Platform v0.24 with Dashmate on Testnet : Issues

[qwizzie]

@pshenmic
@strophy

Source : https://docs.dash.org/en/stable/docs/user/masternodes/setup-testnet.html

qwizzie@HPMN-Testnet:~$ dashmate stop
qwizzie@HPMN-Testnet:~$ npm update -g dashmate
qwizzie@HPMN-Testnet:~$ dashmate reset --platform-only
› Error: Unexpected argument: --platform-only
› See more help with --help

qwizzie@HPMN-Testnet:~$ dashmate --help
Distribution package for Dash node installation

VERSION
dashmate/0.24.4 linux-x64 node-v16.19.1

Why can't i do a Platform-only reset through latest stable Dashmate v0.24.4 iaw instructions here --> https://www.dash.org/blog/dash-core-group-release-announcement-dash-platform-v0-24-on-testnet/
Dashmate does not seem to regnonice the --platform-only argument of a dashmate reset command.
Same goes for command dashmate reset --platform-only --hard.

qwizzie@HPMN-Testnet:~$ dashmate reset --platform-only --hard
› Error: Unexpected argument: --platform-only
› See more help with --help

I am aware there are also versions of Dashmate on v25, but Testnet itself has just been updated to Platform v0.24
and latest stable version of Dashmate on v0.24 (0.24.4) should allow users to do a Platform-only reset command, correct ?

Also after receiving above error, i can no longer issue a dashmate update, dashmate start or a dashmate setup command :

qwizzie@HPMN-Testnet:~$ dashmate update
TypeError: Cannot set properties of undefined (setting 'name')

qwizzie@HPMN-Testnet:~$ dashmate start
TypeError: Cannot set properties of undefined (setting 'name')

qwizzie@HPMN-Testnet:~$ dashmate setup
TypeError: Cannot set properties of undefined (setting 'name')

I got past this by deleting my config.json file, doing a dashmate reset --hard, followed by a new dashmate setup command.
Unfortunetely using dashmate start (after a succesfull dashmate setup) then triggers another error :

dashmate start
❯ Start testnet node
✔ Check node is not started
✖ Docker Compose error: Network dashmate_testnet_default Creating
Network dashmate_testnet_default Error
failed to create network dashmate_testnet_default: Error response from daemon: Pool overlaps with other one on this address space
qwizzie@HPMN-Testnet:~$

I think this is related to my previous dashmate reset --hard command that should have wiped out the following three folders, but did not actually delete them :

Or maybe it is related to something else. How do i get past this Docker Compose error and start dashmate properly ?

Update : I decided to go for the extreme solution and re-installed my testing server (also upgraded from Ubuntu 18.04 to 20.04).
That solved the Docker Compose error, as i needed to install everything from scratch.

PoSe Penalty n/a : incorrect (should show score 0, hpmn's can still get PoSe scored)
Payment queue position : incorrect
Next payment time : incorrect

By the way, there seems to be a difference between Testnet Setup instructions on Docs (see first pic) and the Github instructions of Dashmate
when doing a dashmate reset command.

Source : https://github.com/dashpay/platform/tree/master/packages/dashmate

Should we be using dashmate reset --platform-only or should we be using dashmate reset --platform-only --hard ?
And should we include a dashmate setup command or not ?

 

[qwizzie]

So two Dashmate v0.24.4 issues :

dashmate reset --platform-only
dashmate reset --platform-only --hard
› Error: Unexpected argument: --platform-only

I could not really solve this.

dashmate start
❯ Start testnet node
✔ Check node is not started
✖ Docker Compose error: Network dashmate_testnet_default Creating
Network dashmate_testnet_default Error
failed to create network dashmate_testnet_default: Error response from daemon: Pool overlaps with other one on this address space

Solved by reinstalling my server and installing everything from scratch.

Also it would be nice for Dash Mainnet if we could just copy or move our current (1K) Dash blockchain directly into the (4K) masternode_core_data Docker volume, instead of having to download the whole Dash blockchain through Dashmate, as in most cases Dash users already have the full Dash blockchain on their server (just not in Docker format). I assume copying the Dash blockchain from within the server into a Docker volume is much faster, then dowloading the whole Dash blockchain from scratch into a Docker volume.

 

[pshenmic]

Sup there

› Error: Unexpected argument: --platform-only

I think this command was renamed recently to just `--platform`, could you check that via --help potentially?

TypeError: Cannot set properties of undefined (setting 'name')

This is very interesting, are you able to reproduce this error and could you try to get more info with --verbose flag? I followed your steps but couldn't reproduce

Pool overlaps with other one on this address space

That happened because you wiped your dashmate configuration, while there were already something created by the dashmate in the docker. You should not do it. To resolve that, you can invoke docker system prune, or manage to remove all undeleted resources that are left from previous run manually.
You can try hard reset as well, must fix this too as I remember.

Also it would be nice for Dash Mainnet if we could just copy or move our current (1K) Dash blockchain directly into the (4K) masternode_core_data Docker volume, instead of having to download the whole Dash blockchain through Dashmate, as in most cases Dash users already have the full Dash blockchain on their server (just not in Docker format). I assume copying the Dash blockchain from within the server into a Docker volume is much faster, then dowloading the whole Dash blockchain from scratch into a Docker volume.

I don't think its a good idea to copy chain data inside the containers honestly, there are many things that can mess it up, starting from permissions of the files to the mismatched versions of the software. The sync is being done very fast on both mainnet and testnet, and you can always presync ahead, so I don't think its worth to bother

 

[qwizzie]

Sup there


(1) I think this command was renamed recently to just `--platform`, could you check that via --help potentially?

(2) This is very interesting, are you able to reproduce this error and could you try to get more info with --verbose flag? I followed your steps but couldn't reproduce

(3) That happened because you wiped your dashmate configuration, while there were already something created by the dashmate in the docker. You should not do it. To resolve that, you can invoke docker system prune, or manage to remove all undeleted resources that are left from previous run manually.
You can try hard reset as well, must fix this too as I remember.

(4) I don't think its a good idea to copy chain data inside the containers honestly, there are many things that can mess it up, starting from permissions of the files to the mismatched versions of the software. The sync is being done very fast on both mainnet and testnet, and you can always presync ahead, so I don't think its worth to bother

1 : I currently have my evo node running again on Testnet, but i will keep this in mind if i ever need to do a dashmate reset of platform.
If it is indeed --platform then someone should update the instructions (both on Github and in Docs).

2 : I can't at the moment reproduce anything anymore, as i reinstalled my server to get everything working as intended.

3 : At some point i did try to do a manual reset i.a.w. following code :

docker stop $(docker ps -q)
docker system prune
docker volume prune
rm -rf ~/.dashmate/

Source : https://github.com/dashpay/platform/tree/master/packages/dashmate

But i got stuck on the very first command, it needed an additional argument (something about specifying volume if i remember correctly). So i could not stop Docker and could not proceed.

4 : Valid points.

 

[pshenmic]

But i got stuck on the very first command, it needed an additional argument (something about specifying volume if i remember correctly). So i could not stop Docker and could not proceed.

Its okay if it errors about missing arguments, that means that you don't have any containers running (because the output of docker ps -q will be empty)

You can proceed with system and volume prune at this point

 

[andrewkevin210]

great post