Jump to content

GateHub Update


enej

Recommended Posts

Hi @enej 

i don't think changing the hardware (clusters) fixes anything... the following questions come to mind:

  1. was the 'problem' encountered a software or a protocol problem?
  2. has Ripple made substantive changes to the code base to resolve the problem in the future regardless of hardware and traffic. Note, 'resolve' can include changes made to busy out the 'clients'
  3. are the 'client' software entities properly using the recommended APIs?

I am Generally interested in the nature of the problem and the purported fix.

Are GateHub machines in Amazon, Azure or on premise machines?

Link to comment
Share on other sites

4 minutes ago, Max Entropy said:

Hi @enej 

i don't think changing the hardware (clusters) fixes anything... the following questions come to mind:

  1. was the 'problem' encountered a software or a protocol problem?
  2. has Ripple made substantive changes to the code base to resolve the problem in the future regardless of hardware and traffic. Note, 'resolve' can include changes made to busy out the 'clients'
  3. are the 'client' software entities properly using the recommended APIs?

I am Generally interested in the nature of the problem and the purported fix.

Are GateHub machines in Amazon, Azure or on premise machines?

There was no software or protocol problem whatsoever.

Our wallet client has been connecting to public rippled instances (s1.ripple.com). There was simply not enough servers to handle all of the incoming websocket connections. With increased number of rippled's we can now handle more requests. Same goes for Data API endpoints.

 

Link to comment
Share on other sites

@enej ok understood... thanks...

Did your Gateway software 'busy-out' the clients... the errors reported seem to suggest they did not.

Link to comment
Share on other sites

Guys...

Is anyone familiar with the LastLedgerError... code?

i am generally interested in what is going on with a RippleD servers as it will inform me of the stability of the code.

It is true, that I have not looked at the code. If I have time this weekend, I will.

Link to comment
Share on other sites

LastLedger just means the time between you submitting the order and it making it to the ledger, there were too many other transactions in between.  There is a setting that lets you reject an order if too many other transactions happen in between (high volume trading).  It is there to protect against someone trading on old information (you didn't see a hundred other trades go by), but most of us I think don't really care.  The default is a low number, so I'm guessing that got tripped for a lot of people/clients that didn't set it once trading picked up.  For my own code, I set it pretty high as I don't particularly care how many ledgers go by as long as my order gets there.

Edited by pftq
Link to comment
Share on other sites

Still hangs when I log in. Would like an update on that. Wallet seems OK but everything hangs and I have to close the browser.

Seems solved. Yesterday it was a Chrome problem but today it was Safari. Chrome responsive and working. Fully updated Mac.

Sorry for the fuzz :-).

Edited by PogueMahone
Problem solved
Link to comment
Share on other sites

@pftq - thanks

i am a bit confused...

I will need to look at the code.

Enej's comments above suggests that the 'problem' is related session level TCP sockets timing out (not enough to sockets - above) when connecting to Ripple servers, while the error LastLedgerError... reference seems to be an application layer error.

Link to comment
Share on other sites

Yeah there were multiple problems going on, which makes it confusing to attribute.

For perspective on what I had to do for theworldexchange.net, I increased the maxLedgerOffset to about a hundred (instead of 3 default) to address the MaxLedger error.  Then I added code to detect disconnection (timeout, network unavailable) so that it automatically reconnects to Ripple if something happens, which was a different issue altogether.

Edited by pftq
Link to comment
Share on other sites

Well, this is consistent with my experience and bugs... problems are rarely attributed a single source but rather the unhappy co-incidence of several offending errors.

Edited by Guest
Link to comment
Share on other sites

1 hour ago, enej said:

Hi everyone,

I want to thank you for your support and patience in the past couple of days.

Here are a couple of updates on our end:

1) Last Ledger Exceeded error (tejMaxLedger) has been resolved. We deployed a brand new rippled cluster which will make transaction submission faster and more reliable.

2) Ripple Data API servers (used for price charts, stats and transaction history) were updated by Ripple team to help with high volume of requests. Ripple team is monitoring the servers and will add new ones if needed.

3) GateHub's back-end infrastructure has also been greatly improved yesterday. We deployed more servers and faster database.

It might take a couple of days for us to clear the support backlog but we are working round the clock to get back to you as soon as possible.

For urgent matters feel free to PM me directly.

Thank you,
Enej

 

I just had the fastest transfer of my entire crypto life.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...