Long time to reconnect after upgrade to 0.4.14

Support and Troubleshooting
1

I did the upgrade this morning and this one took a particularly long time for LND to reconnect and be happy. 30 minutes or so from 08:21 to 08:48 my local time (UTC+2).

My LND logs were full of this:

2022-02-12 06:46:28.407 [ERR] RPCS: [/lnrpc.Lightning/ConnectPeer]: server is still in the process of starting                                                                                                                                                                                                             
2022-02-12 06:46:31.841 [ERR] RPCS: [/lnrpc.Lightning/ConnectPeer]: server is still in the process of starting                                                                                                                                                                                                             
2022-02-12 06:46:35.075 [ERR] RPCS: [/lnrpc.Lightning/ConnectPeer]: server is still in the process of starting                                                                                                                                                                                                             
2022-02-12 06:46:39.649 [ERR] RPCS: [/lnrpc.Lightning/ConnectPeer]: server is still in the process of starting                                                                                                                                                                                                             
2022-02-12 06:46:43.308 [ERR] RPCS: [/lnrpc.Lightning/ConnectPeer]: server is still in the process of starting                                                                                                                                                                                                             
2022-02-12 06:46:47.181 [ERR] RPCS: [/lnrpc.Lightning/ConnectPeer]: server is still in the process of starting                                                                                                                                                                                                             
2022-02-12 06:46:51.006 [ERR] RPCS: [/lnrpc.Lightning/ConnectPeer]: server is still in the process of starting                                                                                                                                                                                                             
2022-02-12 06:46:54.214 [ERR] RPCS: [/lnrpc.Lightning/ConnectPeer]: server is still in the process of starting                                                                                                                                                                                                             
2022-02-12 06:46:57.283 [ERR] RPCS: [/lnrpc.Lightning/ConnectPeer]: server is still in the process of starting

I did one restart after the first 10 minutes, not sure that helped, after that I just let it run.

This part of Lightning really worries me: I’m building a service based on this node, but if I have to take it down, I can’t send payments or connections anywhere else.

Are there any real answers to this?

2

Yes, I notice also that is taking a bit longer to catch up.
And this could be due to some factors that I notice:

  • bitcoin sync over Tor takes a while until start finding good peers. A solution could be to follow this guide and activate node sync in hybrid mode (clearnet and Tor). The bitcoin sync part is taking longer than few months ago and that’s why LND is not starting. Is a normal behavior.
  • after bitcoin is synced and indexes catch up, also LND will take a while to start. That’s due to the graph and channels.db file size. So you can start looking to compact that file. If is huge like more than 2GB then could be a problem. If is less than 1GB is not such a big deal.
    Here I found 2 guides about compacting the db. guide 1 and guide 2
  • also you could try adding into your lnd.conf these parameters that can speed up the start.
[Application Options]
# Mark unpayable, unpaid invoices as deleted
gc-canceled-invoices-on-startup=1
gc-canceled-invoices-on-the-fly=1
# Avoid historical graph data sync
ignore-historical-gossip-filters=1
# Avoid slow startup time
sync-freelist=1
# Avoid high startup overhead
stagger-initial-reconnect=1
# Avoid auto force close channels
payments-expiration-grace-period=30m
  • Also, if you are still using RaspPi machine, but you want a faster node, for business use, then is time for you migrate to a strong machine, more powerful, more reliable.
  • Or just ignore all that stuff and just have some more patience. Patience is the key for a node operator.

I hope this will help you.

2 Likes