Hi guys, hope you guys are having a good weekend.
I have a setup of 2 containers, each container having 1 cluster peer and 1 daemon. These 2 containers form a cluster.
Here is my docker-compose file: docker-compose.yml · GitHub
I successfully added file via the cluster HTTP API and got the CID correctly pinned in both 2 peers when having query param local=false
. However, when I used local=true
, the CID was only successfully pinned on the requested peer, while the other got context cancelled
error.
Inspecting the log, I find that this may be because one daemon cannot establish the swarm connection to the other so that the actions are timed-out.
Here is the log:
ipfs-cluster-peer_2 | ipfshttp ipfshttp/ipfshttp.go:682 IPFS request unsuccessful (swarm/connect?arg=%2Fip4%2F172.27.0.2%2Fudp%2F4001%2Fquic%2Fp2p%2F12D3KooWHr8tHvKNn4isWxhRGwEzZCwpDnn9HLKPQkLLRx1CM5rS). Code: 500. Message: connect 12D3KooWHr8tHvKNn4isWxhRGwEzZCwpDnn9HLKPQkLLRx1CM5rS failure: failed to dial 12D3KooWHr8tHvKNn4isWxhRGwEzZCwpDnn9HLKPQkLLRx1CM5rS:
ipfs-cluster-peer_2 | * [/ip4/172.27.0.2/tcp/4001] dial tcp4 0.0.0.0:4001->172.27.0.2:4001: i/o timeout
* [/ip4/172.27.0.2/udp/4001/quic] context deadline exceeded
ipfs-cluster-peer_2 | * [/ip4/113.172.236.221/udp/56885/quic] timeout: no recent network activity
Wonder if there is anything wrong in my config. Really appreciate any help.
Thanks in advance.