This repository has been archived by the owner on Mar 28, 2023. It is now read-only.
ETH - releaseFunds does not change order to RESOLVED #2002
Assignees
Labels
bug
ethereum
Ethereum integration-related issues.
🔍 readyForReview
Issue or PR ready for code review prior to closing.
🚀 🚫
Launch blockers
Projects
Comments
jjeffryes
added
🚀 🚫
|
This continues to happen when trying it with another order disputed by the buyer, and in an order disputed by the seller. Each time ie: after releasing funds 2 times, there are 3, 3 times there are 4, etc. |
|
With some additional testing, whichever party tries to release the funds first is affected by this bug, regardless of which one started the dispute. |
|
@jjeffryes I haven't seen this bug for my disputed test scenarios (i'm on the same branch), with both parties seeing the order as This is how I'm running them:
I'm trying to figure out why we're seeing different behaviour... is the order state stuck in |
|
Of course the other big difference is that you're testing a tiny price with a large quantity, which I don't have a test for yet. |
Jeez that's not good, we better get to the bottom of this @amangale |
|
@hoffmabc mentioned he ran into this same issue over the weekend, but then it stopped happening. So far for me it happens every time. Maybe there's some external factor at work here? The invalid payment bug is also intermittent, though obviously it wasn't happening at the same time as this bug, since these are multisig orders. The two bugs together make me suspect something that can change even when what the client is doing is the same. |
|
@jjeffryes are you working with a fresh node or a v4 upgraded node? |
|
Completely fresh testnet nodes for all 3 nodes involved. |
|
Weird additional detail: I spun up a new seller node to see if maybe that would change anything. but…
Repeating the above after the seller had ETH in their wallet just went back to the bugged behavior where releaseFunds leaves the order in DECIDED. |
|
Update for that last clue, currently the seller, now that they have funds, can accept dispute decisions and releaseFunds works normally. The buyer, however, still shows the bugged behavior if they try to accept a decision. |
|
@drwasho here are the exact steps for this:
So far this bug has happened for every permutation of accepting a dispute payout... except when it doesn't. Why it stopped happening (sometimes) today is a mystery. |
|
A follow up: I just had another test where the seller tried to accept the order and got the insufficient funds error (despite having $0.07 in their wallet, probably due to #2004), in that case when I tried to have the buyer accept the payout they still had the bugged behavior. So it appears the one time it worked for the buyer was unrelated to the seller getting an insufficient funds error. |
|
If it's helpful, here's what was in my terminal the most recent time I got the When I clicked |
|
The multiple ORDER_PAYMENT messages in there seem weird. Should that be happening? After the https://rinkeby.etherscan.io/tx/0x0ced3dfcd16f0af3420af8042e6d529db2c158f382b7c598c3da14854fdc0005 https://rinkeby.etherscan.io/tx/0x3264f66aee8be285715daef23a89aec671773e27cf489b3de7f8e3e38b5da276 Both have a It seems like that might be useful information?
|
|
From the buyer's side, here's the JSON response from /order for the more recent purchase above. |
@jjeffryes what is the moderator's fee? Update: Moderator's fee is 0%, which could be causing the problem: "moderatorInfo": {
"description": "just a test account do not use",
"termsAndConditions": "no real orders will be moderated",
"languages": [
"en-US"
],
"acceptedCurrencies": [
"TETH",
"TBTC",
"TBCH",
"TZEC",
"TLTC"
],
"fee": {
"percentage": 0,
"feeType": "PERCENTAGE"
}
}, |
drwasho
added
the
🔍 readyForReview
cpacia
added a commit
that referenced
this issue
Mar 6, 2020
(#2002, 2005, 2006, 2012) Handle zero mod fee
|
fixed |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
In nth-master, 09c1138, my buyer node accepted the payout for a decided dispute paid for in TETH, but the order remained in the DECIDED state instead of updating to RESOLVED.
The seller, on the other hand, now sees the order as RESOLVED.
Attempting to releaseFunds again doesn't change anything.
After restarting the buyer's node, the order was still in the DECIDED state.
The order was disputed by the buyer, when it was awaiting fulfillment.
This bug does not occur when paying with TLTC.
Result of ob/order before releaseFunds for the buyer:
Result of ob/order after releaseFunds for the buyer:
Results of ob/order after releaseFunds for the seller:
The text was updated successfully, but these errors were encountered: