Issues with GVA nodes

Hi there,

We have several issues with the 1.9 GVA nodes (used by g1nkgo and g1superbot).

At the end, this is quite problematic for the users that are reporting issues (like problems with the balance, and similar). For instance G1nkgo cannot connect to some of these with the ssl issues:

Can the different node maintainers have a look?

Thanks!

cc @pini @Frederic_Renault @guenoel

EDIT: things fixed

[1] SSL issues
SSL_do_handshake() failed (SSL: error:0A000410:SSL routines::sslv3 alert handshake failure:SSL alert number 40) while SSL handshaking to upstream, client: 2400:cb00:366:1000:7117:9c6a:34e9:bf6f, server: g1nkgo.comunes.org, request: “POST /proxy/fania.g1server.net/gva/ HTTP/1.1”, upstream: “https://152.53.16.178:443/gva/”, host: “g1nkgo.comunes.org”, referrer: “https://g1nkgo.comunes.org/
SSL_do_handshake() failed (SSL: error:0A000410:SSL routines::sslv3 alert handshake failure:SSL alert number 40) while SSL handshaking to upstream, client: 2400:cb00:366:1000:7117:9c6a:34e9:bf6f, server: g1nkgo.comunes.org, request: “POST /proxy/fania.g1server.net/gva/ HTTP/1.1”, upstream: “https://[2a0a:4cc0:1:1156:8c1:caff:fe4b:b6d2]:443/gva/”, host: “g1nkgo.comunes.org”, referrer: “https://g1nkgo.comunes.org/
SSL_do_handshake() failed (SSL: error:0A000410:SSL routines::sslv3 alert handshake failure:SSL alert number 40) while SSL handshaking to upstream, client: 2400:cb00:366:1000:7117:9c6a:34e9:bf6f, server: g1nkgo.comunes.org, request: “POST /proxy/gibraleon.g1server.net/gva/ HTTP/1.1”, upstream: “https://152.53.16.178:443/gva/”, host: “g1nkgo.comunes.org”, referrer: “https://g1nkgo.comunes.org/
SSL_do_handshake() failed (SSL: error:0A000410:SSL routines::sslv3 alert handshake failure:SSL alert number 40) while SSL handshaking to upstream, client: 2400:cb00:366:1000:7117:9c6a:34e9:bf6f, server: g1nkgo.comunes.org, request: “POST /proxy/gibraleon.g1server.net/gva/ HTTP/1.1”, upstream: “https://[2a0a:4cc0:1:1156:8c1:caff:fe4b:b6d2]:443/gva/”, host: “g1nkgo.comunes.org”, referrer: “https://g1nkgo.comunes.org/
SSL_do_handshake() failed (SSL: error:0A000410:SSL routines::sslv3 alert handshake failure:SSL alert number 40) while SSL handshaking to upstream, client: 2400:cb00:366:1000:7117:9c6a:34e9:bf6f, server: g1nkgo.comunes.org, request: “POST /proxy/g1.cuates.net/gva/ HTTP/1.1”, upstream: “https://152.53.16.178:443/gva/”, host: “g1nkgo.comunes.org”, referrer: “https://g1nkgo.comunes.org/

SSL_do_handshake() failed (SSL: error:0A000410:SSL routines::sslv3 alert handshake failure:SSL alert number 40) while SSL handshaking to upstream, client: 2400:cb00:366:1000:7117:9c6a:34e9:bf6f, server: g1nkgo.comunes.org, request: “POST /proxy/g1.cuates.net/gva/ HTTP/1.1”, upstream: “https://[2a0a:4cc0:1:1156:8c1:caff:fe4b:b6d2]:443/gva/”, host: “g1nkgo.comunes.org”, referrer: “https://g1nkgo.comunes.org/
SSL_do_handshake() failed (SSL: error:0A000410:SSL routines::sslv3 alert handshake failure:SSL alert number 40) while SSL handshaking to upstream, client: 2400:cb00:366:1000:7117:9c6a:34e9:bf6f, server: g1nkgo.comunes.org, request: “POST /proxy/g1.madeirawonders.com/gva/ HTTP/1.1”, upstream: “https://152.53.16.178:443/gva/”, host: “g1nkgo.comunes.org”, referrer: “https://g1nkgo.comunes.org/

SSL_do_handshake() failed (SSL: error:0A000410:SSL routines::sslv3 alert handshake failure:SSL alert number 40) while SSL handshaking to upstream, client: 2400:cb00:366:1000:7117:9c6a:34e9:bf6f, server: g1nkgo.comunes.org, request: “POST /proxy/g1.madeirawonders.com/gva/ HTTP/1.1”, upstream: “https://[2a0a:4cc0:1:1156:8c1:caff:fe4b:b6d2]:443/gva/”, host: “g1nkgo.comunes.org”, referrer: “https://g1nkgo.comunes.org/

2 Likes

Hi @vjrj
I didn’t know that “bug” was going on.
Does node should resync from scratch to get OK, or there is another method ?

What is the problem abour mirror nodes ?

I do a sync from scratch (it only happened to me once).

That they don’t forge/sign/etc, only mirror the blockchain for queries.

This is taking hours and needs RAM !!

Before trying this method : La synchronisation de Duniter v1 occupe beaucoup de RAM - #21 by Moul

Moving from an “in-sync” node

  • duniter.db
  • peers.db
  • txs.db
  • /data
  • /g1

Have you ever done it this way ??

I thought they can handle transactions…

4 days later, we only have a working 1.9 node (https://duniter-v1.comunes.net).

Any help? TIA

1 Like

Hi @vjrj

It’s clear these problems are troublesome.

I wanted to mention of Astroport.ONE who is using “twin key” system & “primal TX marking” to make Duniter nodes transfer their Web of Trust level to their SSH and IPFS keys (and more) !

This would allow a secured update of database for forked nodes from another node via SFTP over SSH. Combined with Uplanet it brings nice conditions to run our blockchain smoothly with extra services

But nobody seems to like network automation or understand what I am talking about ??

Astroport.OT🤦‍♂️

@Pini il me.semble que tu avais une machine avec un Duniter v1.9 non ? Arrives tu encore a la démarrer ?

Elle n’a jamais (enfin, pas souvent) été arrêtée. C’est duniter.pini.fr:443.

EDIT: Mais comme rapporté dans le post initial, c’est juste un miroir. Je n’ai pas bien compris en quoi ça posait problème.

Can I use a mirror node to do payments? Maybe I wrong but for me a mirror node works read only, so helps with queries but not forging. So in other words currently still we can only use a gva node for payments.

Yes, you can use a mirror node to submit transactions. As long as it is well-connected to the P2P network, it will relay its transactions to other nodes via the WS2P protocol.

3 Likes

Good to know. Thanks for the clarification @elois .

1 Like

is duniter/duniter:dev the latest “stable” GVA docker image to use for duniter 1.9 in docker ?

HI @poka, I used the one shared in this forum by you or @pini I don’t remember well, but probably it’s the same.

1 Like

My node runs pinidh/duniter:dev.

1 Like

I’ve been seeing this error on my 1.9.0 node duniter.pini.fr since 3 days ago:

2025-03-17T09:57:08+00:00 - error: { httpCode: 400,
  uerr:
   { ucode: 1005,
     message: 'Document has unkown fields or wrong line ending format' } }

Any clue?

Same, impossible to keep up to date a GVA node due to this error. I believe the error occur on node startup juste after the sync success, whatever the block number at the end of sync (the sync is up to date when finished)

It doesn’t prevent the node to keep in sync as I understand it. These messages stopped 1h30 after my post.