# \-E network.host doesn't work in VM **URL:** \ **Category:** Elasticsearch\ **Created:** [February 20, 2018, 9:51pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730 "2018-02-20T21:51:10Z")\ **Posts on this page:** 12\ **Page:** 1
**Author:** ![melissachang](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/melissachang/32/36684_2.png) [@melissachang](https://discuss.elastic.co/u/melissachang)\ **Post date:** [February 20, 2018, 9:51pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/1 "2018-02-20T21:51:10Z")
I'm trying to run ElasticSearch and view the webpage from a different machine. On my Debian desktop, `elasticsearch-5.0.1/bin/elasticsearch -E network.host=100.96.149.173` works. I can access machine\_hostname:9200. But running the same command in a Google Compute Engine VM gives: [2018-02-20T21:45:54,967][WARN][o.e.b.ElasticsearchUncaughtExceptionHandler] [] uncaught exception in thre ad [main] org.elasticsearch.bootstrap.StartupException: BindTransportException[Failed to bind to [9300-9400]]; nested : BindException[Cannot assign requested address]; at org.elasticsearch.bootstrap.Elasticsearch.init(Elasticsearch.java:125) ~[elasticsearch-5.0.2.jar :5.0.2] It appears I need a different set of flags for VMs, but I can't figure out what. ---
**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\ **Post date:** [February 20, 2018, 10:07pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/2 "2018-02-20T22:07:36Z")
Can you show the complete command you are using on the VM? ---
**Author:** ![melissachang](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/melissachang/32/36684_2.png) [@melissachang](https://discuss.elastic.co/u/melissachang)\ **Post date:** [February 20, 2018, 10:10pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/3 "2018-02-20T22:10:07Z")
`elasticsearch-5.0.2/bin/elasticsearch -E network.host=` I also tried discovery-gce plugin. If I do `elasticsearch-5.0.2/bin/elasticsearch -E network.host=_gce:hostname_` or `elasticsearch-5.0.2/bin/elasticsearch -E network.host=_gce_`, I can curl against internal IP. But I want to curl against external IP. There doesn't seem to be a discovery-gce option for that. ---
**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\ **Post date:** [February 20, 2018, 10:13pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/4 "2018-02-20T22:13:04Z")
`elasticsearch-5.0.2/bin/elasticsearch -E network.host=0.0.0.0` should bind to all IPs. ---
**Author:** ![melissachang](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/melissachang/32/36684_2.png) [@melissachang](https://discuss.elastic.co/u/melissachang)\ **Post date:** [February 20, 2018, 10:18pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/5 "2018-02-20T22:18:41Z")
That doesn't work. If I run that, I get: [2018-02-20T22:15:58,641][INFO][o.e.t.TransportService] [urZOilT] publish\_address {10.138.0.4:9300}, bo und\_addresses {[::]:9300} 10.138.0.4 is the VM's internal IP. I confirmed if I run `curl 10.138.0.4:9200` inside VM works. But loading external\_ip:9200 from outside the VM doesn't work. This should be easy for you to reproduce, just create a Debian VM on Google Compute Engine. ---
**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\ **Post date:** [February 20, 2018, 10:19pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/6 "2018-02-20T22:19:16Z")
Is the external address actually bound to an interface on the VM then? ---
**Author:** ![melissachang](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/melissachang/32/36684_2.png) [@melissachang](https://discuss.elastic.co/u/melissachang)\ **Post date:** [February 20, 2018, 10:25pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/7 "2018-02-20T22:25:20Z")
Yes, I followed [Running a basic Apache web server](https://cloud.google.com/compute/docs/tutorials/basic-webserver-apache) The Hello World Apache server works: [http://35.230.120.179/](http://35.230.120.179/) ElasticSearch doesn't: [http://35.230.120.179:9200/](http://35.230.120.179:9200/) ---
**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\ **Post date:** [February 20, 2018, 10:28pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/8 "2018-02-20T22:28:59Z")
What does `ip link show` or `ifconfig -a` show? ---
**Author:** ![melissachang](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/melissachang/32/36684_2.png) [@melissachang](https://discuss.elastic.co/u/melissachang)\ **Post date:** [February 20, 2018, 10:33pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/9 "2018-02-20T22:33:18Z")
$ ip link show 1: lo: \ mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: eth0: \ mtu 1460 qdisc mq state UP mode DEFAULT group default qlen 1000 link/ether 42:01:0a:8a:00:04 brd ff:ff:ff:ff:ff:ff $ ifconfig -a -bash: ifconfig: command not found ---
**Author:** ![rjernst](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/rjernst/32/6363_2.png) [@rjernst](https://discuss.elastic.co/u/rjernst)\ **Post date:** [February 21, 2018, 9:48pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/10 "2018-02-21T21:48:11Z")
Did you create a firewall rule allowing port 9200? ---
**Author:** ![melissachang](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/melissachang/32/36684_2.png) [@melissachang](https://discuss.elastic.co/u/melissachang)\ **Post date:** [February 21, 2018, 10:23pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/11 "2018-02-21T22:23:05Z")
I thought I did, but it turns out my firewall rule only allowed 80. I changed to allow all ports. `elasticsearch-5.0.2/bin/elasticsearch -E network.host=0.0.0.0` now works. But `elasticsearch-5.0.2/bin/elasticsearch -E network.host=35.230.120.179` still gives BindTransportException error from above. Should I file a bug? From googling, it seems like there might be a magic combination of flags that work, but I'm not sure what they are. Should be easy for anyone to repro. Just create a Debian VM on Google Compute Engine. You [should not be charged](https://cloud.google.com/free/). ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [March 21, 2018, 10:23pm UTC](https://discuss.elastic.co/t/e-network-host-doesnt-work-in-vm/120730/12 "2018-03-21T22:23:15Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.