# How to recover from a timeout due to network congestion when deploying an upgradeable contract on mainnet?

**URL:** <https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049>\
**Category:** SDK\
**Created:** [June 9, 2020, 3:18am UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049 "2020-06-09T03:18:29Z")\
**Posts on this page:** 12\
**Page:** 1

<div class="post-metadata">

**Author:** ![h00j](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/h00j/32/1725_2.png) [@h00j](https://forum.openzeppelin.com/u/h00j)\
**Post date:** [June 9, 2020, 3:18am UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/1 "2020-06-09T03:18:29Z")

</div>

It seems that there are currently longer wait times for transactions on mainnet lately. My question is, I’ve created a contract at this address [https://etherscan.io/address/0xcc2850f5842127964b18009db1f48033e9fda2c9](https://etherscan.io/address/0xcc2850f5842127964b18009db1f48033e9fda2c9)

I wasn’t given the option to initialize the contract because the CLI timed out due to the long transaction wait time. My question is how do I recover and allow the sdk to call initialize on my function and create my proxy contract?

Here is my project.json

```auto
    {
        "manifestVersion": "2.2",
        "contracts": {
            "MyToken": "MyToken"
         },
         "dependencies": {},
         "name": "MyTokenProject",
         "version": "1.0.2",
         "compiler": {
         "compilerSettings": {
             "optimizer": {
                 "enabled": false,
                 "runs": "200"
             }
          },
         "typechain": {
             "enabled": false
          },
          "manager": "openzeppelin",
          "solcVersion": "0.5.17",
          "artifactsDir": "build/contracts",
          "contractsDir": "contracts"
        },
       "telemetryOptIn": false
    }

```

---

<div class="post-metadata">

**Author:** ![abcoathup](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/abcoathup/32/415_2.png) [@abcoathup](https://forum.openzeppelin.com/u/abcoathup)\
**Post date:** [June 9, 2020, 4:44am UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/2 "2020-06-09T04:44:25Z")

</div>

Hi @h00j,

I am sorry to hear that you are being impacted by network congestion.

When an upgradeable project is first deployed, three contracts are deployed.

1. The logic contract
2. The ProxyAdmin
3. The proxy contract pointing to the logic contract along with calling the initialization logic.

Looking at your transactions, I can only see one contract being deployed, so assume this is the logic contract.

I don’t think this can be easily recovered from, as I assume you still need to deploy the ProxyAdmin and the proxy contract, along with calling the initialization.

You may need to redeploy if you want to interact with the CLI. Though you would need to use an appropriate gas price given the current network congestion. ([https://www.ethgasstation.info/](https://www.ethgasstation.info/)). I noticed that you used 20 Gwei, which appears to be too low during current congestion.

Another community member is also running into this issue: see [2 step process in proxy creation](https://forum.openzeppelin.com/t/2-step-process-in-proxy-creation/3038/2)

---

<div class="post-metadata">

**Author:** ![abcoathup](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/abcoathup/32/415_2.png) [@abcoathup](https://forum.openzeppelin.com/u/abcoathup)\
**Post date:** [June 10, 2020, 3:10am UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/3 "2020-06-10T03:10:40Z")

</div>

Hi @h00j,

As the logic contract has already been deployed (the equivalent of an `oz add` and `oz push`), it should be possible to store the address in the [`<network_name>.json`](https://docs.openzeppelin.com/cli/2.8/configuration#network.json) and then rerun the same command. The CLI should detect that the logic contract is already deployed and won’t deploy it again. This involves manually modifying the `<network_name>.json` file.

The deploy would still involve deploying two transactions, one for the ProxyAdmin and a second for the proxy contract pointing to the logic contract along with calling the initialization logic.

I recommend putting your [Configuration Files in Version Control](https://docs.openzeppelin.com/cli/2.8/configuration#configuration-files-in-version-control) before doing any modifications to your `<network_name>.json` file.

You could check this process first on a public testnet.

Alternatively you could rerun the mainnet deployment.

Please be aware that the network is still congested ([https://www.ethgasstation.info/](https://www.ethgasstation.info/)) so you will need to adjust your gas prices.

---

<div class="post-metadata">

**Author:** ![h00j](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/h00j/32/1725_2.png) [@h00j](https://forum.openzeppelin.com/u/h00j)\
**Post date:** [June 11, 2020, 1:41am UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/4 "2020-06-11T01:41:36Z")

</div>

Thanks I’ll give that a try and let you know how that goes.

---

<div class="post-metadata">

**Author:** ![h00j](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/h00j/32/1725_2.png) [@h00j](https://forum.openzeppelin.com/u/h00j)\
**Post date:** [June 11, 2020, 2:55am UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/5 "2020-06-11T02:55:26Z")

</div>

So I tested doing `oz add` and `oz push` on testnet. But I’m at a loss on how I create the Proxy Contract and ProxyAdmin contract?

Running `oz deploy` creates a brand new contract, and when I run `oz upgrade` I get this output

```auto
Nothing to compile, all contracts are up to date.
All implementations are up to date
No upgradeable contract instances that match contract MyToken were found

```

Even though MyToken has a function with the `initializer` keyword on it.

---

<div class="post-metadata">

**Author:** ![abcoathup](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/abcoathup/32/415_2.png) [@abcoathup](https://forum.openzeppelin.com/u/abcoathup)\
**Post date:** [June 11, 2020, 4:27am UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/6 "2020-06-11T04:27:01Z")

</div>

Hi @h00j,

When deploying an upgradeable project we do the following:

1. The logic contract
2. The ProxyAdmin
3. The proxy contract pointing to the logic contract along with calling the initialization logic.

I assume that your logic contract has been deployed on mainnet but timed out waiting for confirmation of the transaction so the ProxyAdmin and proxy contract haven’t been deployed.

You will likely need to manually update your `mainnet.json` file with the address of your logic contract or create a `mainnet.json` if it doesn’t exist.

If `mainnet.json` doesn’t exist then you could create a `ganache-cli --fork` of mainnet and do an `oz add` and `oz push` to create the `<network>/json` that you could use.  
See [https://blog.openzeppelin.com/testing-real-world-contract-upgrades/](https://blog.openzeppelin.com/testing-real-world-contract-upgrades/) for the concepts.

Once you have `mainnet.json` configured with the address of the logic contract deployed, when you run `npx oz deploy` it should deploy ProxyAdmin and the proxy and allow you to initialize.

* * *

`npx oz upgrade` will only try to upgrade if you already have an upgradeable contract deployed and you have changed the logic contract.

* * *

To avoid having to do the above, you could rerun the mainnet deployment (if you don’t have a `mainnet.json`) but you would have to pay to redeploy the logic contract.

Please be aware that the network is still congested ([https://www.ethgasstation.info/](https://www.ethgasstation.info/)) so you will need to adjust your gas prices.

---

<div class="post-metadata">

**Author:** ![h00j](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/h00j/32/1725_2.png) [@h00j](https://forum.openzeppelin.com/u/h00j)\
**Post date:** [June 13, 2020, 12:03am UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/7 "2020-06-13T00:03:39Z")

</div>

Ok so I tested the process on ropsten. I pushed just the contract and was given the output in my ropsten.json file.

```
"contracts": {
    "MyContract": {
        "address": "0x25fc0F482BfC04EDEF6338D96B48C1773F15DEfD",
        "constructorCode": "6080604052613fb9806100136000396000f3fe",
        "bodyBytecodeHash": "66b4dbc98cdb4359b245a6a37a6d6755a1c2e4ba9c219bd4ce0d64b893caf514",
        "localBytecodeHash": "25d378488af3ba7e78b8b8aafd93c62b415b60da48855181e9a70ff5dfb5545e",
        "deployedBytecodeHash": "25d378488af3ba7e78b8b8aafd93c62b415b60da48855181e9a70ff5dfb5545e",
        "types": { /*omitted for brevity*/ },
        "storage": { /*omitted for brevity*/ },
        "warnings": {
            "hasConstructor": false,
            "hasSelfDestruct": false,
            "hasDelegateCall": false,
            "hasInitialValuesInDeclarations": false,
            "uninitializedBaseContracts": []
         }
    }
}

```

I was able to get the proxy contract and proxy admin after running `npx oz deploy`  
I’m assuming that I can copy the `"contracts"` section of my json output here point it to the address that is on mainnet and run deploy and I should get the proxy contract and proxy admin?

---

<div class="post-metadata">

**Author:** ![abcoathup](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/abcoathup/32/415_2.png) [@abcoathup](https://forum.openzeppelin.com/u/abcoathup)\
**Post date:** [June 15, 2020, 2:33am UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/8 "2020-06-15T02:33:43Z")

</div>

Hi @h00j,

I assume that you can copy the `contracts` section of the network json to mainnet and manually update the address. I recommend testing this against a ganache-cli fork of mainnet first, see [https://blog.openzeppelin.com/testing-real-world-contract-upgrades/](https://blog.openzeppelin.com/testing-real-world-contract-upgrades/) for the concepts.

---

<div class="post-metadata">

**Author:** ![h00j](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/h00j/32/1725_2.png) [@h00j](https://forum.openzeppelin.com/u/h00j)\
**Post date:** [June 16, 2020, 1:34am UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/9 "2020-06-16T01:34:57Z")

</div>

I went through the process of `add`ing `push`ing the contract and then `deploy`ing the proxy admin and proxy contract with testnet and a fork of mainnet and everything worked out smoothly. Going to try it on mainnet next.

---

<div class="post-metadata">

**Author:** ![abcoathup](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/abcoathup/32/415_2.png) [@abcoathup](https://forum.openzeppelin.com/u/abcoathup)\
**Post date:** [June 16, 2020, 1:53am UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/10 "2020-06-16T01:53:37Z")

</div>

Hi @h00j,

Great to hear. Glad you tested on a fork of mainnet.

---

<div class="post-metadata">

**Author:** ![h00j](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/h00j/32/1725_2.png) [@h00j](https://forum.openzeppelin.com/u/h00j)\
**Post date:** [June 16, 2020, 9:42pm UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/11 "2020-06-16T21:42:34Z")

</div>

Just verifying, the solution worked on mainnet also. Thanks!

---

<div class="post-metadata">

**Author:** ![abcoathup](https://sea2.discourse-cdn.com/flex022/user_avatar/forum.openzeppelin.com/abcoathup/32/415_2.png) [@abcoathup](https://forum.openzeppelin.com/u/abcoathup)\
**Post date:** [June 16, 2020, 11:16pm UTC](https://forum.openzeppelin.com/t/how-to-recover-from-a-timeout-due-to-network-congestion-when-deploying-an-upgradeable-contract-on-mainnet/3049/12 "2020-06-16T23:16:44Z")

</div>

Great to hear. Thanks for letting me know.