Use impala-shell to connect to Impala

-B or --delimited

write_delimited=true

Prints query results in plain format as a delimited text file. It is useful for creating data files to be used with other Hadoop components. It also helps to avoid the performance overhead of the pretty-printing output, when running benchmark tests that use queries returning large result sets. To specify the delimiter character, use the --output_delimiter option. To define a file name, utilize the -o or --output_file option

-b or --kerberos_host_fqdn

kerberos_host_fqdn=load-balancer-hostname

Overrides the hostname of the Impala daemon’s Kerberos service principal. impala-shell checks whether the server’s principal matches this hostname. It can be used when impalad is accessed via a load-balancer, but impala-shell should connect to a specific impalad directly

--print_header

print_header=true

Determines whether the header row should be printed

-o filename or --output_file filename

output_file=filename

Specifies an output file name. It is typically used to store the results of a single query executed from the command line with the -q option. It also works for interactive sessions — the messages such as number of rows fetched are displayed but not the actual result set. To suppress these messages (when the -q and -o options are combined), redirect stderr to /dev/null

--output_delimiter=character

output_delimiter=character

Specifies the character to use as a delimiter between fields when query results are printed in plain format (the -B option). The default delimiter is tab (\t). If an output value contains the delimiter character, that field can be quoted, or escaped by doubling quotation marks, or both

-E or --vertical

vertical=true

Prints the output of a query (rows) vertically. This option has no effects if -B is used

-p or --show_profiles

show_profiles=true

Displays the query execution plan (the same output as the EXPLAIN statement) and more detailed execution steps

-h or --help

 — 

Displays help information

 — 

history_max=1000

Sets the maximum number of queries executed in impala-shell to be stored to the history file

-i hostname or

--impalad=hostname[:port_num]

impalad=hostname[:port_num]

Connects to the impalad daemon installed on the specified host. If the port_number parameter is not defined, the default value 21050 is used. This option allows you to connect to any host in the cluster. If you access an impalad instance that is running on a different port (the fe_port parameter is changed), add this port to the option parameter

-q query or --query=query

query=query

Allows you to issue a single query from the command line, without starting the interactive interpreter. You can use this option to run impala-shell from a shell script or when the command is invoked from a Python, Perl, or other scripts. The query should contain a single statement, for example, SELECT, CREATE TABLE, SHOW TABLES, or any other statement recognized by impala-shell. Since you cannot pass the USE statement and another query, fully qualify the names for any table that does not belong to the default database. Alternatively, use the -f option to pass a file with the USE statement followed by other queries

-f query_file or

--query_file=query_file

query_file=path_to_query_file

Allows you to process a file containing multiple SQL statements. For example, a set of DDL statements to create a group of tables and views. Multiple statements must be delimited by the semicolon (;). You can specify a filename as - to represent the standard input. This feature allows you to use impala-shell as part of a Unix pipeline where SQL statements are generated dynamically by other tools

--query_option="option=value" or -Q "option=value"

The header line — [impala.query_options] followed by subsequent lines option=value, one option per line

Sets default query options for impala-shell commands. To set multiple query options at once, use multiple instances of this command-line option. The query option names are not case-sensitive

-k or --kerberos

use_kerberos=true

Determines whether the Kerberos authentication should be used when the shell connects to impalad. If Kerberos is not enabled on the instance of impalad to which impala-shell is connecting, an error is displayed. See Enabling Kerberos Authentication for Impala

-s kerberos_service_name or

--kerberos_service_name=name

kerberos_service_name=name

Specifies the impala-shell name to authenticate to a particular impalad service principal. If kerberos_service_name is not specified, impala is used by default. If this option is used for a connection in which Kerberos is not supported, an error is returned

-V or --verbose

verbose=true

Enables the verbose output

--quiet

verbose=false

Disables the verbose output

-v or --version

version=true

Displays the version information of impala-shell

-c

ignore_query_failure=true

Determines whether to ignore the query failure

-d default_db or

--database=default_db

default_db=default_db

Specifies the initial database. The same as running the USE statement after the connection. If not specified, the DEFAULT database is used

-ssl

ssl=true

Enables TLS/SSL for impala-shell

--ca_cert=path_to_certificate

ca_cert=path_to_certificate

Specifies the local path to the third-party CA certificate, or to a copy of the server certificate for self-signed server certificates. If --ca_cert is not set, impala-shell enables TLS/SSL, but does not validate the server certificate. This is useful for connecting to a trusted Impala that is only running over TLS/SSL, when a copy of the certificate is not available, for example, when debugging customer installations

-l

use_ldap=true

Enables LDAP authentication

-u

user=user_name

Specifies the username, when LDAP authentication is enabled by the -l option. Utilize the short username, not the full LDAP distinguished name. After using this option, the shell prompts for a password

--ldap_password_cmd=command

 — 

Specifies a command to retrieve the LDAP password, when the LDAP authentication is enabled by the -l option. If the command includes space-separated arguments, enclose the command and its arguments in quotation marks

--config_file=path_to_config_file

 — 

Specifies the path of the file containing impala-shell configuration settings. The default is /etc/impalarc. This setting can only be specified on the command line

--live_progress

live_progress=true

Displays a progress bar showing the approximate percentage of completion for each query. The information is updated interactively as the query is processed. See LIVE_PROGRESS Query Option

--disable_live_progress

live_progress=false

Allows you to disable live_progress in the interactive mode

--live_summary

 — 

Shows a detailed report, similar to the SUMMARY command, that shows progress details for each phase of the query execution. The information is updated interactively as the query is processed. See LIVE_PROGRESS Query Option

-var=variable_name=value

 — 

Defines a substitution variable that can be used within the impala-shell session. The variable can be substituted into statements processed by the -q or -f options, or in an interactive shell session. Within a SQL statement, use the following syntax: ${var:variable_name}

--auth_creds_ok_in_clear

 — 

Allows you to use LDAP authentication for an insecure connection to the shell. Note that authentication credentials are sent unencrypted, and may be vulnerable to attack

--protocol=protocol

 — 

--hs2_fp_format=HS2_FP_FORMAT

hs2_fp_format=HS2_FP_FORMAT

Sets the printing format specification for floating point values when using HS2 protocol. The default behavior makes the values handled by Python’s str() built-in method. Use 16G to match Beeswax protocol’s floating-point output format