Meta tags:
description=Set up an internal passthrough Network Load Balancer with VM instance group backends;

Headings (most frequently used words):

gcloud, console, api, configuration, load, configure, create, the, balancer, and, test, backend, your, vms, vm, instance, on, groups, client, with, backends, set, up, network, subnet, rules, use, traffic, ports, connection, start, basic, subnets, ipv6, region, firewall, components, ip, from, forwarding, frontend, review, an, internal, only, bash, script, using, global, access, address, accept, all, multiple, affinity, rule, to, setup, passthrough, group, stay, organized, collections, save, categorize, content, based, preferences, permissions, ipv4, dual, stack, additional, options, what, next, appendix, primer, install, tcp, server, startup, enable, managed, remove, external, addresses, reserved, two, session, tracking, policy, zonal, in, another, subsetting, for, packet, mirroring, products, pricing, support, resources, engage, ping, send, requests, balanced, connect, connectivity, new,

Text of the page (most frequently used words):
the (1147), network (351), and (339), load (290), for (283), compute (271), create (258), west1 (258), #backend (240), #internal (228), ipv6 (221), instance (219), subnet (206), balancer (200), click (197), name (189), allow (186), region (152), rule (144), address (143), you (143), https (140), gcloud (136), this (133), following (131), zone (128), projects (123), project_id (123), health (114), googleapis (113), com (113), vms (112), forwarding (110), cloud (104), group (101), instances (100), ilb (100), that (99), check (98), google (97), from (97), http (96), with (92), select (90), traffic (85), example (84), ports (83), stack (79), passthrough (78), rules (77), www (77), service (76), set (76), ssh (75), client (75), post (74), tcp (72), global (71), add (70), all (67), configure (66), groups (66), enter (65), configuration (64), firewall (64), ranges (64), server (62), use (62), only (61), console (61), range (58), ipv4 (57), networks (56), echo (56), two (55), information (54), type (53), access (53), request (52), page (50), default (50), section (50), balancing (49), regions (48), backends (48), source (46), can (46), test (46), 443 (46), vpc (45), port (45), tags (44), ingress (44), zones (43), dual (43), using (42), script (42), 8008 (40), apache2 (40), are (39), your (39), method (39), curl (39), 8080 (39), 8088 (39), overview (38), frontend (38), connections (38), then (37), see (36), command (36), ssl (36), false (36), external (36), etc (36), table (35), custom (35), connection (35), four (35), unmanaged (35), uses (34), subnetwork (34), debian (34), listen (34), socket (34), target (34), new (33), regional (33), nlisten (32), sed (32), these (31), insert (31), each (31), run (30), zonal (30), mode (30), managed (30), metadata (30), tee (30), direction (30), route (29), incoming (29), europe (29), more (28), local (28), how (28), checks (28), making (28), 8443 (28), prt_conf (28), ens4 (27), balancers (27), update (27), replace (26), ipprotocol (26), subnets (26), image (26), cat (26), services (25), connectivity (25), subnetworks (25), step (24), steps (24), next (24), protocol (24), any (24), conf (24), response (24), action (23), same (23), affinity (23), when (23), get (23), list (22), specified (22), startup (22), udp (22), api (22), file_ports (22), file_http_site (22), file_https_site (22), packets (21), interface (21), enable (21), choose (21), which (21), neg (21), fd20 (20), done (20), routing (19), dev (19), bash (19), options (19), note (19), addresses (19), html (19), 1000 (19), priority (19), available (18), forwardingrules (18), delete (18), vm_hostname (18), need (17), used (17), make (17), multiple (17), true (17), about (16), metric (16), pref (16), medium (16), connect (16), ephemeral (16), sites (16), indicated (16), protocols (16), instancegroups (16), firewalls (16), proto (15), created (15), networking (15), proxy (15), single (15), requests (15), because (15), beta (15), project (15), management (15), targets (15), policy (14), fe80 (14), creates (14), review (14), components (14), filter (14), assigned (14), http_listen_prts (14), http_vh_prts (14), https_listen_prts (14), https_vh_prts (14), prt_conf_2 (14), http_site_conf (14), http_site_conf_2 (14), https_site_conf (14), https_site_conf_2 (14), other (13), apache (13), bind (13), install (13), scheme (13), option (13), template (13), boot (13), specify (13), balanced (13), probe (13), allows (13), hybrid (13), vip (12), based (12), not (12), networktier (12), premium (12), second (12), edit (12), match (12), change (12), static (12), configured (12), repeat (12), startup_script_completed (12), apt (12), server_socket (12), policies (11), sudo (11), kernel (11), show (11), setup (11), application (11), backendservices (11), shows (11), settings (11), one (11), session (11), served (11), index (11), describe (11), url (11), applicable (11), within (11), architecture (10), support (10), its (10), 307 (10), 120c (10), 2000 (10), again (10), packet (10), want (10), advanced (10), required (10), demonstrate (10), spillover (10), tracking (10), behavior (10), values (10), purpose (10), accept (10), internet (10), family (10), bin (10), flavor (10), computemetadata (10), standard (10), ipv6_address (10), expected (10), listens (10), tag (10), system (9), send (9), 2ff (9), fe36 (9), ffbe (9), gateway (9), sends (9), named (9), identify (9), they (9), field (9), patch (9), subsetting (9), number (9), insecure (9), loadbalancingscheme (9), primary (9), verify (9), fields (9), must (9), disabled (9), flag (9), interfaces (9), var (9), ensure (9), like (9), combinations (9), ula (9), ipv4_ipv6 (9), buckets (9), code (8), content (8), via (8), should (8), parameters (8), blue (8), mark (8), before (8), have (8), distributed (8), shared (8), nat (8), 000 (8), _default_ (8), restart (8), disk (8), binds (8), listening (8), interface_name (8), body (8), close (8), necho (8), sourceranges (8), allowed (8), icmp (8), logconfig (8), thumb (7), mtu (7), after (7), different (7), troubleshoot (7), concepts (7), mirroring (7), start (7), specific (7), procedure (7), backendservice (7), version (7), either (7), ratio (7), also (7), automatically (7), remove (7), linux (7), items (7), images (7), located (7), separate (7), testing (7), python (7), virtual (7), contains (7), being (7), apply (7), web (7), guide (7), storage (7), negs (7), mtls (7), capabilities (7), resources (6), output (6), view (6), 100 (6), through (6), failover (6), collector_backend_service (6), where (6), mirror (6), basic (6), public (6), data (6), monitor (6), existing (6), ipaddress (6), finalize (6), there (6), continuing (6), zonal_affinity_spill_cross_zone (6), among (6), reserve (6), accepts (6), exit (6), a2ensite (6), a2enmod (6), systemctl (6), touch (6), gnu (6), demonstrates (6), actual (6), af_inet6 (6), forwarding_rule_vip (6), import (6), urllib (6), headers (6), specifies (6), server_address (6), print (6), client_socket (6), running (6), later (6), distribution (6), addinstances (6), health_check_ranges (6), serverless (6), premises (6), cross (6), terms (5), originating (5), subnet_range (5), matches (5), gateway_address (5), don (5), provide (5), connected (5), cannot (5), regionbackendservices (5), without (5), certificates (5), ipcidrrange (5), clients (5), work (5), instead (5), none (5), look (5), useful (5), maximum (5), necessary (5), connecting (5), client2 (5), machinetype (5), machinetypes (5), canipforward (5), networkinterfaces (5), accessconfigs (5), one_to_one_nat (5), disks (5), persistent (5), read_write (5), autodelete (5), devicename (5), initializeparams (5), sourceimage (5), disktype (5), disktypes (5), disksizegb (5), scheduling (5), preemptible (5), deletionprotection (5), described (5), file (5), always (5), contents (5), destination (5), targettags (5), prefix (5), ping (5), roles (5), troubleshooting (5), metrics (5), redirect (5), português (4), español (4), manage (4), contact (4), down (4), details (4), non (4), lookup (4), outgoing (4), defined (4), forward (4), expires (4), 256 (4), starting (4), logging (4), monitoring (4), dns (4), names (4), important (4), health_check_name (4), checkbox (4), facing (4), follow (4), security (4), might (4), self (4), works (4), persistence (4), unhealthy (4), supported (4), placeholders (4), both (4), shared_loadbalancer_vip (4), numbers (4), save (4), already (4), size (4), templates (4), optional (4), location (4), selected (4), bookworm (4), additional (4), them (4), ipv6_only (4), validate (4), correct (4), simple (4), library (4), thank (4), closes (4), key (4), so_bindtodevice (4), programming (4), get_vm_hostname (4), req (4), sock_stream (4), so_reuseaddr (4), setsockopt (4), sol_socket (4), encode (4), client_address (4), 200 (4), handle (4), times (4), automatic (4), 2600 (4), 2d00 (4), b029 (4), sources (4), manually (4), stacktype (4), balancingmode (4), healthchecks (4), regionhealthchecks (4), napt (4), nprt_conf (4), identical (4), simplicity (4), software (4), 130 (4), 211 (4), 191 (4), ula_ipv6_range (4), guides (4), tools (4), maps (4), logs (4), terraform (4), examples (4), third (3), samples (3), status (3), products (3), last (3), under (3), sysctl (3), net (3), ip_nonlocal_bind (3), follows (3), adds (3), 1460 (3), referenced (3), link (3), provides (3), description (3), commands (3), configuring (3), requires (3), case (3), feature (3), observe (3), signed (3), privateipgoogleaccess (3), another (3), zonal_affinity_stay_within_zone (3), product (3), features (3), timeout (3), rest (3), reference (3), reserved (3), let (3), column (3), common (3), over (3), error (3), current (3), permit (3), allocated (3), download (3), such (3), machine (3), autohealing (3), usage (3), autoscaling (3), production (3), appears (3), enlarge (3), contacts (3), serve (3), text (3), learn (3), appropriate (3), having (3), nature (3), placed (3), omit (3), interprets (3), mean (3), systems (3), blocks (3), deny (3), restrictive (3), just (3), sessions (3), creation (3), private (3), document (3), pass (3), value (3), deliver (3), however (3), comma (3), separated (3), formatted (3), quoted (3), strings (3), communication (3), internalipv6range (3), consider (3), documentation (3), back (3), implemented (3), guest (3), itself (3), responses (3), iam (3), admin (3), engine (3), tls (3), explore (3), tutorials (3), directory (3), registration (3), migrate (3), classic (3), 한국어 (2), 日本語 (2), עברית (2), brasil (2), italiano (2), indonesia (2), français (2), américa (2), latina (2), deutsch (2), english (2), sign (2), site (2), youtube (2), started (2), pricing (2), understand (2), updated (2), 2026 (2), utc (2), licensed (2), license (2), feedback (2), added (2), main (2), 1024 (2), 89sec (2), 24ff (2), feb1 (2), 59b3 (2), multicast (2), ff00 (2), previous (2), 62sec (2), outlined (2), appendix (2), clean (2), accessing (2), fundamentals (2), forwarding_rule_name (2), collector (2), instance_group (2), order (2), lets (2), performance (2), consistent_hash_subsetting (2), specifying (2), scale (2), handled (2), regardless (2), including (2), subject (2), sections (2), pre (2), seconds (2), entry (2), idle_timeout_value (2), connection_persistence_behavior (2), tracking_mode (2), updating (2), backend_service (2), valid (2), idle (2), cli (2), client_ip (2), setting (2), certificate (2), replacement (2), handling (2), alternative (2), strategy (2), installs (2), configures (2), remain (2), adding (2), assign (2), reservation (2), promote (2), deleteaccessconfig (2), replacing (2), config (2), determine (2), removing (2), directly (2), multi (2), modify (2), allowglobalaccess (2), enabled (2), has (2), reboot (2), find (2), sets (2), handles (2), loop (2), binding (2), accepting (2), sending (2), write (2), locally (2), fetch (2), hostname (2), def (2), urlopen (2), return (2), read (2), decode (2), object (2), reusing (2), avoid (2), errors (2), restarting (2), interface_index (2), if_nametoindex (2), argument (2), queued (2), message (2), indicating (2), ready (2), bound (2), while (2), accepted (2), generate (2), construct (2), proper (2), length (2), len (2), plain (2), sendall (2), eof (2), executable (2), chmod (2), python3 (2), stored (2), root (2), folder (2), earlier (2), going (2), takes (2), few (2), minutes (2), scope (2), needs (2), initiate (2), bgp (2), prompted (2), allocate (2), complete (2), fixed (2), cidr (2), block (2), supports (2), 1db0 (2), b882 (2), 802 (2), connectiondraining (2), drainingtimeoutsec (2), httphealthcheck (2), although (2), na2ensite (2), na2enmod (2), nfile_ports (2), nfile_http_site (2), nfile_https_site (2), nhttp_listen_prts (2), nhttp_vh_prts (2), nhttps_listen_prts (2), nhttps_vh_prts (2), nvm_hostname (2), nhttp (2), 169 (2), 254 (2), ntee (2), nprt_conf_2 (2), nhttp_site_conf (2), nhttp_site_conf_2 (2), nhttps_site_conf (2), nhttps_site_conf_2 (2), nsystemctl (2), debian_image_name (2), instructional (2), associated (2), delivered (2), preserving (2), runs (2), 192 (2), 198 (2), ipv6accesstype (2), refer (2), control (2), dynamic_routing_mode (2), selects (2), transmission (2), unit (2), 1500 (2), autocreatesubnetworks (2), enableulainternalipv6 (2), routingconfig (2), routingmode (2), assigns (2), illustrated (2), parameter (2), hops (2), routed (2), makes (2), way (2), sent (2), operating (2), those (2), will (2), conditions (2), task (2), permissions (2), sdk (2), languages (2), frameworks (2), infrastructure (2), costs (2), observability (2), migration (2), industry (2), solutions (2), multicloud (2), databases (2), analytics (2), pipelines (2), hosting (2), development (2), pools (2), optimizations (2), authorization (2), workload (2), identity (2), pool (2), convert (2), capacity (2), app (2), functions (2), constraints (2), technology (2), areas (2), subscribe, newsletter, our, decade, climate, join, cookies, privacy, tech, twitter, events, blog, engage, training, certification, center, getting, github, release, notes, community, forums, sales, marketplace, easy, easytounderstand, solved, problem, solvedmyproblem, otherup, hard, hardtounderstand, incorrect, sample, incorrectinformationorsamplecode, missing, missingtheinformationsamplesineed, otherdown, tell, except, otherwise, noted, java, registered, trademark, oracle, affiliates, developers, creative, commons, attribution, line, telling, 32765, 32766, were, pointing, routes, brief, related, here, context, aid, understanding, primer, issues, peer, what, l3_default, configurations, balance, destinations, copy, collect, collected, help, detect, threats, does, than, 250, enabling, temporarily, disruptive, break, caution, deployment, larger, nonzero, zero, offerings, general, limited, launch, stage, descriptions, preview, maintained, processes, matched, sec, docs, includes, sessionaffinity, types, hash, prefixlength, addresstype, supporting, five, fewer, allports, whose, keyword, hold, pointer, try, encounter, conflicting, resource_type, forwarding_rule, resource_name, existing_forwarding_rule_name, six, certain, determines, component, unlike, doesn, reserving, alternatively, accessconfig, networkinterface, limits, leave, was, could, their, prevents, disconnect, optionally, minimum, autoscale, require, simpler, distribute, did, word, format, whether, include, says, changes, accessible, still, expands, tasks, perform, nano, indicates, serves, called, ipversion, 2001, db8, advertisement, routers, dynamic, bc7, 9a1c, add_box, diagram, means, instructions, creating, originated, grep, inspect, looking, part, but, why, behind, responded, comes, repeated, answered, won, able, removed, processed, environment, state, attempt, notice, out, devices, il4, label, append, colon, come, generating, displayed, virtue, been, tests, equivalent, time, initiating, allowing, looks, subsequent, instanceadmin, securityadmin, networkadmin, role, owner, editor, guidance, teach, familiarize, yourself, categorize, preferences, stay, organized, collections, home, audit, operate, maintain, map, quota, units, proxies, endpoint, draining, customize, quantum, authenticated, user, provided, mutual, encryption, secure, switch, between, deploy, hub, spoke, hop, party, appliances, weighted, optimize, latency, published, domain, faster, improved, protection, best, practices, fail, high, availability, rewrite, header, query, roll, bucket, organization, comparison, resource, model, discover, free, skip,

Text of the page (random words):
echo prt_conf_2 sed s listen 443 https_listen_prts necho prt_conf tee file_ports nhttp_site_conf cat file_http_site nhttp_site_conf_2 echo http_site_conf sed s 80 http_vh_prts necho http_site_conf_2 tee file_http_site nhttps_site_conf cat file_https_site nhttps_site_conf_2 echo https_site_conf sed s _default_ 443 https_vh_prts necho https_site_conf_2 tee file_https_site nsystemctl restart apache2 scheduling preemptible false deletionprotection false create two instance groups by making a post request to the instancegroups insert method post https compute googleapis com compute v1 projects project_id zones us west1 a instancegroups name ig a network https www googleapis com compute v1 projects project_id global networks lb network subnetwork https www googleapis com compute v1 projects project_id regions us west1 subnetworks lb subnet post https compute googleapis com compute v1 projects project_id zones us west1 c instancegroups name ig c network https www googleapis com compute v1 projects project_id global networks lb network subnetwork https www googleapis com compute v1 projects project_id regions us west1 subnetworks lb subnet add instances to each instance group by making a post request to the instancegroups addinstances method post https compute googleapis com compute v1 projects project_id zones us west1 a instancegroups ig a addinstances instances instance https www googleapis com compute v1 projects project_id zones us west1 a instances vm a1 instance https www googleapis com compute v1 projects project_id zones us west1 a instances vm a2 post https compute googleapis com compute v1 projects project_id zones us west1 c instancegroups ig c addinstances instances instance https www googleapis com compute v1 projects project_id zones us west1 c instances vm c1 instance https www googleapis com compute v1 projects project_id zones us west1 c instances vm c2 configure load balancer components these steps configure all of the internal passthrough network load balancer components starting with the health check and backend service and then the frontend components health check in this example you use an http health check that checks for an http 200 ok status code for more information see the health check section backend service because you need to pass http traffic through the internal load balancer you need to use tcp not udp forwarding rule this example creates a single internal forwarding rule internal ip address in this example you specify an internal ip address 10 1 2 99 when you create the forwarding rule console start your configuration in the google cloud console go to the load balancing page go to load balancing click create load balancer for type of load balancer select network load balancer tcp udp ssl and click next for proxy or passthrough select passthrough load balancer and click next for public facing or internal select internal and click next click configure basic configuration on the create internal passthrough network load balancer page enter the following information load balancer name be ilb region us west1 network lb network configure the backends click backend configuration in the health check list click create a health check and enter the following information name hc http 80 protocol http port 80 proxy protocol none request note that when you use the google cloud console to create your load balancer the health check is global if you want to create a regional health check use gcloud or the api click create to handle only ipv4 traffic in the new backend section for ip stack type select the ipv4 single stack option in the instance group list select the ig c instance group and click done click add a backend and repeat the step to add the ig a instance group verify that there is a blue check mark next to backend configuration before continuing configure the frontend click frontend configuration in the new frontend ip and port section do the following for name enter fr ilb for subnetwork select lb subnet in the internal ip purpose section in the ip address list select create ip address enter the following information and then click reserve name ip ilb ip version ipv4 static ip address let me choose custom ip address 10 1 2 99 for ports select multiple and then in port numbers enter 80 8008 8080 and 8088 verify that there is a blue check mark next to frontend configuration before continuing review the configuration click review and finalize review your load balancer configuration settings optional click equivalent code to view the rest api request that will be used to create the load balancer click create gcloud create a new regional http health check to test http connectivity to the vms on port 80 gcloud compute health checks create http hc http 80 region us west1 port 80 create the backend service for http traffic gcloud compute backend services create be ilb load balancing scheme internal protocol tcp region us west1 health checks hc http 80 health checks region us west1 add the two instance groups to the backend service gcloud compute backend services add backend be ilb region us west1 instance group ig a instance group zone us west1 a gcloud compute backend services add backend be ilb region us west1 instance group ig c instance group zone us west1 c create a forwarding rule for the backend service when you create the forwarding rule specify 10 1 2 99 for the internal ip address in the subnet gcloud compute forwarding rules create fr ilb region us west1 load balancing scheme internal network lb network subnet lb subnet address 10 1 2 99 ip protocol tcp ports 80 8008 8080 8088 backend service be ilb backend service region us west1 api create the health check by making a post request to the regionhealthchecks insert method post https compute googleapis com compute v1 projects project_id regions us west1 regionhealthchecks name hc http 80 type http httphealthcheck port 80 create the regional backend service by making a post request to the regionbackendservices insert method post https compute googleapis com compute v1 projects project_id regions us west1 backendservices name be ilb backends group https www googleapis com compute v1 projects project_id zones us west1 a instancegroups ig a balancingmode connection group https www googleapis com compute v1 projects project_id zones us west1 c instancegroups ig c balancingmode connection healthchecks https www googleapis com compute v1 projects project_id regions us west1 healthchecks hc http 80 loadbalancingscheme internal connectiondraining drainingtimeoutsec 0 create the forwarding rule by making a post request to the forwardingrules insert method post https compute googleapis com compute v1 projects project_id regions us west1 forwardingrules name fr ilb ipaddress 10 1 2 99 ipprotocol tcp ports 80 8008 8080 8088 loadbalancingscheme internal subnetwork https www googleapis com compute v1 projects project_id regions us west1 subnetworks lb subnet network https www googleapis com compute v1 projects project_id global networks lb network backendservice https www googleapis com compute v1 projects project_id regions us west1 backendservices be ilb networktier premium test your load balancer these tests show how to validate your load balancer configuration and learn about its expected behavior create a client vm this example creates a client vm vm client in the same region as the backend server vms the client is used to validate the load balancer s configuration and demonstrate expected behavior as described in the testing section console in the google cloud console go to the vm instances page go to vm instances click create instance for name enter vm client for region select us west1 for zone select us west1 a click advanced options click networking and configure the following fields for network tags enter allow ssh for network interfaces select the following network lb network subnet lb subnet click create gcloud the client vm can be in any zone in the same region as the load balancer and it can use any subnet in that region in this example the client is in the us west1 a zone and it uses the same subnet as the backend vms gcloud compute instances create vm client zone us west1 a image family debian 12 image project debian cloud tags allow ssh subnet lb subnet api make a post request to the instances insert method post https compute googleapis com compute v1 projects project_id zones us west1 a instances name vm client tags items allow ssh machinetype https www googleapis com compute v1 projects project_id zones us west1 a machinetypes e2 standard 2 canipforward false networkinterfaces network https www googleapis com compute v1 projects project_id global networks lb network subnetwork https www googleapis com compute v1 projects project_id regions us west1 subnetworks lb subnet accessconfigs type one_to_one_nat name external nat networktier premium disks type persistent boot true mode read_write autodelete true devicename vm client initializeparams sourceimage projects debian cloud global images debian image name disktype projects project_id zones us west1 a disktypes pd standard disksizegb 10 scheduling preemptible false deletionprotection false test connection from client vm this test contacts the load balancer from a separate client vm that is not from a backend vm of the load balancer the expected behavior is for traffic to be distributed among the four backend vms because no session affinity has been configured connect to the client vm instance gcloud compute ssh vm client zone us west1 a make a web request to the load balancer using curl to contact its ip address repeat the request so you can see that responses come from different backend vms the name of the vm generating the response is displayed in the text in the html response by virtue of the contents of var www html index html on each backend vm for example expected responses look like page served from vm a1 and page served from vm a2 curl http 10 1 2 99 the forwarding rule is configured to serve ports 80 8008 8080 and 8088 to send traffic to those other ports append a colon and the port number after the ip address like this curl http 10 1 2 99 8008 if you add a service label to the internal forwarding rule you can use internal dns to contact the load balancer using its service name curl http web test fr ilb il4 us west1 lb project_id internal ping the load balancer s ip address this test demonstrates an expected behavior you cannot ping the ip address of the load balancer this is because internal passthrough network load balancers are implemented in virtual network programming they are not separate devices connect to the client vm instance gcloud compute ssh vm client zone us west1 a attempt to ping the ip address of the load balancer notice that you don t get a response and that the ping command times out after 10 seconds in this example timeout 10 ping 10 1 2 99 send requests from load balanced vms this test demonstrates that when a backend vm sends packets to the ip address of its load balancer s forwarding rule those requests are routed back to itself this is the case regardless of the backend vm s health check state internal passthrough network load balancers are implemented by using virtual network programming and vm configuration in the guest os on linux vms the guest environment creates a route for the load balancer s ip address in the operating system s local routing table because this local route is within the vm itself not a route in the vpc network packets sent to the load balancer s ip address are not processed by the vpc network instead packets sent to the load balancer s ip address remain within the operating system of the vm connect to a backend vm such as vm a1 gcloud compute ssh vm a1 zone us west1 a important you won t be able to connect to a backend vm in this way if you have removed its external ip address for connection options if your backend vms don t have external ip addresses see choose a connection option for internal only vms make a web request to the load balancer by ip address or service name using curl the response comes from the same backend vm that makes the request repeated requests are answered in the same way the expected response when testing from vm a1 is always page served from vm a1 curl http 10 1 2 99 inspect the local routing table looking for a destination that matches the ip address of the load balancer itself 10 1 2 99 this route is a necessary part of an internal passthrough network load balancer but it also demonstrates why a request from a vm behind the load balancer is always responded to by the same vm ip route show table local grep 10 1 2 99 when a backend vm for an internal passthrough network load balancer sends packets to the load balancer s forwarding rule ip address the packets are always routed back to the vm that makes the request this is because an internal passthrough network load balancer is a pass through load balancer and is implemented by creating a local route for the load balancer s ip address within the vm s guest os as indicated in this section if you have a use case where load balanced backends need to send tcp traffic to the load balancer s ip address and you need the traffic to be distributed as if it originated from a non load balanced backend consider using a regional internal proxy network load balancer instead for more information see internal passthrough network load balancers as next hops set up load balancer with dual stack subnets and backends this document shows you how to configure and test an internal passthrough network load balancer that supports both ipv4 and ipv6 traffic the steps in this section describe how to configure the following the example on this page uses a custom mode vpc network named lb network dual stack ipv6 traffic requires a custom mode subnet a dual stack subnet stack type set to ipv4_ipv6 which is required for ipv6 traffic when you create a dual stack subnet on a custom mode vpc network you choose an ipv6 access type for the subnet for this example we set the subnet s ipv6 access type parameter to internal this means new vms on this subnet can be assigned both internal ipv4 addresses and internal ipv6 addresses for instructions see vpc documentation about adding a dual stack subnet firewall rules that allow incoming connections to backend vms the backend instance group which is located in the following region and subnet for this example region us west1 subnet lb subnet with primary ipv4 address range 10 1 2 0 24 although you choose which ipv4 address range to configure on the subnet the ipv6 address range is assigned automatically google provides a fixed size 64 ipv6 cidr block four backend dual stack vms two vms in an unmanaged instance group in zone us west1 a and two vms in an unmanaged instance group in zone us west1 c to demonstrate global access this example creates a second test client vm in a different region and subnet region europe west1 subnet europe subnet with prima...