shardmand

shardmand — Postgres Pro Shardman configuration daemon

Postgres Pro Shardman Start-up with shardmand #

shardmand is a Postgres Pro Shardman configuration daemon. It runs on each node in a Shardman cluster, generates a boot_uuid at every startup, subscribes for changes of shardman/cluster0/data/ladle and shardman/cluster0/data/cluster keys in the etcd store (cluster0 is the default cluster name used by Shardman utils), and manages Postgres Pro Shardman processes on the node where it is running according to the configuration described in these JSON documents.

shardmand manages integrated keepers and sentinels. On startup and when one of the monitored etcd keys changes, shardmand reconfigures them as follows:

  • It calculates the expected node configuration, i. e., the list of keepers and sentinels expected to run and their configurations, from the shardman/cluster0/data/ladle and shardman/cluster0/data/cluster values.

  • It receives the list of running keepers and sentinels with their configurations from the internal process manager.

  • It stops processes that are not expected to run. This can be a process that belongs to a cluster with the same name, but a different UUID, or a process whose description is no longer present in the expected node configuration. For keeper processes, shardmand purges their data directory.

  • If a process should be running, but its settings are different from the expected ones, shardmand updates the configuration and restarts the process. If a process should be running, but it is not running, shardmand starts it.

Also, a separate thread of shardmand periodically updates the shardman/cluster0/data/shardmand/NODENAME etcd key with the ClusterUUID of the last cluster to which the configuration was applied. So, before the shardmanctl nodes add command tries to initialize new clusters for a clover, the command can ensure that no alive threads from a previous cluster configuration are left on all nodes in the clover.

Additionaly, shardmand starts two http servers in separate threads. If servers ports match, a single server running both roles is started. The first server provides following metrics: shardmand_etcd_unavailable_time_seconds, shardmand_healthy_keepers, shardmand_sentinels, shardmand_uptime, shardmand_etcd_errors_total, shardmand_reconfigurations_number_total, shardmand_demotions_number_total. Also server provides a /healthz endpoint for shardmand health-check. The second server provides the following endponts:

  • /shardmand/v1/replica — returns 200 status code if a secondary instance is running on node, 500 status code if a primary instance is running on node, /shardmand/v1/master — returns 200 status code if a primary instance is running on node, 500 status code if a secondary instance is running on node, /shardmand/v1/referee — returns 200 status code if the only instance running on node is referee, 500 status code if the only instance running on node is primary or secondary, 404 if more than one instance is running. If node both primary and secondary instances are running on node /shardmand/v1/replica and shardmand/v1/master endpoints return 404 status code.

  • /shardmand/v1/status — getting information about shardmand status, including boot_uuid.

All Postgres Pro Shardman services are managed by shardmand@cluster0.service, so when it is started, stopped, or restarted, it also starts, stops, or restarts all other Postgres Pro Shardman processes (including DBMS instances).

shardmand Syntax #

shardmand [common_options] [ --system-bus ] [ --user user_name ]

Here common_options are:

[ --cluster-name cluster_name ] [ --log-level error | warn | info | debug ] [ --retries retries_number ] [ --session-timeout seconds ] [ --store-endpoints store_endpoints ] [ --store-ca-file store_ca_file ] [ --store-cert-file store_cert_file ] [ --store-key client_private_key ] [ --store-timeout duration ] [ --version ] [ -h | --help ] [ --log-format ]

Command-line Reference #

This refsection describes shardmand-specific command-line options.

--log-format #

Specifies the log output format, json or text. The default is text.

--system-bus #

Not used. Left for compatibility. Ignored.

--user user_name #

Not used. Left for compatibility. Ignored.

Common Options #

shardmand common options are optional parameters that are not specific to the utility. They specify etcd connection settings, cluster name and a few more settings. By default shardmand tries to connect to the etcd store 127.0.0.1:2379 and use the cluster0 cluster name. The default log level is info.

-h, --help #

Show brief usage information.

--cluster-name cluster_name #

Specifies the name for a cluster to operate on. The default is cluster0.

--log-level level #

Specifies the log verbosity. Possible values of level are (from minimum to maximum): error, warn, info and debug. The default is info.

--retries number #

Specifies how many times shardmanctl retries a failing etcd request. If an etcd request fails, most likely, due to a connectivity issue, shardmanctl retries it the specified number of times before reporting an error. The default is 5.

--session-timeout seconds #

Specifies the session timeout for shardmanctl locks. If there is no connectivity between shardmanctl and the etcd store for the specified number of seconds, the lock is released. The default is 30.

--store-endpoints string #

Specifies the etcd address in the format: http[s]://address[:port](,http[s]://address[:port])*. The default is http://127.0.0.1:2379.

--store-ca-file string #

Verify the certificate of the HTTPS-enabled etcd store server using this CA bundle.

--store-cert-file string #

Specifies the certificate file for client identification by the etcd store.

--store-key string #

Specifies the private key file for client identification by the etcd store.

--store-timeout duration #

Specifies the timeout for a etcd request. The default is 5 seconds.

--monitor-port number #

Specifies the port for the shardmand http server for metrics and probes. The default is 15432.

--api-port number #

Specifies the port for the shardmand http api server. The default is 15432.

--version #

Show shardman-utils version information.

shardmand Environment #

A shardmand service reads the environment from /etc/shardman/shardmand-cluster0.env. The following environment variables affect the behavior of shardmand.

SDM_CLUSTER_NAME #

An alternative to setting the --cluster-name option

SDM_LOG_LEVEL #

An alternative to setting the --log-level option

SDM_RETRIES #

An alternative to setting the --retries option

SDM_SYSTEM_BUS #

An alternative to setting the --system-bus option

SDM_STORE_ENDPOINTS #

An alternative to setting the --store-endpoints option

SDM_STORE_CA_FILE #

An alternative to setting the --store-ca-file option

SDM_STORE_CERT_FILE #

An alternative to setting the --store-cert-file option

SDM_STORE_KEY #

An alternative to setting the --store-key option

SDM_STORE_TIMEOUT #

An alternative to setting the --store-timeout option

SDM_SESSION_TIMEOUT #

An alternative to setting the --session-timeout option

SDM_USER #

An alternative to setting the --user option