(Bug) Transaction runs out of gas during deployment step

[fvictorio]

The deployment step fails because one of the transactions runs out of gas (during the first call to setLastCrowdsaleRecursive, I think).

One possible solution is to remove the gasLimits in the code and rely on the inferred gas cost, that should work fine because these transactions should be predictable.

I can send a PR with this change if you think it's a good solution.

[vbaranov]

gasLimit was set in this commit for contract methods, due to the issue in mainnet: when wizard hangs on the one of the method of contract. If it won't cause problems in the mainnet, I vote for this solution. If not, as a workaround, let's increase gasLimit for this tx.

[fvictorio]

I can test it in Kovan to see that it doesn't causes problems. Is that enough or do you test this kind of things in the mainnet too, using real ether?

[igorbarinov]

@fvictorio we can deploy it on stage without review and I'll test it on mainnet.
Also, please post here your wallet and I'll send you some mainnet eth for testing

[fvictorio]

I've thought about this a little more and I don't think my idea will work. The networks have different gas limits by default, and web3 doesn't have an option to set a new default gas limit: you have to do it in each transaction. This, and the fact that a transaction sometimes exceeds the default of 90000, makes it necessary to use gas limits per transaction anyway.

So, I'll send a PR with updated gas limits for now.

@igorbarinov I created an address at 0x4a9a263d09f3722f5a97619ca77816abdb63d8fd