docs/wiki/introduction/sql.md
It may seem weird at first, but try to think of your operating system as a series of tabular concepts. Each concept becomes a SQL table, like processes, or sockets, the filesystem, a host alias, a running kernel module, etc. There are several informational things — like OS version, CPU features, memory details, UEFI platform vendor details — that are not tabular but rather a body of details with labeled data. We can represent this type of data as a table with a single row and many columns, or a series of key/value rows. When you want to inspect a concept, you SELECT the data, and the associated OS APIs are called in real-time.
Now consider event streams: each event is a row, like a new USB device connection, or file attribute modification. These are the same concepts with an 'event-like' twist. We do not inspect event-time data in real-time, but rather buffer the events as they occur and represent that buffer as a table! Concept 'actions' can be represented too, you perform an action and generate tabular data. Consider stat-ing a file, hashing a blob of data, parsing JSON, reading a SQLite database, traversing a directory, or requesting a user's list of installed browser plugins. Actions use primary keys as input and generate rows as output, and are best used when JOINing.
The world of osquery is centered around SQL: decorating, scheduling, differentials, eventing, targeting. Everything is SQL, and hopefully as expressive as possible. Continue reading our deployment and development guides for a deep-dive into how SQL can power intrusion detection, incident response, process auditing, file integrity monitoring and more.
The osquery SQL language is a superset of SQLite's. Please read SQL as understood by SQLite for reference. This is a great starting place if coming from MySQL, PostgreSQL, or MSSQL.
SELECT only! All mutation-based verbs exist, like INSERT, UPDATE, DELETE, and ALTER, but they do nothing -- unless you're fancy and creating run-time tables or VIEWs, or using an extension. Mutation-based verbs are allowed in extensions, if the extension supports them.
NOTICE: Several tables,
filefor example, require a predicate for one of the columns, and will not work without it. See Tables with arguments for more information.
Before diving into osquery's specific implementation of SQL, please familiarize yourself with the osquery development shell. This shell is designed for ad-hoc exploration of your OS and SQL query prototyping. Then, fire up osqueryi as your user or as a superuser, and try some of the concepts below. Know that this 'shell' does not connect to a remote server; it is completely standalone.
Within the shell, try: .help
$ osqueryi
Using a virtual database. Need help, type '.help'
osquery> .help
Welcome to the osquery shell. Please explore your OS!
You are connected to a transient 'in-memory' virtual database.
.all [TABLE] Select all from a table
.bail ON|OFF Stop after hitting an error
.connect PATH Connect to an osquery extension socket
.disconnect Disconnect from a connected extension socket
.echo ON|OFF Turn command echo on or off
[...]
osquery>
Try the meta-commands .tables and .schema to list all of the tables and their schema. The schema meta-command takes an argument that helps limit the output to a partial string match.
osquery> .schema process
[...]
CREATE TABLE process_memory_map(`pid` INTEGER, `start` TEXT, `end` TEXT, `permissions` TEXT, `offset` BIGINT, `device` TEXT, `inode` INTEGER, `path` TEXT, `pseudo` INTEGER, PRIMARY KEY (`pid`)) WITHOUT ROWID;
CREATE TABLE process_open_files(`pid` BIGINT, `fd` BIGINT, `path` TEXT, PRIMARY KEY (`pid`)) WITHOUT ROWID;
CREATE TABLE process_open_sockets(`pid` INTEGER, `fd` BIGINT, `socket` BIGINT, `family` INTEGER, `protocol` INTEGER, `local_address` TEXT, `remote_address` TEXT, `local_port` INTEGER, `remote_port` INTEGER, `path` TEXT, `state` TEXT, `net_namespace` TEXT HIDDEN, PRIMARY KEY (`pid`, `fd`, `socket`, `family`, `protocol`, `local_address`, `remote_address`, `local_port`, `remote_port`, `path`, `state`, `net_namespace`)) WITHOUT ROWID;
CREATE TABLE processes(`pid` BIGINT, `name` TEXT, `path` TEXT, `cmdline` TEXT, `state` TEXT, `cwd` TEXT, `root` TEXT, `uid` BIGINT, `gid` BIGINT, `euid` BIGINT, `egid` BIGINT, `suid` BIGINT, `sgid` BIGINT, `on_disk` INTEGER, `wired_size` BIGINT, `resident_size` BIGINT, `total_size` BIGINT, `user_time` BIGINT, `system_time` BIGINT, `disk_bytes_read` BIGINT, `disk_bytes_written` BIGINT, `start_time` BIGINT, `parent` BIGINT, `pgroup` BIGINT, `threads` INTEGER, `nice` INTEGER, `elevated_token` INTEGER HIDDEN, `secure_process` INTEGER HIDDEN, `protection_type` TEXT HIDDEN, `virtual_process` INTEGER HIDDEN, `elapsed_time` BIGINT HIDDEN, `handle_count` BIGINT HIDDEN, `percent_processor_time` BIGINT HIDDEN, `upid` BIGINT, `uppid` BIGINT, `cpu_type` INTEGER, `cpu_subtype` INTEGER, `translated` INTEGER, `cgroup_path` TEXT HIDDEN, PRIMARY KEY (`pid`)) WITHOUT ROWID;
This complete schema for all supported platforms is available on the homepage. To see schema in your shell for tables foreign to your OS, like kernel modules on macOS, use the --enable_foreign command line flag.
On macOS (or Linux), select 1 process's pid, name, and path. Then change the display mode and issue the same query:
osquery> SELECT pid, name, path FROM processes WHERE pid > 0 LIMIT 1;
+-----+---------+---------------+
| pid | name | path |
+-----+---------+---------------+
| 1 | launchd | /sbin/launchd |
+-----+---------+---------------+
osquery> .mode line
osquery> SELECT pid, name, path FROM processes WHERE pid > 0 LIMIT 1;
pid = 1
name = launchd
path = /sbin/launchd
osquery> .mode pretty
Then try: SELECT pid, name, path FROM processes ORDER BY start_time DESC LIMIT 1; several times and you will continue to select the last-most-recent process to start. This data is equivalent to ps and is a real-time representation of processes.
To really hammer home the real-time representation, try SELECT * FROM time;. Feel free to inspect other concepts/tables. Use .mode line for the best output within smaller terminal views.
Then, let's look at a "meta" table that provides details to osquery about itself. These tables are prefixed with osquery_:
osquery> .mode line
osquery> SELECT * FROM osquery_info;
pid = 15982
uuid = 4892E1C6-F800-5F8E-92B1-BC2216C29D4F
instance_id = 94c004b0-49e5-4ece-93e6-96c1939c0f83
version = 2.4.6
config_hash =
config_valid = 0
extensions = active
build_platform = darwin
build_distro = 10.12
start_time = 1496552549
watcher = -1
This will always show the current PID of the running osquery process, shell or otherwise.
Let's use this to demonstrate JOINing:
osquery> SELECT pid, name, path FROM osquery_info JOIN processes USING (pid);
pid = 15982
name = osqueryi
path = /usr/local/bin/osqueryi
Now let's get fancy and complicated, by performing two JOINs and adding a WHERE clause:
osquery> SELECT p.pid, name, p.path as process_path, pf.path as open_path
...> FROM osquery_info i
...> JOIN processes p ON p.pid = i.pid
...> JOIN process_open_files pf ON pf.pid = p.pid
...> WHERE pf.path LIKE '/dev/%';
pid = 15982
name = osqueryi
process_path = /usr/local/bin/osqueryi
open_path = /dev/ttys000
pid = 15982
name = osqueryi
process_path = /usr/local/bin/osqueryi
open_path = /dev/ttys000
pid = 15982
name = osqueryi
process_path = /usr/local/bin/osqueryi
open_path = /dev/ttys000
pid = 15982
name = osqueryi
process_path = /usr/local/bin/osqueryi
open_path = /dev/null
We can expand upon this later using subqueries and more tables.
Several tables, file for example, require arguments to be queried. Consider SELECT * FROM file: you would not want this to trigger a complete walk of the mounted file systems. The query must be constrained with some sort of argument or input parameter. Tables like this are indicated by a pushpin icon in the schema documentation next to the columns that act as arguments to constrain a query.
As an example of this concept, let's exercise the file table:
osquery> .mode line
osquery> SELECT * FROM file;
osquery> SELECT * FROM file WHERE path = '/dev/zero';
path = /dev/zero
directory = /dev
filename = zero
inode = 304
uid = 0
gid = 0
mode = 0666
device = 50331651
size = 0
block_size = 131072
atime = 1463786341
mtime = 1463786341
ctime = 1463786341
btime = 0
hard_links = 1
type = character
osquery> SELECT count(1) FROM file WHERE path LIKE '/dev/%';
count(1) = 568
The documentation for file says both path and directory can be used as input parameters. In most cases these columns and tables should "do the right thing" and respond to various operators. String data, like paths, are not easily compared so = or <> and LIKE are the only operators that make sense.
Let's get semi-fancy:
osquery> .mode line
osquery> SELECT path, inode, size, type
...> FROM file
...> WHERE path IN (SELECT '/dev/zero');
path = /dev/zero
inode = 304
size = 0
type = character
Now let's introduce the hash table and hopefully show something useful, like the hash of the last file modified in /etc:
osquery> .mode line
osquery> SELECT path, mtime, sha256
...> FROM file
...> JOIN hash USING (path)
...> WHERE file.directory = '/etc'
...> ORDER BY mtime DESC LIMIT 1;
path = /etc/krb5.keytab
mtime = 1464730624
sha256 = e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855
osquery includes various 'additional' SQL functions and aggregations. We try to balance SQL feature requests using a locality question: "does this make sense to execute on a single host, or many?" If the aggregation or function would be better suited with information from a fleet or group of hosts, find a mechanism to perform the function/aggregation after-the-fact.
osquery includes the following C-math functions: sqrt, log, log10, ceil, floor, power, pi.
osquery> .mode line
osquery> select disk_size as disk_size from disk_info;
disk_size = 107372805120
osquery> select sqrt(disk_size) as disk_size from disk_info;
disk_size = 327677.898430761
osquery> select log(disk_size) as disk_size from disk_info;
disk_size = 25.3995727757846
osquery> select log10(disk_size) as disk_size from disk_info;
disk_size = 11.0308942992233
osquery> select ceil(disk_size) as disk_size from disk_info;
disk_size = 107372805120
osquery> select floor(disk_size) as disk_size from disk_info;
disk_size = 107372805120
osquery> select power(disk_size, 2) as disk_size from disk_info;
disk_size = 1.15289192793375e+22
osquery> select pi() * disk_size as disk_size from disk_info;
disk_size = 337321615760.32
The following trig functions: sin, cos, tan, cot, asin, acos, atan, and radians to degrees conversions.
osquery .mode line
osquery> select sin(30);
sin(30) = -0.988031624092862
osquery> select cos(30);
cos(30) = 0.154251449887584
osquery> select tan(30);
tan(30) = -6.40533119664628
osquery> select cot(30);
cot(30) = -0.156119952161659
osquery> select asin(.5);
asin(.5) = 0.523598775598299
osquery> select acos(.5);
acos(.5) = 1.0471975511966
osquery> select atan(.5);
atan(.5) = 0.463647609000806
osquery> select radians(60);
radians(60) = 1.0471975511966
osquery> select degrees(1.3);
degrees(1.3) = 74.484513367007
concat(ARG1, ARG2, ARG3...): Concatenate arguments, ignoring nulls.
osquery> .mode line
osquery> select concat('hello', NULL, ' ', 'world');
concat('hello', NULL, ' ', 'world') = hello world
concat_ws(SEPARATOR, ARG1, ARG2, ARG3...): Concatenate arguments, ignoring nulls, and interleaved with SEPARATOR.
osquery> .mode line
osquery> select concat_ws(' ', 'hello', NULL, 'world');
concat_ws(' ', 'hello', NULL, 'world') = hello world
split(COLUMN, TOKENS, INDEX): split COLUMN using any character token from TOKENS and return the INDEX result. If an INDEX result does not exist, a NULL type is returned.
osquery> .mode line
osquery> select uid from users;
uid = 500
uid = 1001
osquery> select split(uid, 1, 0) from users;
split(uid, 1, 0) = 500
split(uid, 1, 0) = 00
regex_split(COLUMN, PATTERN, INDEX): similar to split, but instead of TOKENS, apply the POSIX regex PATTERN (as interpreted by std::regex).
osquery> .mode line
osquery> select uid from users;
uid = 500
uid = 1001
osquery> select regex_split(uid, "[1-5]", 0) from users;
split(uid, 1, 0) = 00
split(uid, 1, 0) = 00
regex_match(COLUMN, PATTERN, INDEX): Runs regex match across the column, and returns matched subgroups. (The 0 index is the full match, subsequent numbers are the groups).
osquery> .mode line
osquery> select regex_match('hello world. Goodbye', '(\w+) \w+', 0) as m0,
regex_match('hello world. Goodbye', '(\w+) \w+', 1) as m1;
m0 = hello world
m1 = hello
inet_aton(IPv4_STRING): return the integer representation of an IPv4 string.
osquery> .mode line
osquery> select inet_aton("1.0.1.5") as ipInt
ipInt = 16777477
version_compare(LEFT_VERSION, RIGHT_VERSION, COMPARE_FLAVOR): return -1 if the left version is less than the right, 0 if they are equal, or 1 if the left version is greater than the right. COMPARE_FLAVOR is optional, and can be of value (ARCH, DPKG, or RHEL), which equates to the respective linux distribution package versioning. The default COMPARE_FLAVOR is semantic versioning.
osquery> .mode line
osquery> select version_compare('1.0', '1.0');
version_compare('1.0', '1.0') = 0
osquery> select version_compare('4:1.1.0', '4:1.1.0-3', 'ARCH');
version_compare('4:1.1.0', '4:1.1.0-3', 'ARCH') = 0
osquery> select version_compare('50.4.1b', '50.4.1c');
version_compare('50.4.1b', '50.4.1c') = -1
osquery> select version_compare('1.0.0~rc2^2021', '1.0.0', 'RHEL');
version_compare('1.0.0~rc2^2021', '1.0.0', 3) = -1
osquery> select version_compare('1:1.2.13-2', '4.2.1', 'ARCH');
version_compare('1:1.2.13-2', '4.2.1', 1) = 1
We have added sha1, sha256, and md5 functions that take a single argument and return the hashed value.
osquery> .mode line
osquery> select username from users;
username = Guest
username = System
osquery> select sha1(username) as usernameHash from users;
usernameHash = face83ee3014bdc8f98203cc94e2e89222452e90
usernameHash = 29d43743c43bda9873fc7a79c99f2ec4b6b442b1
osquery> select sha256(username) as usernameHash from users;
usernameHash = a835887ac13e6558ea6cb404aae6a35b7cbff6796af813d72f7b8d08f3fa0ec9
usernameHash = 4d2c882abd33183be08ec6f4b47a1f09d3dd211de7556d9b587f7e34eec5ed0b
osquery> select md5(username) as usernameHash from users;
usernameHash = 7d4ef62de50874a4db33e6da3ff79f75
usernameHash = 2a44946d16fe86e63a7e078744c58d56
community_id_v1(SOURCE_ADDR, DEST_ADDR, SOURCE_PORT, DEST_PORT, PROTOCOL, SEED): returns the Community ID v1 hash of the network connection. This can be used to match with the Community ID generated by other tools such as Zeek and Suricata. SEED is optional and will be set to 0 if omitted. If some values are missing or cannot be parsed, the function will return an empty result and log a warning. For strict error checking (resulting in failure of the query), use community_id_v1_strict.
osquery> .mode line
osquery> select community_id_v1('66.35.250.204', '128.232.110.120', 80, 34855, 6) AS community_id;
community_id = 1:LQU9qZlK+B5F3KDmev6m5PMibrg=
osquery> select community_id_v1('66.35.250.204', '2001:0:3238:DFE1:63::FEFB', 80, 2347, 6) AS community_id;
community_id = 1:rxU6O+b2d9kbSWjRmVDoBbowx6g=
osquery> select community_id_v1('66.35.250.204', '2001:0:3238:DFE1:63::FEFB', 80, 2347, 6, 37) AS community_id_with_seed;
community_id_with_seed = 1:jmJ2ORP31di4mtsQPIKzyoEb3yo=
osquery> select community_id_v1(local_address,remote_address,local_port,remote_port,protocol) as community_id from process_open_sockets limit 2;
community_id = 1:PaAbtXl8lgQoYFUShUQwXpcNVfw=
W0129 10:34:47.759569 195012032 sqlite_hashing.cpp:226] Community ID saddr cannot be parsed as IP
community_id =
There are also encoding functions available, to process query results.
to_base64: base64 encode a string.
osquery> .mode line
osquery> select device_id from cpu_info;
device_id = CPU0
osquery> select to_base64(device_id) as device_id from cpu_info;
device_id = Q1BVMA==
from_base64: Decode a base64 encoded string. If the string is not valid base64 an empty string is returned.
osquery> .mode line
osquery> select device_id from cpu_info;
device_id = CPU0
osquery> select to_base64(device_id) as device_id from cpu_info;
device_id = Q1BVMA==
select from_base64(to_base64(device_id)) as device_id from cpu_info;
device_id = CPU0
conditional_to_base64: Encode a string if and only if the string contains non-printable ASCII characters. (Specifically, it considers 0x20 through 0x7e printable)
osquery> .mode line
osquery> select device_id from cpu_info;
device_id = CPU0
osquery> select conditional_to_base64(device_id) as device_id from cpu_info;
device_id = CPU0
osquery> select conditional_to_base64(device_id || char(183)) as device_id from cpu_info;
device_id = Q1BVMMK3
in_cidr_block(CIDR_RANGE, IP_ADDRESS): return 1 if the IP address is within the CIDR block, otherwise 0.
osquery> .mode line
osquery> SELECT in_cidr_block('10.0.0.0/26', '10.0.0.24');
in_cidr_block('10.0.0.0/26', '10.0.0.24') = 1
osquery> SELECT in_cidr_block('2001:db8::/48', '2001:db8:0:ffff:ffff:ffff:ffff:ffff');
in_cidr_block('2001:db8::/48', '2001:db8:0:ffff:ffff:ffff:ffff:ffff') = 1
version:
osquery> .mode line
osquery> SELECT '1.0' = '1.0' COLLATE VERSION;
'1.0' = '1.0' COLLATE VERSION = 1
osquery> SELECT '50.4.1b' < '50.4.1c' COLLATE VERSION;
'50.4.1b' < '50.4.1c' COLLATE VERSION = 1
osquery> SELECT '20.10a' > '20.102' COLLATE VERSION;
'20.10a' > '20.102' COLLATE VERSION = 1
version_arch:
osquery> .mode line
osquery> SELECT '4:2' = '4:2-1' COLLATE VERSION_ARCH;
'4:2' = '4:2-1' COLLATE VERSION_ARCH = 1
osquery> SELECT '2-2pre' < '2-2rc' COLLATE VERSION_ARCH;
'2-2pre' < '2-2rc' COLLATE VERSION_ARCH = 1
osquery> SELECT '42.2-1' > '42.1-2' COLLATE VERSION_ARCH;
'42.2-1' > '42.1-2' COLLATE VERSION_ARCH = 1
version_dpkg:
osquery> .mode line
osquery> SELECT '1:2.0-10' = '1:2.0-10' COLLATE VERSION_DPKG;
'1:2.0-10' = '1:2.0-10' COLLATE VERSION_DPKG = 1
osquery> SELECT '22.07.5-2ubuntu1.3' < '22.07.5-2ubuntu1.4' COLLATE VERSION_DPKG;
'22.07.5-2ubuntu1.3' < '22.07.5-2ubuntu1.4' COLLATE VERSION_DPKG = 1
osquery> SELECT '2:8.2.3995-1ubuntu2.9' > '2:8.2.3995-1ubuntu2.3' COLLATE VERSION_DPKG;
'2:8.2.3995-1ubuntu2.9' > '2:8.2.3995-1ubuntu2.3' COLLATE VERSION_DPKG = 1
version_rhel:
osquery> .mode line
osquery> SELECT '0.5.0~rc1^202' = '0.5.0~rc1^202' COLLATE VERSION_RHEL;
'0.5.0~rc1^202' = '0.5.0~rc1^202' COLLATE VERSION_RHEL = 1
osquery> SELECT '1.1.0~BETA2' < '1.1.0~CR1' COLLATE VERSION_RHEL;
'1.1.0~BETA2' < '1.1.0~CR1' COLLATE VERSION_RHEL = 1
osquery> SELECT '1.0.0' > '1.0.0~rc2' COLLATE VERSION_RHEL;
'1.0.0' > '1.0.0~rc2' COLLATE VERSION_RHEL = 1
Over time it may make sense to rename tables and columns. osquery tries to apply plurals to table names and achieve the easiest foreign key JOIN syntax. This often means slightly skewing concept attributes or biasing towards diction used by POSIX.
osquery makes an effort to mark deprecated tables and create 'clone' VIEWs so that previously scheduled queries continue to work. Similarly, for old column names, the column will be marked HIDDEN and only returned if explicitly selected. This does not make queries using * future-proof, as they will begin using the new column names when the client is updated. All of these changes are considered osquery API changes and marked as such in release notes on GitHub.