3bot and eVDC provisioning unreliable

Dear Community,

We want to give you an update on an issue we are having on the TFGrid 2.x.

For months now we experience issues with the Stellar payment network, and this sometimes results in failures when users want to deploy a VDC (virtual datacenter) or 3Bot (our SDK system). If it happens you can retry and often it will work but its not a nice situation.

Issues we encounter and solutions we tried:

  • we have run against rate limiting issues at first, we have worked around those
  • we have lots of timeouts on talking to the stellar blockchain, we have changed code over last months to deal with this, actually too much code, its better but still not ok
  • we tried 3 times to run a full blockchain node for stellar on our grid but didn’t succeed yet, timeouts on syncing and the machine needs to be really really big.

We managed to make some progress on it but its not good enough.

It looks like Stellar is getting overloaded on their network and there is only so much we can do.

One of the reasons we are pushing ThreeFold grid 3.0 is to also deal with this issue.
In TFGrid 3.0 everything runs on our own Substrate blockchain called TFChain.

Yesterday we have ordered new fast SSD’s and have reserved a big server to try and get a full Stellar blockchain node (4e attempt), this potentially will resolve the issue on 2.x as well. We will keep you posted.

We apologise for the inconveniences this causes on TFGrid 2.x.

thanks for your understanding

More info about TFGrid 3.0: Announcement of TFGrid 3.0

4 Likes