-
Notifications
You must be signed in to change notification settings - Fork 20.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Blockchain download time #1442
Comments
Downloading the chain is not the issue, is problem is processing it. 3-9 hours is a very optimistic estimate. (i.e. throw a i7+ 4/8 cores at it, and maybe it will achieve that performance). On a moderately powerful computer it will take way more time. As far as I know, a t2.medium is not enough with the test net spam. Maybe if you get through all the junk to the head it will stay in sync, but it will take quite some time to get there. (e.g. on an older core 2 CPU (as far as I know more powerful than t2.medium) it took at least 1-2 days to sync up). So, this isn't really an issue with geth, but rather the amount of junk we accumulated in the test network. Running full nodes will always be quite computational heavy, but hopefully after Frontier we can release eth/63 and enable light clients to join the network and not require so many resources. Until then however, please have patience :), and maybe throw a buigger instance at it while it finishes up syncing and then reduce it to see how much is just enough to keep in sync :) |
Or ask someone you trust to make you a copy of the chain |
A copy would be appreciated. :) |
OK, so I'm now running on an M4 Large. Still creeping along. |
Ensure the volume that the datadir resides on has sufficient I/O. Very low default iops can result in disk bottlenecks |
Feel like I'm chasing a dream. Has anyone actually recently (last few days) downloaded the current blockchain start to finish or are we just guessing and suggesting more horsepower and faster disk? Hope production looks better than this. |
I did a full sync about 3 days ago on a Core2 laptop. Took me 2 days (on an On Wed, Jul 8, 2015 at 8:43 PM, Gagabits [email protected] wrote:
|
I sync'd the 3 Go Bootnodes listed on stats.ethdev.com and my 13" Macbook Pro all within the last week, all from scratch |
* Make tx pool fee check logic match the logic in state_transition.go * Improve non-native currency fee tx handling (ethereum#2) Improves multi-currency fee support in the tx pool queue and pending lists. Previously when paying for a tx in a non-native currency, the balance of the account would only be checked when inserting the transaction into the tx pool, but as the balance of the account can change after the tx is admitted, transactions are periodically checked for validity in `txList.Filter`. This PR changes txList to properly track balances in non-native currencies. This also adds support for checking the gas price minimum in the txList. This PR also converts gas prices to CELO prior to comparing if a new transaction at the same nonce should override an old transaction at the same nonce. It also refactors calculations of `tx.Fee()` * Fix nil pointer exception in txList Filter This occurred because gasPriceMinimums contains the gasPriceMinimum for every fee currency, not just the fee currencies of the transfer. This changes the iteration to go through each gasPriceFloor and then get the gasPriceMinimum from the fee currency associated with the floor. This works because gasPriceMinimums contains every whitelisted fee currency and the tx has already been validated to pay in a fee currency. * Fix panic in 'ConvertToGold()' Co-authored-by: Joshua Gutow <[email protected]> Co-authored-by: Joshua Gutow <[email protected]>
Stephen Tual advises that the download should take from 3-9 hours to complete. Have been running since 1:15 GMT and currently (14:00 GMT) at 619K and crawling. Running on amazon ec2 medium. Load average is steady over 3. Server is 'Caloosahatchee' on the stats page. Have not yet been able to download a complete blockchain.
The text was updated successfully, but these errors were encountered: