MultiChain Developer Q&A - Recent questions tagged sync

How to increase the synchronisation speed ?

Thu, 01 Oct 2020 13:46:41 +0000

We have a business case where we have to frequently write and query multichain streams. To improve the performance, we are writing to a primary node and querying to a secondary node of blockchain network.

But the problem is that when primary node is writing, sync speed is almost 1 block per minute. And when primary node is idle it's still 3 blocks per minute. Sometimes it takes weeks for secondary node to catch up which is not idle for our business case where customers can query the data within 2-3 days of writing.

Our customers are little unhappy with this situation. Please suggest, is there is any solution to increase the speed of syncing or if there is any other solution to this problem.

Our blockchain has around 4,00,000 blocks currently.

peer network with different startingheight value

Sat, 28 Sep 2019 15:43:07 +0000

Dear Team,

I have a network composed by 4 nodes, one seed node and other 3 nodes connected to it. I checked they are in sync with the <getblockchaininfo> command. All of them show the same values for the following parameters:
"blocks" : 477,
"headers" : 477,
"bestblockhash" : "00110f363f98ff344e93d5491b79644783abd2f58a7b257540be7c416596a562"

However, if I log in in the seed node and I send the command <getpeerinfo>, I get the list of the 3 peers with a different value for one of them:
[
{
"id" : 1,
"startingheight" : 462,
"banscore" : 0,
"synced_headers" : 476,
"synced_blocks" : 476,
},
{
"id" : 3,
"startingheight" : 477,
"banscore" : 0,
"synced_headers" : 477,
"synced_blocks" : 477,
},
{
"id" : 4,
"startingheight" : 477,
"banscore" : 0,
"synced_headers" : 477,
"synced_blocks" : 477,
}
]

Could you kindly help me in understanding? Thanks, Fabio

Generate automatic events on create/update of item/stream so that polling is not required

Fri, 06 Sep 2019 07:09:04 +0000

I have multichain configured on server. So, like in Ethereum, we can create event in the smart contract which can keep listening to events for new actions, likewise, can we subscribe to stream events using Multichain's APIs without explicitely keep polling from subscriber's end?

Timeout downloading block from peer, disconnecting

Wed, 30 Jan 2019 09:24:29 +0000

Hi,

I have a master node with 900+ blocks. Now I try to connect a new node, but when it starts syncing up with the master, it disconnects after some blocks complaining about a timeout:

receive version message: /MultiChain:0.2.0.6/: version 70002, blocks=959, us=174.16.213.31:53738, peer=2
Added time data, samples 2, offset +0 (+0 minutes)
mchn: Parameter set from peer=2 verified
Pre-allocating up to position 0x100000 in rev00000.dat
UpdateTip: new best=00f4c23c213136008518e74414f17605c024ad3cf34d81cb53f8130266ee815b height=1 log2_work=9 tx=2 date=2019-01-28 08:48:07 progress=1.000000 cache=0
UpdateTip: new best=0000b25e3031363046cc69e138f209868484bbd15f6fee3f0dc42b5f51be3565 height=2 log2_work=9.5849625 tx=3 date=2019-01-28 08:48:08 progress=1.000000 cache=0
...
...
UpdateTip: new best=00113d9451141fc321bc88e5bc2f2388580f683fa3c9af4705a754485c3fbdac height=126 log2_work=14.988685 tx=1284 date=2019-01-28 14:10:10 progress=1.000000 cache=0
UpdateTip: new best=005f0110f6d32bf8e68a9598dd7533852a30cff92ec65341e25e55149f976eff height=127 log2_work=15 tx=1540 date=2019-01-28 14:10:14 progress=1.000000 cache=0
UpdateTip: new best=00f59a8a840365aa1c3bac5fe717753c07f4cf27f6e95926423531ab33488af8 height=128 log2_work=15.011227 tx=1760 date=2019-01-28 14:10:16 progress=1.000000 cache=0
ResendWalletTransactions()
UpdateTip: new best=008a866cb66d2f98894fd98ebe2b7eb78f5d9ff5460d6e723ef0db38ba6757f2 height=129 log2_work=15.022368 tx=1908 date=2019-01-28 14:10:18 progress=1.000000 cache=0
UpdateTip: new best=00e7442c16a2008681791394d95966db86e5155cee8668ff25566c332953d6f0 height=130 log2_work=15.033423 tx=2015 date=2019-01-28 14:10:20 progress=1.000000 cache=0
UpdateTip: new best=0012883f0df88e470804ec4f5aefe78fbbfd0ced34b02737a321bb496b0f63d2 height=131 log2_work=15.044394 tx=2112 date=2019-01-28 14:10:23 progress=1.000000 cache=0
UpdateTip: new best=00d26a69b99e592f5687a8704a217b87be3516f4f72dfdb80d734d939aa52663 height=132 log2_work=15.055282 tx=2273 date=2019-01-28 14:10:25 progress=1.000000 cache=0
UpdateTip: new best=00a163113a22b08250698cd44280b44e4a7f952ddf9dbb9aa33f88a8c57741d9 height=133 log2_work=15.066089 tx=2410 date=2019-01-28 14:10:27 progress=1.000000 cache=0
UpdateTip: new best=006862bab537d132727e6abc9f1adef8e2520c9d26d5d1f5f727d7ab24f4e8b4 height=134 log2_work=15.076816 tx=2521 date=2019-01-28 14:10:28 progress=1.000000 cache=0
UpdateTip: new best=00c77418d1248a4118363b5df2c620023d2e3fd28ed6b83f3366eeed99a04118 height=135 log2_work=15.087463 tx=2613 date=2019-01-28 14:10:30 progress=1.000000 cache=0
ResendWalletTransactions()
UpdateTip: new best=0081d6e61d0b716b1b318ddd48260d1c620be5b100fdad271e2732311fe7b049 height=136 log2_work=15.098032 tx=2789 date=2019-01-28 14:10:33 progress=1.000000 cache=0
Timeout downloading block 00e9dfcfc91067a07d32a80e241d944089d33a5aeea12335422970dd270dcbe1 from peer=2, disconnecting

What is this timeout about? Why does it happen? Is there any parameter that allows me to increase such timeout?

Thanks!

Sync Wallets Between Nodes

Mon, 05 Nov 2018 14:17:14 +0000

Hello,

I started learning MC 2.0 I noticed something I'd like to ask about.

My setup: 2 nodes on two different computers running locally.

When I created a new account on one node, I couldn't interact or use the address from the other node unless I imported the address or the wallet.

Is there a way to keep all addresses/wallets synced between the nodes? Or is there a method to add an address to all wallets on all the nodes so the user can interact with the blockchain using any node.

Thank you for your time.

[edit] I tried setting up cluster nodes using -initprivkey. However, when I imported a new account, it wasn't added to the other nodes.

Is it possible to disable transactions validation on a second node?

Sat, 19 May 2018 22:44:35 +0000

Hello guys!

A second node is joining an existing chain. The sync process takes lot of time. Is there any possibility to speed up the sync process?

I think when I disable the validation process of each transactions on the second node could speed up the sync process a little bit. Could be this true? If yes, how can I disable it?

I set mine-empty-rounds to 0 to reduce the blocks, but still I have many blocks which need to be synchronized on another node.

Many thanks in advance!

Block syncing to new nodes.

Tue, 14 Nov 2017 06:07:57 +0000

Hi team,

I 've a 2 node multichain. The nodes were in sync and working fine, I started with load testing of the nodes and gradually on the 2nd node I started to get memory errors like below. Anyway that's another story.

EXCEPTION: St9bad_alloc
std::bad_alloc
bitcoin in ProcessMessages()

So after this I deleted the ~/.multichain/chain1 from my 2nd node and started the multichain on 2nd node again. It took some time to get the blocks synced (the inflight section of getpeerinfo didn't show any blocks so I assumed they are synced. )

Now on the 2nd node when I do getinfo and getpeerinfo I see respectively;

"blocks" : 1287,

"inbound" : true,
"startingheight" : 1287,
"banscore" : 0,
"synced_headers" : 1287,
"synced_blocks" : 1287,

Seems alright.

But when I do the same on 1st node.

"blocks" : 1287,

"inbound" : false,
"startingheight" : 632,
"banscore" : 0,
"synced_headers" : 634,
"synced_blocks" : -1,

Doesn't seem alright to me.

1. What does the inbound imply here?

2. why my sync header and block is different on node 1?

Thanks.

How exactly does MultiChain broadcast the information?

Fri, 30 Jun 2017 13:48:23 +0000

What kind of code is involved? Is it using the boost c++ library's signals? How exactly do those work? Thanks!

How to Increase the speed of blocks synchronization

Wed, 08 Mar 2017 07:10:04 +0000

Hi,

I am joining an existing chain that has over 150,000 blocks already and blocks syncronisation to my local node takes lot of time.. it's already 4 hours and blocks synched 40,000 only.

Is there anyway that we could improve the speed of sync?.

getblockchaininfo

{
    "chain" : "main",
    "chainname" : "chain1",
    "description" : "MultiChain chain1",
    "protocol" : "multichain",
    "setupblocks" : 60,
    "reindex" : false,
    "blocks" : 40273,
    "headers" : 154576,
    "bestblockhash" : "00007953b668e13658f2a0242e4fc061a506fb51d4ad466cb71d3a48e
d4b7c6a",
    "difficulty" : 0.00001526,
    "verificationprogress" : 1.00000000,
    "chainwork" : "000000000000000000000000000000000000000000000000000000009d529
d52"

}

Thanks.

Node catchup seems to be rolling back blocks continuously

Thu, 13 Oct 2016 14:23:59 +0000

HI

I blew away one of my multichain nodes and then re-created it my peering it with a node running in the cloud. I've been watching it and two things look strange:

1) It is gradually increasing the number of blocks that are in the syncing node, but it is continously rolling back blocks and then re-adding them.....it's like taking three steps forward followed by two back......would you exepct this......feels like there is an issue and the sync is now taking a very long time!

2) The syncing node seems to be mining....it has permissions to mine but my assumption is that it won't request the mempool until it catches up the existing mined blocks, which it hasn't done so.....so how could it be mining??

Cheers

Marty