LBRYCommunity/hub
1
0
Fork 0
Code Issues 35 Pull requests 6 Projects Releases Packages Wiki Activity

update readme and docker files

Browse source
This commit is contained in:
Jack Robison 2022-05-04 15:24:47 -04:00
parent 10baf47c02
commit 53206a0861
No known key found for this signature in database
GPG key ID: DF25C68FE0239BB2
4 changed files with 248 additions and 66 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

194
README.md
View file

@ -64,15 +64,203 @@ With options for high performance, if you have 64gb of memory and 12 cores, ever
As of block 1147423 (4/21/22) the size of the scribe rocksdb database is 120GB and the size of the elasticsearch volume is 63GB. As of block 1147423 (4/21/22) the size of the scribe rocksdb database is 120GB and the size of the elasticsearch volume is 63GB.
### docker-compose ### docker-compose
The recommended way to run a scribe hub is with docker. The recommended way to run a scribe hub is with docker.
If you have the resources to run all of the services on one machine (at least 300gb of fast storage, preferably nvme, 64gb of RAM, 12 fast cores), see [this](https://github.com/lbryio/scribe/blob/master/docker/docker-compose.yml) docker-compose example.
#### Cluster environment #### Cluster environment
For best performance the recommended setup uses three server instances, these can be rented VPSs, self hosted VMs (ideally not on one physical host unless the host is sufficiently powerful), or physical computers. One is a dedicated lbcd node, one an elasticsearch server, and the third runs the scribe services. With this configuration the lbcd and elasticsearch servers can be shared between multiple scribe hub servers - more on that later.
Server Requirements (space requirements are at least double what's needed so it's possible to copy snapshots into place or make snapshots):
- lbcd: 2 cores, 8gb ram (slightly more may be required syncing from scratch, from a snapshot 8 is plenty), 150gb of NVMe storage
- elasticsearch: 8 cores, 9gb of ram (8gb minimum given to ES), 150gb of SSD speed storage
- scribe: 8 cores, 32gb of ram, 200gb of NVMe storage
[Example docker-compose.yml](https://github.com/lbryio/scribe/blob/master/docker/elastic-compose.yml) for the elasticsearch instance All servers are assumed to be running ubuntu 20.04 with user named `lbry` with passwordless sudo and docker group permissions, ssh configured, ulimits set high, and docker + docker-compose installed. The server running elasticsearch should have swap disabled. The three servers need to be able to communicate with each other, they can be on a local network together or communicate over the internet. This guide will assume the three servers are on the internet.
[Example docker-compose.yml](https://github.com/lbryio/scribe/blob/master/docker/hub-compose.yml) for the hub instance ##### Setting up the lbcd instance
Log in to the lbcd instance and perform the following steps:
- Build the lbcd docker image by running
```
git clone https://github.com/lbryio/lbcd.git
cd lbcd
docker build . -t lbry/lbcd:latest
```
- Copy the following to `~/docker-compose.yml`
```
version: "3"
volumes:
lbcd:
services:
lbcd:
image: lbry/lbcd:latest
restart: always
network_mode: host
command:
- "--rpcuser=lbry"
- "--rpcpass=lbry"
- "--rpclisten=127.0.0.1"
volumes:
- "lbcd:/root/.lbcd"
ports:
- "127.0.0.1:9245:9245"
- "9246:9246" # p2p port
```
- Start lbcd by running `docker-compose up -d`
- Check the progress with `docker-compose logs -f --tail 100`
##### Setting up the elasticsearch instance
Log in to the elasticsearch instance and perform the following steps:
- Copy the following to `~/docker-compose.yml`
```
version: "3"
volumes:
es01:
services:
es01:
image: docker.elastic.co/elasticsearch/elasticsearch:7.16.0
container_name: es01
environment:
- node.name=es01
- discovery.type=single-node
- indices.query.bool.max_clause_count=8192
- bootstrap.memory_lock=true
- "ES_JAVA_OPTS=-Dlog4j2.formatMsgNoLookups=true -Xms8g -Xmx8g" # no more than 32, remember to disable swap
ulimits:
memlock:
soft: -1
hard: -1
volumes:
- "es01:/usr/share/elasticsearch/data"
ports:
- "127.0.0.1:9200:9200"
```
- Start elasticsearch by running `docker-compose up -d`
- Check the status with `docker-compose logs -f --tail 100`
##### Setting up the scribe hub instance
- Log in (ssh) to the scribe instance and generate and print out a ssh key, this is needed to set up port forwards to the other two instances. Copy the output of the following:
```
ssh-keygen -q -t ed25519 -N '' -f ~/.ssh/id_ed25519 <<<y >/dev/null 2>&1
```
- After copying the above key, log out of the scribe hub instance.
- Log in to the elasticsearch instance add the copied key to `~/.ssh/authorized_keys` (see [this](https://stackoverflow.com/questions/6377009/adding-a-public-key-to-ssh-authorized-keys-does-not-log-me-in-automatically) if confused). Log out of the elasticsearch instance once done.
- Log in to the lbcd instance and add the copied key to `~/.ssh/authorized_keys`, log out when done.
- Log in to the scribe instance and copy the following to `/etc/systemd/system/es-tunnel.service`, replacing `lbry` with your user and `your-elastic-ip` with your elasticsearch instance ip.
```
[Unit]
Description=Persistent SSH Tunnel for ES
After=network.target
[Service]
Restart=on-failure
RestartSec=5
ExecStart=/usr/bin/ssh -NTC -o ServerAliveInterval=60 -o ExitOnForwardFailure=yes -L 127.0.0.1:9200:127.0.0.1:9200 lbry@your-elastic-ip
User=lbry
Group=lbry
[Install]
WantedBy=multi-user.target
```
- Next, copy the following to `/etc/systemd/system/lbcd-tunnel.service` on the scribe instance, replacing `lbry` with your user and `your-lbcd-ip` with your lbcd instance ip.
```
[Unit]
Description=Persistent SSH Tunnel for lbcd
After=network.target
[Service]
Restart=on-failure
RestartSec=5
ExecStart=/usr/bin/ssh -NTC -o ServerAliveInterval=60 -o ExitOnForwardFailure=yes -L 127.0.0.1:9245:127.0.0.1:9245 lbry@your-lbcd-ip
User=lbry
Group=lbry
[Install]
WantedBy=multi-user.target
```
- Verify you can ssh in to the elasticsearch and lbcd instances from the scribe instance
- Enable and start the ssh port forward services on the scribe instance
```
sudo systemctl enable es-tunnel.service
sudo systemctl enable lbcd-tunnel.service
sudo systemctl start es-tunnel.service
sudo systemctl start lbcd-tunnel.service
```
- Build the scribe docker image on the scribe hub instance by running the following:
```
git clone https://github.com/lbryio/scribe.git
cd scribe
docker build -f ./docker/Dockerfile.scribe -t lbry/scribe:development .
```
- Copy the following to `~/docker-compose.yml` on the scribe instance
```
version: "3"
volumes:
lbry_rocksdb:
services:
scribe:
depends_on:
- scribe_elastic_sync
image: lbry/scribe:${SCRIBE_TAG:-development}
restart: always
network_mode: host
volumes:
- "lbry_rocksdb:/database"
environment:
- HUB_COMMAND=scribe
command:
- "--daemon_url=http://lbry:lbry@127.0.0.1:9245"
- "--max_query_workers=2"
- "--cache_all_tx_hashes"
scribe_elastic_sync:
image: lbry/scribe:${SCRIBE_TAG:-development}
restart: always
network_mode: host
ports:
- "127.0.0.1:19080:19080" # elastic notifier port
volumes:
- "lbry_rocksdb:/database"
environment:
- HUB_COMMAND=scribe-elastic-sync
command:
- "--elastic_host=127.0.0.1"
- "--elastic_port=9200"
- "--max_query_workers=2"
- "--filtering_channel_ids=770bd7ecba84fd2f7607fb15aedd2b172c2e153f 95e5db68a3101df19763f3a5182e4b12ba393ee8"
- "--blocking_channel_ids=dd687b357950f6f271999971f43c785e8067c3a9 06871aa438032244202840ec59a469b303257cad b4a2528f436eca1bf3bf3e10ff3f98c57bd6c4c6"
scribe_hub:
depends_on:
- scribe_elastic_sync
- scribe
image: lbry/scribe:${SCRIBE_TAG:-development}
restart: always
network_mode: host
ports:
- "50001:50001" # electrum rpc port and udp ping port
- "2112:2112" # comment out to disable prometheus metrics
volumes:
- "lbry_rocksdb:/database"
environment:
- HUB_COMMAND=scribe-hub
command:
- "--daemon_url=http://lbry:lbry@127.0.0.1:9245"
- "--elastic_host=127.0.0.1"
- "--elastic_port=9200"
- "--max_query_workers=4"
- "--host=0.0.0.0"
- "--max_sessions=100000"
- "--filtering_channel_ids=770bd7ecba84fd2f7607fb15aedd2b172c2e153f 95e5db68a3101df19763f3a5182e4b12ba393ee8"
- "--blocking_channel_ids=dd687b357950f6f271999971f43c785e8067c3a9 06871aa438032244202840ec59a469b303257cad b4a2528f436eca1bf3bf3e10ff3f98c57bd6c4c6"
- "--prometheus_port=2112" # comment out to disable prometheus metrics
```
- Start the scribe hub services by running `docker-compose up -d`
- Check the status with `docker-compose logs -f --tail 100`
### From source ### From source

7
docker/Dockerfile
View file

@ -39,15 +39,8 @@ RUN python3.9 docker/set_build.py
RUN rm ~/.cache -rf RUN rm ~/.cache -rf
# entry point # entry point
ARG host=localhost
ARG tcp_port=50001
ARG daemon_url=http://lbry:lbry@localhost:9245/
VOLUME $db_dir VOLUME $db_dir
ENV TCP_PORT=$tcp_port
ENV HOST=$host
ENV DAEMON_URL=$daemon_url
ENV DB_DIRECTORY=$db_dir ENV DB_DIRECTORY=$db_dir
ENV MAX_SESSIONS=100000
ENV MAX_SEND=1000000000000000000 ENV MAX_SEND=1000000000000000000
ENV MAX_RECEIVE=1000000000000000000 ENV MAX_RECEIVE=1000000000000000000

59
docker/docker-compose.yml
View file

@ -10,41 +10,43 @@ services:
depends_on: depends_on:
- lbcd - lbcd
- scribe_elastic_sync - scribe_elastic_sync
image: lbry/scribe:${SCRIBE_TAG:-latest-release} image: lbry/scribe:${SCRIBE_TAG:-latest}
restart: always restart: always
network_mode: host network_mode: host
volumes: volumes:
- "lbry_rocksdb:/database" - "lbry_rocksdb:/database"
environment: # for full options, see `scribe --help` environment:
- HUB_COMMAND=scribe - HUB_COMMAND=scribe
- DAEMON_URL=http://lbry:lbry@127.0.0.1:9245 command: # for full options, see `scribe --help`
- MAX_QUERY_WORKERS=2 - "--daemon_url=http://lbry:lbry@127.0.0.1:9245"
#- CACHE_ALL_TX_HASHES=Yes # uncomment to keep an index of all tx hashes in memory. This uses lots (10+g) of memory but substantially improves performance. - "--max_query_workers=2"
# - "--cache_all_tx_hashes" # uncomment to keep an index of all tx hashes in memory. This uses lots (10+g) of memory but substantially improves performance.
scribe_elastic_sync: scribe_elastic_sync:
depends_on: depends_on:
- es01 - es01
image: lbry/scribe:${SCRIBE_TAG:-latest-release} image: lbry/scribe:${SCRIBE_TAG:-latest}
restart: always restart: always
network_mode: host network_mode: host
ports: ports:
- "127.0.0.1:19080:19080" # elastic notifier port - "127.0.0.1:19080:19080" # elastic notifier port
volumes: volumes:
- "lbry_rocksdb:/database" - "lbry_rocksdb:/database"
environment: # for full options, see `scribe-elastic-sync --help` environment:
- HUB_COMMAND=scribe-elastic-sync - HUB_COMMAND=scribe-elastic-sync
- MAX_QUERY_WORKERS=2 command: # for full options, see `scribe-elastic-sync --help`
- FILTERING_CHANNEL_IDS=770bd7ecba84fd2f7607fb15aedd2b172c2e153f 95e5db68a3101df19763f3a5182e4b12ba393ee8 - "--max_query_workers=2"
- BLOCKING_CHANNEL_IDS=dd687b357950f6f271999971f43c785e8067c3a9 06871aa438032244202840ec59a469b303257cad b4a2528f436eca1bf3bf3e10ff3f98c57bd6c4c6 - "--filtering_channel_ids=770bd7ecba84fd2f7607fb15aedd2b172c2e153f 95e5db68a3101df19763f3a5182e4b12ba393ee8"
#- ELASTIC_HOST=127.0.0.1 # elasticsearch host - "--blocking_channel_ids=dd687b357950f6f271999971f43c785e8067c3a9 06871aa438032244202840ec59a469b303257cad b4a2528f436eca1bf3bf3e10ff3f98c57bd6c4c6"
#- ELASTIC_PORT=9200 # elasticsearch port - "--elastic_host=127.0.0.1" # elasticsearch host
#- ELASTIC_NOTIFIER_HOST=127.0.0.1 # address for the elastic sync notifier to connect to - "--elastic_port=9200" # elasticsearch port
#- ELASTIC_NOTIFIER_PORT=19080 - "--elastic_notifier_host=127.0.0.1" # address for the elastic sync notifier to connect to
- "--elastic_notifier_port=19080"
scribe_hub: scribe_hub:
depends_on: depends_on:
- lbcd - lbcd
- scribe_elastic_sync - scribe_elastic_sync
- scribe - scribe
image: lbry/scribe:${SCRIBE_TAG:-latest-release} image: lbry/scribe:${SCRIBE_TAG:-latest}
restart: always restart: always
network_mode: host network_mode: host
ports: ports:
@ -52,20 +54,21 @@ services:
- "2112:2112" # comment out to disable prometheus metrics - "2112:2112" # comment out to disable prometheus metrics
volumes: volumes:
- "lbry_rocksdb:/database" - "lbry_rocksdb:/database"
environment: # for full options, see `scribe-hub --help` environment:
- HUB_COMMAND=scribe-hub - HUB_COMMAND=scribe-hub
- DAEMON_URL=http://lbry:lbry@127.0.0.1:9245 # used for broadcasting transactions command: # for full options, see `scribe-hub --help`
- HOST=0.0.0.0 - "--daemon_url=http://lbry:lbry@127.0.0.1:9245"
- MAX_QUERY_WORKERS=4 # reader threads - "--max_query_workers=4"
- PROMETHEUS_PORT=2112 # comment out to disable prometheus metrics - "--host=0.0.0.0"
- FILTERING_CHANNEL_IDS=770bd7ecba84fd2f7607fb15aedd2b172c2e153f 95e5db68a3101df19763f3a5182e4b12ba393ee8 - "--filtering_channel_ids=770bd7ecba84fd2f7607fb15aedd2b172c2e153f 95e5db68a3101df19763f3a5182e4b12ba393ee8"
- BLOCKING_CHANNEL_IDS=dd687b357950f6f271999971f43c785e8067c3a9 06871aa438032244202840ec59a469b303257cad b4a2528f436eca1bf3bf3e10ff3f98c57bd6c4c6 - "--blocking_channel_ids=dd687b357950f6f271999971f43c785e8067c3a9 06871aa438032244202840ec59a469b303257cad b4a2528f436eca1bf3bf3e10ff3f98c57bd6c4c6"
#- MAX_SESSIONS=100000 # uncomment to increase the maximum number of electrum connections, defaults to 1000 - "--prometheus_port=2112" # comment out to disable prometheus metrics
#- ALLOW_LAN_UDP=Yes # uncomment to reply to clients on the local network # - "--elastic_host=127.0.0.1" # elasticsearch host
#- ELASTIC_HOST=127.0.0.1 # elasticsearch host # - "--elastic_port=9200" # elasticsearch port
#- ELASTIC_PORT=9200 # elasticsearch port # - "--elastic_notifier_host=127.0.0.1" # address for the elastic sync notifier to connect to
#- ELASTIC_NOTIFIER_HOST=127.0.0.1 # address for the elastic sync notifier to connect to # - "--elastic_notifier_port=19080"
#- ELASTIC_NOTIFIER_PORT=19080 # - "--max_sessions=100000 # uncomment to increase the maximum number of electrum connections, defaults to 1000
# - "--allow_lan_udp" # uncomment to reply to clients on the local network
es01: es01:
image: docker.elastic.co/elasticsearch/elasticsearch:7.16.0 image: docker.elastic.co/elasticsearch/elasticsearch:7.16.0
container_name: es01 container_name: es01

54
docker/hub-compose.yml
View file

@ -7,38 +7,38 @@ services:
scribe: scribe:
depends_on: depends_on:
- scribe_elastic_sync - scribe_elastic_sync
image: lbry/scribe:${SCRIBE_TAG:-latest-release} image: lbry/scribe:${SCRIBE_TAG:-development}
restart: always restart: always
network_mode: host network_mode: host
volumes: volumes:
- "lbry_rocksdb:/database" - "lbry_rocksdb:/database"
environment: # for full options, see `scribe --help` environment:
- HUB_COMMAND=scribe - HUB_COMMAND=scribe
- DAEMON_URL=http://lbry:lbry@127.0.0.1:9245 command:
- MAX_QUERY_WORKERS=2 - "--daemon_url=http://lbry:lbry@127.0.0.1:9245"
#- CACHE_ALL_TX_HASHES=Yes # uncomment to keep an index of all tx hashes in memory. This uses lots (10+g) of memory but substantially improves performance. - "--max_query_workers=2"
- "--cache_all_tx_hashes"
scribe_elastic_sync: scribe_elastic_sync:
image: lbry/scribe:${SCRIBE_TAG:-latest-release} image: lbry/scribe:${SCRIBE_TAG:-development}
restart: always restart: always
network_mode: host network_mode: host
ports: ports:
- "127.0.0.1:19080:19080" # elastic notifier port - "127.0.0.1:19080:19080" # elastic notifier port
volumes: volumes:
- "lbry_rocksdb:/database" - "lbry_rocksdb:/database"
environment: # for full options, see `scribe-elastic-sync --help` environment:
- HUB_COMMAND=scribe-elastic-sync - HUB_COMMAND=scribe-elastic-sync
- MAX_QUERY_WORKERS=2 command:
- FILTERING_CHANNEL_IDS=770bd7ecba84fd2f7607fb15aedd2b172c2e153f 95e5db68a3101df19763f3a5182e4b12ba393ee8 - "--elastic_host=127.0.0.1"
- BLOCKING_CHANNEL_IDS=dd687b357950f6f271999971f43c785e8067c3a9 06871aa438032244202840ec59a469b303257cad b4a2528f436eca1bf3bf3e10ff3f98c57bd6c4c6 - "--elastic_port=9200"
#- ELASTIC_HOST=127.0.0.1 # elasticsearch host - "--max_query_workers=2"
#- ELASTIC_PORT=9200 # elasticsearch port - "--filtering_channel_ids=770bd7ecba84fd2f7607fb15aedd2b172c2e153f 95e5db68a3101df19763f3a5182e4b12ba393ee8"
#- ELASTIC_NOTIFIER_HOST=127.0.0.1 # address for the elastic sync notifier to connect to - "--blocking_channel_ids=dd687b357950f6f271999971f43c785e8067c3a9 06871aa438032244202840ec59a469b303257cad b4a2528f436eca1bf3bf3e10ff3f98c57bd6c4c6"
#- ELASTIC_NOTIFIER_PORT=19080
scribe_hub: scribe_hub:
depends_on: depends_on:
- scribe_elastic_sync - scribe_elastic_sync
- scribe - scribe
image: lbry/scribe:${SCRIBE_TAG:-latest-release} image: lbry/scribe:${SCRIBE_TAG:-development}
restart: always restart: always
network_mode: host network_mode: host
ports: ports:
@ -46,17 +46,15 @@ services:
- "2112:2112" # comment out to disable prometheus metrics - "2112:2112" # comment out to disable prometheus metrics
volumes: volumes:
- "lbry_rocksdb:/database" - "lbry_rocksdb:/database"
environment: # for full options, see `scribe-hub --help` environment:
- HUB_COMMAND=scribe-hub - HUB_COMMAND=scribe-hub
- DAEMON_URL=http://lbry:lbry@127.0.0.1:9245 # used for broadcasting transactions command:
- HOST=0.0.0.0 - "--daemon_url=http://lbry:lbry@127.0.0.1:9245"
- MAX_QUERY_WORKERS=4 # reader threads - "--elastic_host=127.0.0.1"
- PROMETHEUS_PORT=2112 # comment out to disable prometheus metrics - "--elastic_port=9200"
- FILTERING_CHANNEL_IDS=770bd7ecba84fd2f7607fb15aedd2b172c2e153f 95e5db68a3101df19763f3a5182e4b12ba393ee8 - "--max_query_workers=4"
- BLOCKING_CHANNEL_IDS=dd687b357950f6f271999971f43c785e8067c3a9 06871aa438032244202840ec59a469b303257cad b4a2528f436eca1bf3bf3e10ff3f98c57bd6c4c6 - "--host=0.0.0.0"
#- MAX_SESSIONS=100000 # uncomment to increase the maximum number of electrum connections, defaults to 1000 - "--max_sessions=100000"
#- ALLOW_LAN_UDP=Yes # uncomment to reply to clients on the local network - "--filtering_channel_ids=770bd7ecba84fd2f7607fb15aedd2b172c2e153f 95e5db68a3101df19763f3a5182e4b12ba393ee8"
#- ELASTIC_HOST=127.0.0.1 # elasticsearch host - "--blocking_channel_ids=dd687b357950f6f271999971f43c785e8067c3a9 06871aa438032244202840ec59a469b303257cad b4a2528f436eca1bf3bf3e10ff3f98c57bd6c4c6"
#- ELASTIC_PORT=9200 # elasticsearch port - "--prometheus_port=2112" # comment out to disable prometheus metrics
#- ELASTIC_NOTIFIER_HOST=127.0.0.1 # address for the elastic sync notifier to connect to
#- ELASTIC_NOTIFIER_PORT=19080