[ANNOUNCE] Abe 0.7: Open Source Block Explorer Knockoff - page 5.

John Tobey

hero member

Activity: 481

Merit: 529

Quote from: pointychimp on July 13, 2013, 11:40:01 AM

Is there any more information you need? Is there someway I can tell abe to connect to bitcoind with RPC over a specific port?

Sorry for the delayed reply. Please post the last few lines of output before "catch_up_rpc: abort" or "Exception: RPC load failed." Have you run Bitcoin with -reindex since adding txindex to bitcoin.conf?

RPC is slower for the initial load, I think over twice as slow as blockfiles. Once caught up, it usually inserts blocks faster than the blockfile method, since RPC has already loaded most transactions by the time the block comes (assuming that you arrange for it to catch up frequently).

Abe normally does not "read the blockfiles over and over again": it stores the current position in the database.

pointychimp

newbie

Activity: 20

Merit: 0

Hey I've been having more luck since starting fresh with abe and the blockchain. I'm trying to get abe to use RPC but it always says either "catch_up_rpc: abort" or "Exception: RPC load failed." The bitcoin client is the first time I've ever seen RPC so I am very unfamiliar with it. It does seem, however to be a better way for abe to update the database instead of reading the blockfiles over and over again.

So my question is this. How do I make it work? I have a ~/.bitcoin/bitcoin.conf which contains the rpc username and password, and told it to listen for rpc on port 8332 (plus other settings). That port is correctly forwarded in my router. In abe.conf I have (among other things)

datadir += [{
"dirname": "/home//.bitcoin",
"loader": "rpc", # See the comments for default-loader below.
"chain": "Bitcoin"
}]

Is there any more information you need? Is there someway I can tell abe to connect to bitcoind with RPC over a specific port?

Thanks for your help.

Matt

coinpipe

newbie

Activity: 2

Merit: 0

Ah got it! Thanks a bunch.

John Tobey

hero member

Activity: 481

Merit: 529

Quote from: coinpipe on June 25, 2013, 02:03:16 PM

Thanks for ABE. I've been trying it out with a few different alt-coins. I could do with a hand understanding address_version.

For most coins it seems to be obvious, just the unicode of the first character of an address. But some coins have addresses like this (feathercoin in this example)

6nh62XDjhbqjjit78jfStECNcHz89xbNeq
72HzQ77Xbpg67MadosyWDjd5YZsVHz4Npt

etc...

Plugging it into decode_address, for example:

http://explorer.doubloons.net/chain/Doubloons/q/decode_address/6nh62XDjhbqjjit78jfStECNcHz89xbNeq

gives "0e" as the hex version byte for both those addresses, so I would think it is "\u000e".

coinpipe

newbie

Activity: 2

Merit: 0

Thanks for ABE. I've been trying it out with a few different alt-coins. I could do with a hand understanding address_version.

For most coins it seems to be obvious, just the unicode of the first character of an address. But some coins have addresses like this (feathercoin in this example)

6nh62XDjhbqjjit78jfStECNcHz89xbNeq
72HzQ77Xbpg67MadosyWDjd5YZsVHz4Npt

etc...

I've tried settings an address_version of "\u0036" (a unicode '6') and this seems to work fine, however I'm worried that is not right. Is \u0036 ok, or should I do something else?

macintosh

newbie

Activity: 28

Merit: 0

thank you for this!

pointychimp

newbie

Activity: 20

Merit: 0

Quote from: John Tobey on June 17, 2013, 09:27:43 PM

Quote from: pointychimp on June 17, 2013, 04:22:29 PM

I keep getting a "RuntimeError: maximum recursion depth exceeded" error. It happens every time I start abe up, and every time I access a page the error shows up in the console.

This has been on the to-do list for a long time. I'll give it another look, but as I recall, it would take the better part of a day to code and test.

First, run Abe once with --rescan in case that solves it. If you know Python, you can try sys.setrecursionlimit(300000) as suggested here. (Replace 300000 with a few more than the number of blocks in your chain, or try lower numbers until either success or another kind of crash.)

I think this happens when the block files are "very" out of sequence: when a sequence of 1000 or more blocks all occur before one of their common ancestors. Could this be the case with your block files? Maybe this is normal for the latest Bitcoin versions, which I have not tested. Is there anything unusual about how your block files were created, or did you download the chain again after running Abe the first time?

I don't think I did anything unusual. I think I did copy block files from one of my computers to my server in order to speed up the chain download time. Problem is I don't remember if I did that before running abe for the first time or after. I'm currently letting abe rescan. If this doesn't work, I might just bite the bullet, remove everything from the mySQL database, delete my .bitcoin/ directory, download the blockchain the conventional way, and try again.

Thanks for the help. I'll try to remember to report back if I ever get this solved!

John Tobey

hero member

Activity: 481

Merit: 529

Quote from: bongwater on June 17, 2013, 09:34:39 PM

Open source block explorer? can you explain what this is?

https://en.bitcoin.it/wiki/Block_chain_browser

macintosh

newbie

Activity: 28

Merit: 0

lovin this idea, goodluck

bongwater

newbie

Activity: 14

Merit: 0

Open source block explorer? can you explain what this is?

John Tobey

hero member

Activity: 481

Merit: 529

Quote from: pointychimp on June 17, 2013, 04:22:29 PM

I keep getting a "RuntimeError: maximum recursion depth exceeded" error. It happens every time I start abe up, and every time I access a page the error shows up in the console.

This has been on the to-do list for a long time. I'll give it another look, but as I recall, it would take the better part of a day to code and test.

First, run Abe once with --rescan in case that solves it. If you know Python, you can try sys.setrecursionlimit(300000) as suggested here. (Replace 300000 with a few more than the number of blocks in your chain, or try lower numbers until either success or another kind of crash.)

I think this happens when the block files are "very" out of sequence: when a sequence of 1000 or more blocks all occur before one of their common ancestors. Could this be the case with your block files? Maybe this is normal for the latest Bitcoin versions, which I have not tested. Is there anything unusual about how your block files were created, or did you download the chain again after running Abe the first time?

pointychimp

newbie

Activity: 20

Merit: 0

I keep getting a "RuntimeError: maximum recursion depth exceeded" error. It happens every time I start abe up, and every time I access a page the error shows up in the console. I start abe with the following command:

python -m Abe.abe --config abe-my.conf

and the (censored) contents of abe-my.conf are as follows:

dbtype MySQLdb
connect-args {"user":"*****","db":"*****","passwd":"*****"}
upgrade
host *****
port 2750
datadir = /****/****/.bitcoin

What more info do you need, if any? Is there something in python I have to change?

mullick

legendary

Activity: 1064

Merit: 1002

Any chance this would be easy enough to adapt for a newbie with a tutorial? I have only ever done a hello world and simple calculator in java but get enough to follow simple directions enough. I could offer someone a few Ltc to make a tutorial. I may be way out of my league he but It's worth a shot

John Tobey

hero member

Activity: 481

Merit: 529

Quote from: ttsda on April 16, 2013, 04:37:49 PM

Quote from: John Tobey on April 16, 2013, 03:54:41 PM

If you have good SQL skills and a few hours to spare, I can outline the steps. But if you can afford to wait for the reload, that will be much easier.

Please do! I think I can do it if you give me a few instructions

Answered via PM. If anything useful results from this exercise, one of us will post it.

ttsda

newbie

Activity: 8

Merit: 0

Quote from: John Tobey on April 16, 2013, 03:54:41 PM

If you have good SQL skills and a few hours to spare, I can outline the steps. But if you can afford to wait for the reload, that will be much easier.

Please do! I think I can do it if you give me a few instructions

John Tobey

hero member

Activity: 481

Merit: 529

Quote from: ttsda on April 16, 2013, 03:26:54 PM

I loaded the blockchain to an Abe sqlite DB, but now I want to use Postgres.
Is there a way to convert the sqlite DB to pg, or do I have to rescan the whole blockchain again?

This would be a nice feature but is not supported currently. If you have good SQL skills and a few hours to spare, I can outline the steps. But if you can afford to wait for the reload, that will be much easier.

ttsda

newbie

Activity: 8

Merit: 0

I loaded the blockchain to an Abe sqlite DB, but now I want to use Postgres.
Is there a way to convert the sqlite DB to pg, or do I have to rescan the whole blockchain again?

akaihola

newbie

Activity: 13

Merit: 0

Quote from: John Tobey on March 20, 2013, 09:09:01 AM

Quote from: akaihola on March 19, 2013, 11:39:43 PM

Does the "trim spent outputs" feature in the no-statistics branch mean that you can't see the total amount of BTC received for each account?

Any particular use case in mind?

Tracking the total amount received for a bunch of addresses.

John Tobey

hero member

Activity: 481

Merit: 529

Quote from: akaihola on March 19, 2013, 11:39:43 PM

Does the "trim spent outputs" feature in the no-statistics branch mean that you can't see the total amount of BTC received for each account?

Yes, that information is lost. All you see is total received minus total sent. It would be possible to store the total received by each address, but it would only be a lower bound on the real value, since addresses themselves are trimmed when empty.

Any particular use case in mind?

akaihola

newbie

Activity: 13

Merit: 0

Does the "trim spent outputs" feature in the no-statistics branch mean that you can't see the total amount of BTC received for each account?

Topic: [ANNOUNCE] Abe 0.7: Open Source Block Explorer Knockoff - page 5. (Read 28975 times)