0

clightning runs for a while great. However, every second day the computer locks up,

I use docker-compose for that:

version: "3"
services:
  bitcoind:
    image: nicolasdorier/docker-bitcoin:0.16.3
    container_name: bitcoind
    environment:
      BITCOIN_EXTRA_ARGS: |
        testnet=0
        whitelist=0.0.0.0/0
        server=1
        rpcuser=*****
        rpcpassword=*****
    expose:
      - "8333"
    ports:
      - "0.0.0.0:8333:8333"
    volumes:
      - "/media/ronald/data/scratch/bitcoin/mainnet/bitcoind:/data"

  clightning_bitcoin:
    image: elementsproject/lightningd
    container_name: lightning
    command:
      - --bitcoin-rpcconnect=bitcoind
      - --bitcoin-rpcuser=****
      - --bitcoin-rpcpassword=****
      - --network=bitcoin
      - --alias=TWronald✅
      - --rgb=15C315
      - --log-level=debug
      - --announce-addr=220.133.223.206:9735
      - --bind-addr=0.0.0.0:9735
    environment:
      EXPOSE_TCP: "true"
    expose:
      - "9735"
    ports:
      - "0.0.0.0:9735:9735"
    volumes:
      - "/media/ronald/data/scratch/bitcoin/mainnet/clightning:/root/.lightning"
      - "/media/ronald/data/scratch/bitcoin/mainnet/bitcoind:/etc/bitcoin"
    links:
      - bitcoind

lightning-cli getinfo

{
  "id": "035aef5661e1a6e370db60dc0455796800afd5b51fbc12a0a8b34836b15f5d7ef6", 
  "alias": "TWronald✅", 
  "color": "15c315", 
  "num_peers": 15, 
  "num_pending_channels": 0, 
  "num_active_channels": 15, 
  "num_inactive_channels": 0, 
  "address": [
    {
      "type": "ipv4", 
      "address": "220.133.223.206", 
      "port": 9735
    }
  ], 
  "binding": [
    {
      "type": "ipv4", 
      "address": "0.0.0.0", 
      "port": 9735
    }
  ], 
  "version": "", 
  "blockheight": 560987, 
  "network": "bitcoin", 
  "msatoshi_fees_collected": 0
}

Why the version of lightning is "" ??? scratch/bitcoin/mainet/clightning/config is empty.

Currently it is running fine. ps ax |grep lightn

 7278 ?        S      0:03 /usr/bin/lightning_channeld
 8524 pts/20   S+     0:00 grep --color=auto lightn
16579 ?        S      0:06 /usr/bin/lightning_channeld
31660 ?        S      1:19 lightningd --bitcoin-rpcconnect=bitcoind --bitcoin-rpcuser=**** --bitcoin-rpcpassword=**** --network=bitcoin --alias=TWronald✅ --rgb=15C315 --log-level=debug --announce-addr=220.133.223.206:9735 --bind-addr=0.0.0.0:9735
31665 ?        Z      0:00 [lightning_chann] <defunct>
31666 ?        Z      0:00 [lightning_closi] <defunct>
31667 ?        Z      0:00 [lightning_conne] <defunct>
31668 ?        Z      0:00 [lightning_gossi] <defunct>
31669 ?        Z      0:00 [lightning_hsmd] <defunct>
31670 ?        Z      0:00 [lightning_oncha] <defunct>
31671 ?        Z      0:00 [lightning_openi] <defunct>
31675 ?        S      0:00 /usr/bin/lightning_hsmd
31677 ?        S      0:00 /usr/bin/lightning_connectd
31678 ?        S      5:45 /usr/bin/lightning_gossipd
31862 ?        S      0:00 socat TCP4-listen:9835,fork,reuseaddr UNIX-CONNECT:/root/.lightning/lightning-rpc
31868 ?        S      0:06 /usr/bin/lightning_channeld
31869 ?        S      0:09 /usr/bin/lightning_channeld
31870 ?        S      0:06 /usr/bin/lightning_channeld
31871 ?        S      0:07 /usr/bin/lightning_channeld
31872 ?        S      0:07 /usr/bin/lightning_channeld
31874 ?        S      0:10 /usr/bin/lightning_channeld
31875 ?        S      0:10 /usr/bin/lightning_channeld
31876 ?        S      0:06 /usr/bin/lightning_channeld
31879 ?        S      0:10 /usr/bin/lightning_channeld
31988 ?        S      0:06 /usr/bin/lightning_channeld
31989 ?        S      0:06 /usr/bin/lightning_channeld
31992 ?        S      0:07 /usr/bin/lightning_channeld

... but still some Zombies.

When it grinds to a hold, it uses most with gossip.

Can you spot some changes I should make to improve my system?

1

This is most likely the gossipd bug that was causing a request from a peer to be processed very slowly. It was fixed in PR 2268. You don't have that fix in your installation yet, which can be seen by the 5 daemons that are not being reaped correctly (the <defunct> processes in the list). This was also fixed in PR 2224.

Both fixes are not yet part of a release, but we will release them soon.

  • I just updated the image you are using to include the mentioned fixes, so a simple docker pull elementsproject/lightningd should get you going again. I'm investigating why the automatic image build broke on Docker Hub. – cdecker Feb 1 at 9:54
  • Thank you very much. I updated the image and wait for the version fix, which is not that a priority. – Ronald Wiplinger Feb 1 at 12:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.