[Passed] LiquidBits term 7 grant (updated) - Automatic Liquidity Pool (ALP)

[Cybnate]

Even a small ALP operation requires an operator. The operator fees are then in a bad ratio to the ALP liquidity volume.

Not sure if I understand this right? NuBots also have operator fees. I’m good to reduce the LiquidBits amount from 150 NBT/60 day to 100 NBT/60 days given the amount of work I have on it. However the difference is relatively small, I suppose you are referring to other operators?

[masterOfDisaster]

i was just trying to say that a small ALP operation still requires an operator - in difference to completely removing an ALP operation from an exchange
And I think a standard ALP operator can have more efforts than a passively operating NuBot manager.
So I don’t consider it wrong if an ALP operator requires a higher fee than a NuBot operator.

[Cybnate]

Automatic payment failed yesterday due to Nu deamon misbehaving and requiring a restart. Today it went through fine.
Will do a manual payout for yesterday’s rewards in next 25h.

[Cybnate]

The server has been restarted. The server crashed with a known issue:

ERROR: exception caught in main loop: list index out of range : list index out of range

The bounty to fix this still stands: 100 US-NBT Bounty to fix issue beyond error message ALP pool software

[masterOfDisaster]

Isn’t it about time to sit that error out and plan the transition towards ALP2 with NuBot?

[Cybnate]

I’m aware that the NuBot integration has been released a few days ago, but what about the ALPv2 software? Apparently it has been finished, but haven’t seen an announcement with a link to the latest version. Quick search on the forum and a 3 months scrollback on Gitter didn’t deliver a link other than a test instance.

Do you know where to find a link? Did I miss something, have I been out of the loop given your response?

[masterOfDisaster]

Do you know where to find a link? Did I miss something, have I been out of the loop given your response?

No I don’t, but can have a look.
I don’t think you’ve been out of the loop.
I was trying to say: don’t waste efforts for a system that will likely be discontinued

[Cybnate]

No I don’t, but can have a look.I don’t think you’ve been out of the loop.

Phew.

I was trying to say: don’t waste efforts for a system that will likely be discontinued

Appreciate the concern. But I think it is only discontinued when there is a working and stable replacement available under an open license. So until then I’m interested in fixing issues in current application.

[masterOfDisaster]

So until then I’m interested in fixing issues in current application.

You are right.
Sometimes I might be thinking ahead too far…

[zoro]

Liquidbits server has continuous problems i see. is this connected somehow with running pybot for Poloniex?

[Cybnate]

No, this is an old problem which started to appear in October last year as far as I know. I can’t get my finger on why it happens as it is very random in time. That’s why there is the 100 NBT reward on it, it is not easy to fix.

The other issues are related with Nu daemon terminating/not responding, which started to appear since 2.x. As long as it not more than once a week I need to restart something I’m fine with it. Just inconvenient.

[Cybnate]

Due to the name change of the trading pair and API from NBT to US-NBT on CCEDK it appears to be no longer possible to add new orders with the current instance of LiquidBits on CCEDK. The Bittrex and Poloniex are still working fine at this stage.

Will try to apply a quick-fix after the scheduled payout in 2 hours time. There is a risk if it is not working that a longer outage might be required and that I have to set up a different instance for my PyBots on Bittrex and Poloniex. The software might not be able to easily differentiate the base pairs across exchanges.

My apologies for any issues and downtime in advance. This was an unscheduled change by CCEDK on request of the community and I will try as good as I can to accommodate it.

[Cybnate]

LiquidBits is still down. Issues are proving challenging, server doesn’t respond properly any longer. Logs are not generated.
Will post further details when making progress.

Edit: 7:57 UTC: Server remains unresponsive. Possible other issues. Creating snapshot to secure logs and reverting to backup from earlier today. This may take a while.

Edit: 9:22 UTC: Server restored. PyBots still don’t connect. Server started, no logs. Very strange. Suspect hangover issues from DNS issues 2 days ago. Will restore backup from last week as last resort. Will do this overnight (NZ time).

[masterOfDisaster]

I’m aware that this already creates a lot of efforts, but…
…could this be an opportunity to make a hard cut over to ALPv2?
I mean, the ALP clients need to upgrade, because of the changed CCEDK API, right?

A properly scheduled and announced change would be way better.
But it seems that it’s already too late for that.

[willy]

I mean, the ALP clients need to upgrade, because of the changed CCEDK API, right?

Mh. There are orders on the books and the pair ID’s on the public API haven’t changed, just the names.

I’m not sure what to think of that.

[Cybnate]

I’m not sure what to think of that.

It has me puzzled too. The server has been down for 12h now. They haven’t been cancelled clearly. Doing another attempt to revive the server before I have to attend to other commitments for most of the day.

.could this be an opportunity to make a hard cut over to ALPv2?

It is not that it is ready to fly, had planned to look into it over Easter. It will need to be tested and understood before it runs safely and secure. I’m not a developer and not a full-time system administrator. It will take time.

Edit: it is unclear what is going on, just did a full restore from a backup one week old. I’m unable to start the server.py successfully for CCEDK or Bittrex/Poloniex. The PID runs but there is no logging output and its bots can’t see it. This is really weird. I’m running out of options. Will make another attempt tonight with an even older backup.

[Cybnate]

The services has been restored after the issues with the change from NBT to US-NBT to CCEDK. However an upgrade of the software is mandatory to continue using PyBot on CCEDK.

Existing users, see for a quick update here:

CCEDK trading pairs - other exchanges need to follow

Just published the required changes to support US-NBT on CCEDK with PyBot in a new release on Github here: https://github.com/Cybnate/nu-pool/releases/tag/1.5LB Instructions to update existing installation: 1. Unzip release 2. Copy trading.py and exchanges.py to the folder where your PyBot software resides. Choose overwrite when asked. On Linux, this is usually in your folder under nu-pool in the folder Python. 3. Restart Python, the easiest and surest way to do this is to restart your PC…

Details for a complete upgrade are on the website here: http://cybnate.github.io/index-liquidbits.html#Instruction_for_installing_and_using

[zoro]

I can’t make it work on windows 7.
I am getting the error: unable to receive balance for nbt on ccedk.
I have removed old python installation 2.7.9, perform a restart and install the new python 2.7.11
Has anyone succeeded?

[Cybnate]

The release appeared to be incomplete, made an post here about the problem. CCEDK trading pairs - other exchanges need to follow

[mhps]

CCEDk changed US-NBT to USNBT. does the client need to be changed?