gotcha, will give it a try.
cheers.
1 Like
After a bit of searchingā¦ gcp cloud run seems to be a good heroku replacement with a free tier that still lets me be sufficiently lazy. Didnāt try Render.com yet though.
1 Like
maybe this helps
there is a separate module for scheduling jobs for parse server
you can also use cron jobs directly on your own server
1 Like
Hi, I searched and could not find anywhere the technical requirements for a server to self-host moralis server.
E.g. to run an Avalanche RPC: https://docs.avax.network/nodes/build/set-up-node-with-installer
Requirements for an Avalanche RPC:
- CPU: Equivalent of 8 AWS vCPU
- RAM: 16 GiB
- Storage: 1 TiB
- OS: Ubuntu 20.04 or MacOS >= 12
- Network: sustained 5Mbps up/down bandwidth
Somewhere else I see more āhumbleā requiremnts: https://gist.github.com/cgcardona/4a5c9d0dbe7fdbe6ffce77fa05e9e580
- 4 CPUs
- 8 GB
- 160 GB SSD Disk
- 5 TB transfer
(I donāt believe that 160 GB disk are enoughā¦ Iād calculate at least 300 GB for avalanche)
So I wonder, if I need a moralis server supporting both Avalanche and other chainsā¦ do I need to calculate for each chain like 500 GB- 1 TB of space depending on the chain, and equivalent RAM?
Or is Moralis smart also for this and is fetching data from a central place and thus doesnāt need to validate blocks like a full node does?
So, realistically, i I have a VPS with 1.5TB, CPU: 14 cores, 32 GB RAM, how many chains can this support?
Iād like to support these chains:
- Avalanche
- BSC
- Polygon
The main features I need are:
- track NFT transfers
- track various events (both fetch events from the past and real-time)
I was considering to install my own RPC for each chain I need, but moralis makes things easier than using RPCs.
I also was wondering if a Moralis server has an RPC for each chain and if I have access to itā¦ or how does it work behind the scenes?
Thanks for your answers, especially the technical requirements (my suggestion for you) should be in the documentation: https://docs.moralis.io/docs/v1-server-self-hosting?utm_source=blog&utm_medium=post&utm_campaign=Introducing%20Self-Hosted%20Moralis%20Servers
Cheers
this should be more then enough for you
you will have to use Streams API to sync events and nft transfers, you can use your own custom backend and database if you want too, you donāt have to necessarily use parse server unless you want it to work with moralis v1 sdk
Thanks mate, Iām eyeing VPS server on netcup: https://www.netcup.eu/vserver/vps.php#v-server-details
If I could get a lower specs:
- 12 vCore
- 24 GB RAM
- 960 GB SSD (RAID10)
Or:
- 10 vCore
- 18 GB RAM
- 640 GB SSD (RAID10)
How does moralis work behind the scenes? Where does the moralias server fetch the blockchain data if it doensāt run a full node behind the scenes?
it does run a full node behind the scenes, but if you want to use the API from moralis then you will get only the info that you need
if you want to run a full node then you will probably need higher specs
1 Like
Do I guess it right that it runs a node behind the scenes but then it āprunesā the blocks to save up space and saves the info in mongo DB for easier/less expensive in terms of queirying fetching of data?
So basically what needs a lot of space is the blockchain data, but do you know how much space is needed right now on a moralias server for each chain?
For me the advantage of using a self hosted moralis server, despite being āv1ā, is to avoid paying for queries. I see the free account has some limits, so if I query my own server like crazy I would not need to pay for itā¦
Btw, do you know if the serve hosted moralis server is gonna have long time support (LTS) or if it will eventually be deprecated as well?
If you tell me the self hosted server has an incertain destiny, then I guess Iāll study more v2. But if itās being supported, then itās definitively interesting.
you can self host with parse server, current moralis server uses parse server behind the scenes
with parse server it is easy to use cloud code and it uses by default mongo db, but you can use any other backend or programming language or technology if you want
you will need a way to get the data on the server in the first place, that can be done with web3api, with streams API or with another method that you find
we plan to provide support for web3api and streams API in the future for a long time
1 Like
We currently have a lot of things in Cloud Functions. What is the quickiest way to migrate the cloud functions code to self-hosted cloud functions server?
For example,
Moralis.web3ByChain does not exist
Moralis.Cloud does not exist
many other errors from typescript
Moralis.Cloud can be replaced with Parse.Cloud
Moralis.web3ByChain doesnāt work now either, you can import directly ethers or web3 in cloud code
1 Like
Hi everyone,
I actually followed the tutorial step-by-step and deployed on Heroku, but the server url is returning an error as shown in the image below as well as the build log.
-----> Building on the Heroku-22 stack
-----> Determining which buildpack to use for this app
-----> Node.js app detected
-----> Creating runtime environment
NPM_CONFIG_LOGLEVEL=error
USE_YARN_CACHE=true
NODE_VERBOSE=false
NODE_ENV=production
NODE_MODULES_CACHE=true
-----> Installing binaries
engines.node (package.json): unspecified
engines.npm (package.json): unspecified (use default)
engines.yarn (package.json): unspecified (use default)
Resolving node version 18.x...
Downloading and installing node 18.11.0...
Using default npm version: 8.19.2
Resolving yarn version 1.22.x...
Downloading and installing yarn (1.22.19)
Installed yarn 1.22.19
-----> Installing dependencies
Installing node modules (yarn.lock)
yarn install v1.22.19
[1/4] Resolving packages...
[2/4] Fetching packages...
warning Pattern ["debug@>= 2.6.9 < 3.0.0 || >= ^3.1.0"] is trying to unpack in the same destination "/tmp/yarncache.zNC7V/v6/npm-debug-3.2.7-72580b7e9145fb39b6676f9c5e5fb100b934179a-integrity/node_modules/debug" as pattern ["debug@^3.1.0","debug@^3.2.7","debug@^3.2.7"]. This could result in non-deterministic behavior, skipping.
[3/4] Linking dependencies...
[4/4] Building fresh packages...
Done in 28.82s.
-----> Build
Running build (yarn)
yarn run v1.22.19
$ tsc
Done in 2.92s.
-----> Pruning devDependencies
yarn install v1.22.19
[1/4] Resolving packages...
[2/4] Fetching packages...
warning Pattern ["debug@>= 2.6.9 < 3.0.0 || >= ^3.1.0"] is trying to unpack in the same destination "/tmp/yarncache.zNC7V/v6/npm-debug-3.2.7-72580b7e9145fb39b6676f9c5e5fb100b934179a-integrity/node_modules/debug" as pattern ["debug@^3.1.0","debug@^3.2.7","debug@^3.2.7"]. This could result in non-deterministic behavior, skipping.
[3/4] Linking dependencies...
[4/4] Building fresh packages...
warning Ignored scripts due to flag.
Done in 3.57s.
-----> Caching build
- yarn cache
-----> Build succeeded!
-----> Discovering process types
Procfile declares types -> (none)
Default types for buildpack -> web
-----> Compressing...
Done: 70.9M
-----> Launching...
Released v3
https://myparseserver.herokuapp.com/ deployed to Heroku
Any form of pointer to what the issue is would be greatly appreciated.
Thank you
Could you provide an example?
did you set the environment variables?
you can use ethers as you use it directly in nodejs, when you self host you can import any external library in cloud code
Yes, I did that, and I have as well restarted the server (dyno), yet it still throws the same error
it is expected to use node 18?
and how about solving the āanyā type returned by Web3API? . For example, this line
var NFTs = await Parse.Web3API.account.getNFTs(options);
when we try to access the object, like
NFTs.result[i].token_id
the build says
I donāt know this part, you can use directly Moralis v2 sdk in cloud code if you want
you are calling this in cloud?