ipecho: add ipecho client, server, tile, and shred version commands

Author: mmcgee-jump

book/api/cli.md

@@ -142,6 +142,16 @@ $ fdctl version
 0.101.11814
 ```
 
+## `shred-version`
+Prints the current shred version of the cluster being joined, according
+to the entrypoints, to standard output and exits. The command writes
+diagnostic messages from logs to `stderr`.
+
+```sh [bash]
+$ fdctl shred-version
+9065
+```
+
 ## `set-identity`
 Changes the identity key of a running validator. The `<keypair>`
 argument is required and must be the path to an Agave style

book/api/metrics-generated.md

@@ -899,3 +899,18 @@
 | <span class="metrics-name">snapin_&#8203;accounts_&#8203;inserted</span> | gauge | Number of accounts inserted during snpashot loading. Might decrease if snapshot load is aborted and restarted |
 
 </div>
+
+## Ipecho Tile
+
+<div class="metrics">
+
+| Metric | Type | Description |
+|--------|------|-------------|
+| <span class="metrics-name">ipecho_&#8203;shred_&#8203;version</span> | gauge | The current shred version used by the validator |
+| <span class="metrics-name">ipecho_&#8203;connection_&#8203;count</span> | gauge | The number of active connections to the ipecho service |
+| <span class="metrics-name">ipecho_&#8203;connections_&#8203;closed_&#8203;ok</span> | counter | The number of connections to the ipecho service that have been made and closed normally |
+| <span class="metrics-name">ipecho_&#8203;connections_&#8203;closed_&#8203;error</span> | counter | The number of connections to the ipecho service that have been made and closed abnormally |
+| <span class="metrics-name">ipecho_&#8203;bytes_&#8203;read</span> | counter | The total number of bytes read from all connections to the ipecho service |
+| <span class="metrics-name">ipecho_&#8203;bytes_&#8203;written</span> | counter | The total number of bytes written to all connections to the ipecho service |
+
+</div>

src/app/firedancer-dev/Local.mk

@@ -15,6 +15,7 @@ $(call add-objs,commands/sim,fd_firedancer_dev)
 $(call add-objs,commands/backtest,fd_firedancer_dev)
 $(call add-objs,commands/snapshot_load,fd_firedancer_dev)
 $(call add-objs,commands/repair,fd_firedancer_dev)
+$(call add-objs,commands/ipecho_server,fd_firedancer_dev)
 
 $(call make-bin,firedancer-dev,main,fd_firedancer_dev fd_firedancer fddev_shared fdctl_shared fdctl_platform fd_discof fd_disco fd_choreo fd_flamenco fd_funk fd_quic fd_tls fd_reedsol fd_waltz fd_tango fd_ballet fd_util firedancer_version,$(SECP256K1_LIBS) $(ROCKSDB_LIBS) $(OPENSSL_LIBS))
 

src/app/firedancer-dev/commands/backtest.c

@@ -350,7 +350,7 @@ backtest_cmd_topo( config_t * config ) {
   backtest_topo( config );
 }
 
-args_t
+static args_t
 configure_args( void ) {
   args_t args = {
     .configure.command = CONFIGURE_CMD_INIT,

src/app/firedancer-dev/commands/ipecho_server.c

@@ -0,0 +1,121 @@
+
+#include "../../shared/commands/configure/configure.h"
+#include "../../shared/commands/run/run.h"
+#include "../../../disco/metrics/fd_metrics.h"
+#include "../../../disco/topo/fd_topob.h"
+
+#include <time.h>
+#include <stdio.h>
+#include <unistd.h>
+
+#define NAME "ipecho-server"
+
+extern fd_topo_obj_callbacks_t * CALLBACKS[];
+
+fd_topo_run_tile_t
+fdctl_tile_run( fd_topo_tile_t const * tile );
+
+static void
+ipecho_topo( fd_topo_t *  topo,
+             char const * name ) {
+  fd_topob_new( topo, name );
+  topo->max_page_size = 1UL<<21UL;
+
+  fd_topob_wksp( topo, "all" );
+  fd_topo_link_t * link = fd_topob_link( topo, "ipecho_out", "all", 4UL, 0UL, 1UL );
+  link->permit_no_consumers = 1;
+  fd_topo_tile_t * tile = fd_topob_tile( topo, "ipecho", "all", "all", 0UL, 0, 0 );
+  tile->ipecho.expected_shred_version = 32;
+  tile->ipecho.entrypoints_cnt = 0UL;
+  tile->ipecho.bind_address = FD_IP4_ADDR(127,0,0,1);
+  tile->ipecho.bind_port    = 12008;
+  fd_topob_tile_out( topo, "ipecho", 0UL, "ipecho_out", 0UL );
+
+  fd_topob_auto_layout( topo, 0 );
+  fd_topob_finish( topo, CALLBACKS );
+}
+
+extern int * fd_log_private_shared_lock;
+
+static void
+ipecho_server_cmd_topo( config_t * config ) {
+  ipecho_topo( &config->topo, config->name );
+}
+
+static args_t
+configure_args( void ) {
+  args_t args = {
+    .configure.command = CONFIGURE_CMD_INIT,
+  };
+
+  ulong stage_idx = 0UL;
+  args.configure.stages[ stage_idx++ ] = &fd_cfg_stage_hugetlbfs;
+  args.configure.stages[ stage_idx++ ] = NULL;
+
+  return args;
+}
+
+void
+ipecho_server_cmd_perm( args_t *    args FD_PARAM_UNUSED,
+                   fd_cap_chk_t *   chk,
+                   config_t const * config ) {
+  args_t c_args = configure_args();
+  configure_cmd_perm( &c_args, chk, config );
+  run_cmd_perm( NULL, chk, config );
+}
+
+static void
+ipecho_server_cmd_fn( args_t *   args,
+                      config_t * config ) {
+  (void)args;
+
+  args_t c_args = configure_args();
+  configure_cmd_fn( &c_args, config );
+
+  run_firedancer_init( config, 1 );
+
+  fd_log_private_shared_lock[ 1 ] = 0;
+  fd_topo_join_workspaces( &config->topo, FD_SHMEM_JOIN_MODE_READ_WRITE );
+  fd_topo_fill( &config->topo );
+
+  ulong tile_idx1 = fd_topo_find_tile( &config->topo, "ipecho", 0UL );
+  fd_topo_tile_t * ipecho_tile1 = &config->topo.tiles[ tile_idx1 ];
+  ulong volatile * const ipecho_metrics1 = fd_metrics_tile( ipecho_tile1->metrics );
+  (void)ipecho_metrics1;
+
+  fd_topo_run_single_process( &config->topo, 2, config->uid, config->gid, fdctl_tile_run );
+
+  ulong tile_idx = fd_topo_find_tile( &config->topo, "ipecho", 0UL );
+  FD_TEST( tile_idx!=ULONG_MAX );
+  fd_topo_tile_t * ipecho_tile = &config->topo.tiles[ tile_idx ];
+
+  sleep(1);
+  ulong volatile * const ipecho_metrics = fd_metrics_tile( ipecho_tile->metrics );
+
+  ulong last_conns = ULONG_MAX;
+  ulong last_closed_ok = ULONG_MAX;
+  ulong last_closed_error = ULONG_MAX;
+
+  for(;;) {
+    ulong ipecho_conns = FD_VOLATILE_CONST( ipecho_metrics[ MIDX( GAUGE, IPECHO, CONNECTION_COUNT ) ] );
+    ulong ipecho_closed_ok = FD_VOLATILE_CONST( ipecho_metrics[ MIDX( COUNTER, IPECHO, CONNECTIONS_CLOSED_OK ) ] );
+    ulong ipecho_closed_error = FD_VOLATILE_CONST( ipecho_metrics[ MIDX( COUNTER, IPECHO, CONNECTIONS_CLOSED_ERROR ) ] );
+
+    if( FD_UNLIKELY( ipecho_conns!=last_conns || ipecho_closed_ok!=last_closed_ok || ipecho_closed_error!=last_closed_error ) ) {
+      FD_LOG_NOTICE(( "connections=%lu closed_ok=%lu closed_err=%lu", ipecho_conns, ipecho_closed_ok, ipecho_closed_error ));
+      last_conns = ipecho_conns;
+      last_closed_ok = ipecho_closed_ok;
+      last_closed_error = ipecho_closed_error;
+    }
+
+    nanosleep( &(struct timespec){ .tv_sec=0, .tv_nsec=1000L*1000L }, NULL );
+  }
+}
+
+action_t fd_action_ipecho_server = {
+  .name = NAME,
+  .args = NULL,
+  .perm = ipecho_server_cmd_perm,
+  .fn   = ipecho_server_cmd_fn,
+  .topo = ipecho_server_cmd_topo,
+};

src/app/firedancer-dev/main.c

@@ -86,6 +86,7 @@ extern fd_topo_run_tile_t fd_tile_benchs;
 extern fd_topo_run_tile_t fd_tile_bundle;
 extern fd_topo_run_tile_t fd_tile_pktgen;
 extern fd_topo_run_tile_t fd_tile_udpecho;
+extern fd_topo_run_tile_t fd_tile_ipecho;
 
 extern fd_topo_run_tile_t fd_tile_gossip;
 extern fd_topo_run_tile_t fd_tile_repair;
@@ -149,6 +150,7 @@ fd_topo_run_tile_t * TILES[] = {
   &fd_tile_snaprd,
   &fd_tile_snapdc,
   &fd_tile_snapin,
+  &fd_tile_ipecho,
   NULL,
 };
 
@@ -179,6 +181,8 @@ extern action_t fd_action_sim;
 extern action_t fd_action_backtest;
 extern action_t fd_action_snapshot_load;
 extern action_t fd_action_repair;
+extern action_t fd_action_shred_version;
+extern action_t fd_action_ipecho_server;
 
 action_t * ACTIONS[] = {
   &fd_action_run,
@@ -208,6 +212,8 @@ action_t * ACTIONS[] = {
   &fd_action_backtest,
   &fd_action_snapshot_load,
   &fd_action_repair,
+  &fd_action_shred_version,
+  &fd_action_ipecho_server,
   NULL,
 };
 

src/app/firedancer/Local.mk

@@ -36,6 +36,9 @@ $(call add-objs,topology,fd_firedancer)
 $(call add-objs,config,fd_firedancer)
 $(call add-objs,callbacks,fd_firedancer)
 
+# commands
+$(call add-objs,commands/shred_version,fd_firedancer)
+
 # version
 $(call make-lib,firedancer_version)
 $(call add-objs,version,firedancer_version)

src/app/firedancer/commands/shred_version.c

@@ -0,0 +1,41 @@
+#include "../../shared/fd_config.h"
+#include "../../shared/fd_action.h"
+
+#include "../../../disco/topo/fd_topo.h"
+#include "../../../discof/ipecho/fd_ipecho_client.h"
+
+#include <unistd.h>
+
+void
+shred_version_cmd_fn( args_t *   args,
+                      config_t * config ) {
+  (void)args;
+
+  void * _client = aligned_alloc( FD_IPECHO_CLIENT_ALIGN, fd_ipecho_client_footprint() );
+  FD_TEST( _client );
+  fd_ipecho_client_t * client = fd_ipecho_client_join( fd_ipecho_client_new( _client ) );
+  FD_TEST( client );
+
+  ulong tile_idx = fd_topo_find_tile( &config->topo, "gossip", 0UL );
+  FD_TEST( tile_idx!=ULONG_MAX );
+
+  fd_topo_tile_t * tile = &config->topo.tiles[ tile_idx ];
+  fd_ipecho_client_init( client, tile->gossip.entrypoints, tile->gossip.entrypoints_cnt );
+
+  for(;;) {
+    ushort shred_version = 0;
+    int _charge_busy;
+    int err = fd_ipecho_client_poll( client, &shred_version, &_charge_busy );
+    if( FD_UNLIKELY( -1==err ) ) FD_LOG_ERR(( "couldn't get shred version" ));
+    if( FD_UNLIKELY( !err) ) {
+      FD_LOG_STDOUT(( "%hu\n", shred_version ));
+      break;
+    }
+  }
+}
+
+action_t fd_action_shred_version = {
+  .name        = "shred-version",
+  .fn          = shred_version_cmd_fn,
+  .description = "Retrieve the current shred version from the entrypoints",
+};

src/app/firedancer/config/default.toml

@@ -349,11 +349,14 @@ user = ""
     # The shred version is a small hash of the genesis block and any
     # subsequent hard forks.  The validator client uses it to filter
     # out any shred traffic from validators that disagree with this
-    # validator on the genesis hash or the set of hard forks.  If
-    # nonzero, ignore any shreds that have a different shred version
-    # than this value.  If zero, the expected shred version is
-    # automatically determined by copying the shred version that the
-    # entrypoint validator is using.
+    # validator on the genesis hash or the set of hard forks.
+    #
+    # If set to zero, the expected shred version is automatically
+    # determined at startup by requesting the shred version from the
+    # gossip entrypoints.  If set to a nonzero value, the validator will
+    # still request the shred version from the gossip entrypoints, but
+    # if the shred version is not equal to this value, the validator
+    # will abort with an error.
     expected_shred_version = 0
 
 # This section configures the "funk" account database.  Currently, funk

src/app/firedancer/main.c

@@ -84,6 +84,7 @@ extern fd_topo_run_tile_t fd_tile_poh;
 extern fd_topo_run_tile_t fd_tile_send;
 extern fd_topo_run_tile_t fd_tile_tower;
 extern fd_topo_run_tile_t fd_tile_rpcserv;
+extern fd_topo_run_tile_t fd_tile_ipecho;
 
 fd_topo_run_tile_t * TILES[] = {
   &fd_tile_net,
@@ -110,6 +111,7 @@ fd_topo_run_tile_t * TILES[] = {
   &fd_tile_send,
   &fd_tile_tower,
   &fd_tile_rpcserv,
+  &fd_tile_ipecho,
   NULL,
 };
 
@@ -124,6 +126,7 @@ extern action_t fd_action_netconf;
 extern action_t fd_action_set_identity;
 extern action_t fd_action_help;
 extern action_t fd_action_version;
+extern action_t fd_action_shred_version;
 
 action_t * ACTIONS[] = {
   &fd_action_run,
@@ -137,6 +140,7 @@ action_t * ACTIONS[] = {
   &fd_action_set_identity,
   &fd_action_help,
   &fd_action_version,
+  &fd_action_shred_version,
   NULL,
 };