effect depends on OS settings.
max_files_per_process
Parameter description: Specifies the maximum number of files that can be opened simultaneously by each server process. If the operating system kernel enforces a reasonable limit, then this parameter does not need to be set.
However, on some platforms (especially most BSD systems), the kernel allows independent processes to open far more files than the system can actually support. If users encounter failures such as "Too many open files", they should try reduce the setting. Typically, the system must meet this requirement: Number of file descriptors ≥ Maximum number of concurrent statements x Number of primary DNs on the current server x max_files_per_process x 3.
Type: POSTMASTER
Value range: an integer ranging from 25 to INT_MAX
Default value: 1000
max_files_per_node
Parameter description: Specifies the maximum number of files that can be opened by a single SQL statement on a single node. Generally, you do not need to set this parameter. This parameter is supported only by clusters of version 8.1.3 or later.
Type: SUSET
Value range: an integer ranging from –1 to INT_MAX. The value –1 indicates that the maximum number is limited.
Default value: –1
- The default value of this parameter is –1 in a new cluster. In an upgrade scenario, the default value of this parameter is retained for forward compatibility.
- If error message "The last file name is [%s] and %d files have already been opened on data node [%s] with a maximum of %d files." is displayed during statement execution, increase the value of max_files_per_node.
enable_fd_check
Parameter description: Specifies whether to perform verification when FD is used. This parameter is supported only in 8.2.1.300 and later versions.
Type: SIGHUP
Value range: Boolean
- on indicates that FD verification is enabled.
- off indicates that FD verification is enabled.
Default value: on
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.