elective-stereophonic
elective-stereophonic
uMGW Technical Discussion - Phase 1 - PoC - "Testnet -> Mainnet" - Doge and VRC  
Please login or register.

Login with username, password and session length
Advanced search  

News:

Latest Stable Nxt Client: Nxt 1.12.1 Upgrade before block 2870000 is mandatory!

Pages: 1 [2] 3 4 ... 12  All

Author Topic: uMGW Technical Discussion - Phase 1 - PoC - "Testnet -> Mainnet" - Doge and VRC  (Read 25487 times)

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

Oooooh..okay.  I'll put txindex into the DOGE.conf.

However, I'm thinking I figured out the issue.  It looks like when I switched from doge testnet to mainnet, I needed to delete everything out of my ~/.dogecoin folder(except the .conf file).  After loading dogecoind backup, and run UMGW, again - NO ISSUES! :D  Lesson learned! :D

Btw, is there a graceful way to exit uMGW?  I've been using ctrl-c.
Logged

Steve

  • Sr. Member
  • ****
  • Karma: +28/-0
  • Offline Offline
  • Posts: 349
  • Steve please
    • View Profile

Will take note of this!
Logged

jl777

  • Hero Member
  • *****
  • Karma: +718/-123
  • Offline Offline
  • Posts: 6170
    • View Profile

Oooooh..okay.  I'll put txindex into the DOGE.conf.

However, I'm thinking I figured out the issue.  It looks like when I switched from doge testnet to mainnet, I needed to delete everything out of my ~/.dogecoin folder(except the .conf file).  After loading dogecoind backup, and run UMGW, again - NO ISSUES! :D  Lesson learned! :D

Btw, is there a graceful way to exit uMGW?  I've been using ctrl-c.
please document everything! I know there are a lot of things I do to make it work, but I dont even know what I know that lets me make it work.

pkill -15 MGW from a different window is probably safest
Logged
There are over 1000 people in SuperNET slack! http://slackinvite.supernet.org/ automatically sends you an invite

I am just a simple C programmer

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

Thanks!  I'm making quick and dirty notes of all the gotchas like this. :)

So, I'm still seeing that error for the account:

multigateway new RTblock 1 height.140294 t.20751158 RTtimestamp.20784449
cmd.(requestType=getAccount&account=3547125209856232764) -> {"errorCode":5,"errorDescription":"Unknown account"}

I verified that from my home PC, I can get http://127.0.0.1:6876/nxt?requestType=getAccount&account=1564966558655164996 to report info on the account just fine.  It reports:

{"publicKey":"4644d0f51c128e7cb603ed7761fc0c93edecc696c234b8e6d2639c8e3858fa49","guaranteedBalanceNQT":"0","balanceNQT":"29900000000","accountRS":"NXT-5EL6-TEM2-ND98-32ZFD","account":"1564966558655164996","effectiveBalanceNXT":0,"unconfirmedBalanceNQT":"29900000000","forgedBalanceNQT":"0"}

My MGW.conf file looks just like I've shown you before.  All strictly JSON formatted.  I can send it to you again in a PM?
Logged

jl777

  • Hero Member
  • *****
  • Karma: +718/-123
  • Offline Offline
  • Posts: 6170
    • View Profile

Thanks!  I'm making quick and dirty notes of all the gotchas like this. :)

So, I'm still seeing that error for the account:

multigateway new RTblock 1 height.140294 t.20751158 RTtimestamp.20784449
cmd.(requestType=getAccount&account=3547125209856232764) -> {"errorCode":5,"errorDescription":"Unknown account"}

I verified that from my home PC, I can get http://127.0.0.1:6876/nxt?requestType=getAccount&account=1564966558655164996 to report info on the account just fine.  It reports:

{"publicKey":"4644d0f51c128e7cb603ed7761fc0c93edecc696c234b8e6d2639c8e3858fa49","guaranteedBalanceNQT":"0","balanceNQT":"29900000000","accountRS":"NXT-5EL6-TEM2-ND98-32ZFD","account":"1564966558655164996","effectiveBalanceNXT":0,"unconfirmedBalanceNQT":"29900000000","forgedBalanceNQT":"0"}

My MGW.conf file looks just like I've shown you before.  All strictly JSON formatted.  I can send it to you again in a PM?
testnet is around 20000 blocks ahead of block 140294
as of that block, my guess is that your acct wasnt there yet
just wait for blockchain to catch up, also might have to restart uMGW, it doesnt always handle starting up while blockchain is loading perfectly
Logged
There are over 1000 people in SuperNET slack! http://slackinvite.supernet.org/ automatically sends you an invite

I am just a simple C programmer

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

Steve, can you get your NRS 1.1.6 on testnet passed block 140000?  Mine keeps reporting done at that block...I'm thinking that's what's causing me grief with my account.  Might have to do with the new code for the digital assets store...I saw something similar with my client on my PC.

UPDATE: I blew away my nrs database so I could start from scratch, thinking maybe it got corrupted with all the shutdowns, rebooting, etc.  It's been siting at  Checksum passed at block 76500 for twenty minutes.  I'm going to let it load overnight, and check it in the morning.  You having better luck with testnet under 1.1.6, Steve?
« Last Edit: July 23, 2014, 02:37:02 am by cobaltskky »
Logged

LocoMB

  • Hero Member
  • *****
  • Karma: +101/-37
  • Offline Offline
  • Posts: 751
    • View Profile

Steve, can you get your NRS 1.1.6 on testnet passed block 140000?  Mine keeps reporting done at that block...I'm thinking that's what's causing me grief with my account.  Might have to do with the new code for the digital assets store...I saw something similar with my client on my PC.

UPDATE: I blew away my nrs database so I could start from scratch, thinking maybe it got corrupted with all the shutdowns, rebooting, etc.  It's been siting at  Checksum passed at block 76500 for twenty minutes.  I'm going to let it load overnight, and check it in the morning.  You having better luck with testnet under 1.1.6, Steve?

Don't know if it is related - I may be having similar issues:

I am running 1.2.1e on testNet, and it is getting stuck at
2014-07-23 07:58:55 INFO: Nxt server 1.2.1e started successfully.
2014-07-23 07:58:55 INFO: RUNNING ON TESTNET - DO NOT USE REAL ACCOUNTS!
2014-07-23 07:58:58 INFO: nxt.apiServerEnforcePOST = "true"
2014-07-23 08:01:57 INFO: Checksum passed at block 30000
2014-07-23 08:09:07 INFO: Checksum passed at block 76500

- but when I log in to the client, it has blocks up to date, only with insane high base targets and only two accounts generating the blocks.

Are we seeing the same thing?





upsate : the asset I had issued has disappeared, so it looks like i was on a fork at some point at least !
« Last Edit: July 23, 2014, 08:43:01 am by l8orre »
Logged
TOX
90E54E5B5213290EE616D425CADC473038CFABFA53C913271AA8559D1937DC4AF3A354A9E4E5

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

l80rre, on my main PC, I have a height of 163039 with NXT-AMPQ-B6ZZ-S8TL-EFDBH generating most of the blocks.  Looks like your NRS on testnet hasn't DLed the full block chain, or you're on a fork....Nlock gen time for me is 91 seconds.  Are you on 1.2.0e?  I find that the 1.2.0e NRS gets passed block 140000 just fine on my NRS nodes.  It's 1.1.6 that has the problems with stalling.
« Last Edit: July 23, 2014, 10:43:55 am by cobaltskky »
Logged

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

James, I'm pretty sure the issue that I'm running into on testnet is the inability for 1.1.6 to make it past block 140000.  It simply stops processing.  I saw this on my main PC too - 1.1.6 would start barking about blocks with Digital Goods info in them.

And I can't run uMGW against 1.2.0e because of the change in API calls.

What are your thoughts on how we can get around this?  How much risk would be involved with setting up uMGW directly on Mainnet, and testing with it there?   I'm thinking that the only way someone can get to the assets from the NXT client would be by using longzai's latest update to the UI, and adding the uMGW info to their client anyway - so I think we'd be pretty safe bringing it up on mainnet.

I guess another question would be - if I bring up my cluster on mainnet for doge, test it, then want to replace two of my nodes with nodes operated by other people, will that cause problems swapping out nodes like (assuming the MGW.conf file is configured correctly)that?
Logged

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

James, please regard my above questions.  I think I have a better idea of how all this works now. :)  I'm thinking we'll trash the idea of me getting my own cluster up on testnet, or my own cluster up on mainnet, and we'll move forward with just trying to get a co-owned cluster up on mainnet.  It will be hidden from our customers anyway until the MGW asset is added to their MGWlite client.

I had a quick question - can the gateway_sig contain alpha characters?
Logged

jl777

  • Hero Member
  • *****
  • Karma: +718/-123
  • Offline Offline
  • Posts: 6170
    • View Profile

James, please regard my above questions.  I think I have a better idea of how all this works now. :)  I'm thinking we'll trash the idea of me getting my own cluster up on testnet, or my own cluster up on mainnet, and we'll move forward with just trying to get a co-owned cluster up on mainnet.  It will be hidden from our customers anyway until the MGW asset is added to their MGWlite client.

I had a quick question - can the gateway_sig contain alpha characters?
Please stick to MGW design
three independent operators
gateway_sig must be numerical
Logged
There are over 1000 people in SuperNET slack! http://slackinvite.supernet.org/ automatically sends you an invite

I am just a simple C programmer

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

Agreed, sticking with original design of three independent operators.
Logged

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

Alright, James - I did as you suggested.  I got my cluster up with uMGW using MIC as the asset and dogecoind as the coin daemon.  They all came up, they seem to have been able to log into the new NXT accounts that I created for this purpose.  I funded the NXT accounts with some MIC and some NXT.  The output that I see from uMGW looks like this:

multigateway new RTblock 13 height.192004 t.20862493 RTtimestamp.20863577
finished multigateway new RTblock 13 height.192004 t.20862493 RTtimestamp.20863577
F.1 H.1 t.20862545 height.192005 192012 lastblock.(14710856213943008774) -> nextblock.(8030700231067307146)

multigateway new RTblock 14 height.192005 t.20862545 RTtimestamp.20863837
finished multigateway new RTblock 14 height.192005 t.20862545 RTtimestamp.20863837
F.1 H.1 t.20863067 height.192006 192012 lastblock.(8030700231067307146) -> nextblock.(5616807512449057635)

multigateway new RTblock 15 height.192006 t.20863067 RTtimestamp.20863837
finished multigateway new RTblock 15 height.192006 t.20863067 RTtimestamp.20863837

And it just keeps doing that.  Is that the "ready state" that you'd expect?
Logged

jl777

  • Hero Member
  • *****
  • Karma: +718/-123
  • Offline Offline
  • Posts: 6170
    • View Profile

Alright, James - I did as you suggested.  I got my cluster up with uMGW using MIC as the asset and dogecoind as the coin daemon.  They all came up, they seem to have been able to log into the new NXT accounts that I created for this purpose.  I funded the NXT accounts with some MIC and some NXT.  The output that I see from uMGW looks like this:

multigateway new RTblock 13 height.192004 t.20862493 RTtimestamp.20863577
finished multigateway new RTblock 13 height.192004 t.20862493 RTtimestamp.20863577
F.1 H.1 t.20862545 height.192005 192012 lastblock.(14710856213943008774) -> nextblock.(8030700231067307146)

multigateway new RTblock 14 height.192005 t.20862545 RTtimestamp.20863837
finished multigateway new RTblock 14 height.192005 t.20862545 RTtimestamp.20863837
F.1 H.1 t.20863067 height.192006 192012 lastblock.(8030700231067307146) -> nextblock.(5616807512449057635)

multigateway new RTblock 15 height.192006 t.20863067 RTtimestamp.20863837
finished multigateway new RTblock 15 height.192006 t.20863067 RTtimestamp.20863837

And it just keeps doing that.  Is that the "ready state" that you'd expect?
yes!
When you start making deposits, more printouts will appear
Logged
There are over 1000 people in SuperNET slack! http://slackinvite.supernet.org/ automatically sends you an invite

I am just a simple C programmer

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

YAY. :}
Logged

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

FYI - turns out this error:

Code: [Select]
null rawtransaction
error decoding.((null))
DOGE post_process_bitcoind_RPC (getrawtransaction) error.{"result":null,"error"},"id":"jl777"}

null rawtransaction
error decoding.((null))
DOGE post_process_bitcoind_RPC (getrawtransaction) error.{"result":null,"error"},"id":"jl777"}

null rawtransaction
error decoding.((null))
ignorelist.0x7fa905a3c010 lastignore.309274 Start RTcoinloop.DOGE cp->height 309309 vs latest 309312
calc_nxt64bits: illegal char.(e 101) in (true).3
WARNING: got height.76753 vs expected h.76938 for 10441688513471843606
WARNING: got height.76752 vs expected h.76937 for 15386774980726572929
WARNING: got height.76751 vs expected h.76936 for 11496175141813315609
WARNING: got height.76750 vs expected h.76935 for 5304899106037837118

wasn't caused by switching from doge's testnet to mainnet.  It was caused by me using ctrl-c to terminate the uMGW daemon.  WHOOPS.

pkill -15 uMGW from another screen session appears to prevent this from happening, so far.
Logged

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

Hey, James - it looks like I didn't actually have umgw up and running correctly before.  Long story.  But I think I got dogecoind and NXt 1.1.6 on mainnet running properly, with the fullblockchain DLed and up to date.

Now when I execute .uMGW, it runs through a bunch of the processing that I see before, then it starts a bunch of line like this:

DOGE block.307749 blockhash.2e37945f5ec31be764fe206071ace9e42cdb109410773269182c1f78f02b4b84 | lag.3180
DOGE block.307750 blockhash.750686e39464ffa80e853a3fd32f5d468da1ede8f901833e072dfe9c2d80a083 | lag.3179
DOGE block.307751 blockhash.d7bb4f38a6477570f6ea11154d84a5f8f43c49d3eee74322a7e6cfb4b8593ae3 | lag.3178
DOGE block.307752 blockhash.1db3027eba3f1f43c196f67cbf9a1ea7f7953695e4e00881dceb1393399ba9b9 | lag.3177
DOGE block.307753 blockhash.23adb3bb293d21b3101a2d516320b14bb8621cd6a3d5ba3cc0544c25d51f0280 | lag.3176
DOGE block.307754 blockhash.c6fecab549423d2fbd53587e494da470d152d827226038001cffe5e6724cd590 | lag.3175
Segmentation fault (core dumped)

And then it throws that segmentation fault.  It's the same block (307754) on each of my nodes.  What are your thoughts?
Logged

jl777

  • Hero Member
  • *****
  • Karma: +718/-123
  • Offline Offline
  • Posts: 6170
    • View Profile

Hey, James - it looks like I didn't actually have umgw up and running correctly before.  Long story.  But I think I got dogecoind and NXt 1.1.6 on mainnet running properly, with the fullblockchain DLed and up to date.

Now when I execute .uMGW, it runs through a bunch of the processing that I see before, then it starts a bunch of line like this:

DOGE block.307749 blockhash.2e37945f5ec31be764fe206071ace9e42cdb109410773269182c1f78f02b4b84 | lag.3180
DOGE block.307750 blockhash.750686e39464ffa80e853a3fd32f5d468da1ede8f901833e072dfe9c2d80a083 | lag.3179
DOGE block.307751 blockhash.d7bb4f38a6477570f6ea11154d84a5f8f43c49d3eee74322a7e6cfb4b8593ae3 | lag.3178
DOGE block.307752 blockhash.1db3027eba3f1f43c196f67cbf9a1ea7f7953695e4e00881dceb1393399ba9b9 | lag.3177
DOGE block.307753 blockhash.23adb3bb293d21b3101a2d516320b14bb8621cd6a3d5ba3cc0544c25d51f0280 | lag.3176
DOGE block.307754 blockhash.c6fecab549423d2fbd53587e494da470d152d827226038001cffe5e6724cd590 | lag.3175
Segmentation fault (core dumped)

And then it throws that segmentation fault.  It's the same block (307754) on each of my nodes.  What are your thoughts?
strange. If its always the same block, try setting the blockheight for DOGE past that block, I will investigate this block and see if it has anything funny in it
Logged
There are over 1000 people in SuperNET slack! http://slackinvite.supernet.org/ automatically sends you an invite

I am just a simple C programmer

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

I set it for 308000, now my nodes bomb at 309249.  Maybe something funky in those blocks...I'll try even higher just to see if I can get these loaded.
Logged

cobaltskky

  • Board Moderator
  • Hero Member
  • ****
  • Karma: +226/-138
  • Offline Offline
  • Posts: 2344
    • View Profile

Bombed at 310395 this time.
Logged
Pages: 1 [2] 3 4 ... 12  All
 

elective-stereophonic
elective-stereophonic
assembly
assembly