wallet: introduce setfeerate (an improved settxfee, in sat/vB) by jonatack · Pull Request #20391 · bitcoin/bitcoin

jonatack · 2020-11-14T22:23:20Z

Continuing the work on issue #19543, this proposes a setfeerate RPC in sat/vB and makes settxfee a hidden RPC per the plan described in #20484 (comment).

jonatack · 2020-11-14T22:24:00Z

Thanks to @xekyo for suggesting this in #20305 (comment).

DrahtBot · 2020-11-15T00:07:37Z

The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.

Conflicts

Reviewers, this pull request conflicts with the following ones:

test: Run mempool_limit.py even with wallet disabled #20874 (test: Run mempool_limit.py even with wallet disabled by stackman27)
wallet: check for non-representable CFeeRates #20546 (wallet: check for non-representable CFeeRates by jonatack)
util: Avoid invalid integer negation in FormatMoney and ValueFromAmount #20406 (util: Avoid invalid integer negation in FormatMoney and ValueFromAmount by practicalswift)
wallet: Migrate legacy wallets to descriptor wallets #19602 (wallet: Migrate legacy wallets to descriptor wallets by achow101)
rpc: have raw transaction decoding infer output descriptors #16795 (rpc: have raw transaction decoding infer output descriptors by instagibbs)

If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first.

ghost · 2020-11-15T03:28:42Z

In the 'Successful responses' and 'Error responses' above shouldn't it be sat/vB instead of sat/B?

jonatack · 2020-11-15T07:21:46Z

@prayank23 thanks for looking. Those will be automatically updated from sat/B to sat/vB by #20305, which updates the ToString() helper in CFeeRate. I updated the PR descpription to clarify that (thanks!)

src/wallet/rpcwallet.cpp

jonatack · 2020-11-27T17:36:15Z

Rebased and updated per the plan outlined in #20484 (comment).

maflcko

Concept ACK on adding the feature for users to set -paytxfee via rpc at runtime in sat/b

test/functional/rpc_getblockstats.py

src/wallet/rpcwallet.cpp

Make settxfee a hidden rpc to avoid confusion for new users and guide them to use the setfeerate rpc in sat/vB, while allowing existing scripts based on settxfee in BTC/kvB to continue using it without breaking. A hidden RPC means that `bitcoin-cli help settxfee` continues to display the help documentation while removing it from the results of the general `bitcoin-cli help`. Updates the release note.

per PR 20403 review feedback by MarcoFalke

S3RK · 2021-02-11T20:16:05Z

I start reviewing this one.

S3RK

I did a high level pass, didn't look at functional tests at all yet. Overall seems good, but I need to give it more thorough look. Please check my comments.

S3RK · 2021-02-11T20:51:59Z

src/policy/feerate.h

        static_assert(std::is_integral<I>::value, "CFeeRate should be used without floats");
    }
+    static CFeeRate FromSatB(CAmount fee_rate);
+    static CFeeRate FromBtcKb(CAmount fee_rate);


On line 47 I believe it should be (nFeePaid * 1e3 / 1e8) == (nFeePaid / 1e5)
and not (nFeePaid * 1e8 / 1e3) == (nFeePaid / 1e5)

S3RK · 2021-02-11T20:58:09Z

src/policy/feerate.cpp

+        switch (mode) {
+        case FeeEstimateMode::SAT_VB:
+            return strprintf("%d.%03d %s/vB", nSatoshisPerK / 1000, nSatoshisPerK % 1000, CURRENCY_ATOM);
+        default:


Do we want to assert some other FeeEstimateMode isn't passed by mistake?

S3RK · 2021-02-11T21:11:24Z

src/test/amount_tests.cpp

+    BOOST_CHECK_EQUAL(CFeeRate(3141).ToString(FeeEstimateMode::SAT_VB, /* with_units */ false), "3.141");
+    BOOST_CHECK_EQUAL(CFeeRate(10002).ToString(FeeEstimateMode::SAT_VB, /* with_units */ false), "10.002");
+    BOOST_CHECK_EQUAL(CFeeRate(3141).ToString(FeeEstimateMode::BTC_KVB, /* with_units */ false), "0.00003141");
+    BOOST_CHECK_EQUAL(CFeeRate(10002).ToString(FeeEstimateMode::BTC_KVB, /* with_units */ false), "0.00010002");


what happens if we pass an absurdly big number?

S3RK · 2021-02-12T08:23:35Z

src/test/rpc_tests.cpp

    BOOST_CHECK_THROW(AmountFromValue(ValueFromString("93e+9")), UniValue); //overflow error
 }

+BOOST_AUTO_TEST_CASE(rpc_parse_fee_rate_values)


Aren't CFeeRate various constructors and ToString() already covered by other unit tests? Unit tests have a non-zero maintenance cost and I'm missing the point of such duplication. The ValueFromFeeRate fn is pretty trivial and I believe the test should be trivial as well.

S3RK · 2021-02-12T08:31:20Z

src/policy/feerate.cpp

-    default:                      return strprintf("%d.%08d %s/kvB", nSatoshisPerK / COIN, nSatoshisPerK % COIN, CURRENCY_UNIT);
+    if (with_units) {
+        switch (mode) {
+        case FeeEstimateMode::SAT_VB:


I'm confused why do we put fee estimate mode and fee units, which are different things, into the same enum. I understand this is a bit out of scope for this PR, but I'd appreciate if you could provide some historical context for educational purposes.

IIRC that was done in #11413, but good point. I'll look at fixing that.

Ah, just remembered that I wrote a patch in November 2020 that does that but I have not opened it yet to not have too many PRs open.

S3RK · 2021-02-12T08:51:41Z

src/wallet/rpcwallet.cpp

+            LOCK(wallet.cs_wallet);
+
+            const CFeeRate amount{CFeeRate::FromSatB(AmountFromValue(request.params[0]))};
+            const CFeeRate relay_min_feerate{wallet.chain().relayMinFee().GetFeePerK()};


Do we need GetFeePerK() call here and below?

S3RK · 2021-02-12T09:18:23Z

src/test/amount_tests.cpp

+    BOOST_CHECK(CFeeRate::FromSatB(CAmount(100000)) == CFeeRate(1));
+    BOOST_CHECK(CFeeRate::FromSatB(CAmount(100001)) == CFeeRate(1));
+    BOOST_CHECK(CFeeRate::FromSatB(CAmount(2690000)) == CFeeRate(26));
+    BOOST_CHECK(CFeeRate::FromSatB(CAmount(123456789)) == CFeeRate(1234));


what happens if we pass an absurdly big number?

jonatack · 2021-02-12T16:04:32Z

Thank you @S3RK for your feedback, much appreciated. I'll reply soon.

S3RK

I finished the review. A bunch of small questions/comments, but nothing serious. Happy to discuss and do further reviews.

S3RK · 2021-02-17T08:48:34Z

test/functional/rpc_fundrawtransaction.py

@@ -717,7 +719,7 @@ def test_option_feerate(self):

        result = node.fundrawtransaction(rawtx)  # uses self.min_relay_tx_fee (set by settxfee)
        btc_kvb_to_sat_vb = 100000  # (1e5)


This is already defined earlier

S3RK · 2021-02-17T09:00:13Z

test/functional/wallet_basic.py

        fee_per_byte = Decimal('0.001') / 1000
-        self.nodes[2].settxfee(fee_per_byte * 1000)


Suggested change

fee_per_byte = Decimal('0.001') / 1000

self.nodes[2].settxfee(fee_per_byte * 1000)

fee_sat_per_byte = 100

self.nodes[2].setfeerate(fee_sat_per_byte)

S3RK · 2021-02-17T09:04:19Z

test/functional/rpc_fundrawtransaction.py

@@ -717,7 +719,7 @@ def test_option_feerate(self):

        result = node.fundrawtransaction(rawtx)  # uses self.min_relay_tx_fee (set by settxfee)


Suggested change

result = node.fundrawtransaction(rawtx) # uses self.min_relay_tx_fee (set by settxfee)

result = node.fundrawtransaction(rawtx) # uses self.min_relay_tx_fee (set by setfeerate)

S3RK · 2021-02-17T09:05:07Z

test/functional/rpc_fundrawtransaction.py

+        result = [self.nodes[3].fundrawtransaction(rawtx),  # uses self.min_relay_tx_fee (set by setfeerate)
            self.nodes[3].fundrawtransaction(rawtx, {"subtractFeeFromOutputs": []}),  # empty subtraction list
-            self.nodes[3].fundrawtransaction(rawtx, {"subtractFeeFromOutputs": [0]}),  # uses self.min_relay_tx_fee (set by settxfee)
+            self.nodes[3].fundrawtransaction(rawtx, {"subtractFeeFromOutputs": [0]}),  # uses self.min_relay_tx_fee (set by setfeerate)


There are couple more comments to fix below

S3RK · 2021-02-18T08:20:05Z

test/functional/wallet_bumpfee.py

+    test_response(requested=fee_rate, expected=fee_rate, msg="Fee rate for transactions with this wallet successfully set to 25.000 sat/vB")
+    bumped_tx = rbf_node.bumpfee(rbfid)
+    actual_feerate = bumped_tx["fee"] * COIN / rbf_node.getrawtransaction(bumped_tx["txid"], True)["vsize"]
+    assert_greater_than(Decimal("0.01"), abs(fee_rate - actual_feerate))


Why this specific value of 0.01? In the test case with settxfee we have a different value of 0.00001000. Should we have a testcase-wise constant?

Even 0.01 is too small for this. Signatures can be 71-73 bytes long, but even 72 bytes throws this off.

Suggested change

assert_greater_than(Decimal("0.01"), abs(fee_rate - actual_feerate))

bumped_txdetails = rbf_node.getrawtransaction(bumped_tx["txid"], True)

allow_for_bytes_offset = len(bumped_txdetails['vout']) * 2 # potentially up to 2 bytes per output

actual_fee = bumped_tx["fee"] * COIN

assert_approx(actual_fee, fee_rate * bumped_txdetails['vsize'], fee_rate * allow_for_bytes_offset)

S3RK · 2021-02-18T08:22:54Z

test/functional/wallet_bumpfee.py

+    def test_response(*, wallet="RBF wallet", requested=0, expected=0, error=None, msg):
+        assert_equal(rbf_node.setfeerate(requested), {"wallet_name": wallet, "fee_rate": expected, ("error" if error else "result"): msg})
+
+    # Test setfeerate with too high/low values returns expected errors


This test has nothing to do with bumpfee RPC. I believe it's better to have it in a separate test. wallet_bumpfee.py is already huge

S3RK · 2021-02-18T08:24:51Z

test/functional/wallet_bumpfee.py

+    assert_greater_than(Decimal("0.01"), abs(fee_rate - actual_feerate))
+    test_response(msg="Fee rate for transactions with this wallet successfully unset. By default, automatic fee selection will be used.")
+
+    # Test setfeerate with a different -maxtxfee


Same as above. This belongs to setfeerate test, not bumpfee test

S3RK · 2021-02-18T08:55:35Z

src/core_write.cpp

            strprintf("%s%d.%08d", sign ? "-" : "", quotient, remainder));
 }

+UniValue ValueFromFeeRate(const CFeeRate& fee_rate, FeeEstimateMode mode)


While I think it's a great idea to have this function, I'm a bit concerned that it'll be not used and forgotten. Do you have plans to do a follow up and put it to use by replacing ValuefromAmount in appropriate places ?

S3RK · 2021-02-18T09:10:49Z

src/wallet/rpcwallet.cpp

            throw JSONRPCError(RPC_INVALID_PARAMETER, "Cannot specify both estimate_mode and fee_rate");
        }
-        cc.m_feerate = CFeeRate(AmountFromValue(fee_rate), COIN);
+        cc.m_feerate = CFeeRate::FromSatB(AmountFromValue(fee_rate));


I love the idea of named constructors. Should we strive for consistency and use them in more places?

S3RK · 2021-02-18T09:12:13Z

test/functional/wallet_create_tx.py

+        def test_response(*, requested=0, expected=0, error=None, msg):
+            assert_equal(node.setfeerate(requested), {"wallet_name": "test_wallet", "fee_rate": expected, ("error" if error else "result"): msg})
+
+        # Test setfeerate with 10.0001 (CFeeRate rounding), "10.001" and "4" sat/vB


Again I believe those check would be better places in a separate setfeerate test

DrahtBot · 2021-03-03T19:25:01Z

🐙 This pull request conflicts with the target branch and needs rebase.

_{Want to unsubscribe from rebase notifications on this pull request? Just convert this pull request to a "draft".}

ryanofsky · 2021-07-13T17:59:11Z

src/wallet/rpcwallet.cpp

            throw JSONRPCError(RPC_INVALID_PARAMETER, "Cannot specify both estimate_mode and fee_rate");
        }
-        cc.m_feerate = CFeeRate(AmountFromValue(fee_rate), COIN);
+        cc.m_feerate = CFeeRate::FromSatB(AmountFromValue(fee_rate));


In commit "wallet: update rpcwallet to CFeeRate named constructors" (c72083c)

Wow wow. I see that this is doing the right thing, but these constructor names seem absurdly misleading! The AmountFromValue function converts a floating point number to a fixed-point representation of that number multiplied by 10e8. So even though the fee_rate univalue argument has units of sat/B, the AmountFromValue integer return value is no longer in the same units, because it has been multiplied by 100000000. So FromSatB should really be called CFeeRate::FromSatB_Fixed_Point_e8 or something like that because it is taking sat/B values with a very specific, atypical number representation.

An analogous thing is happening with FromBtcKb below. It doesn't really take a BTC/Kb argument. That's basically false advertising. What it really takes is a BTC/Kb*10e8 argument which is the same as a sat/Kb argument, so this constructor would be more accurately called CFeeRate::FromSatKb

I guess I'm having a negative reaction to these constructor names, but I do think this could be cleaned up pretty easily. I think it would be better if CFeeRate class had no involvement with univalue parsing or fixed point representations, and if it didn't (mis)use the CAmount type as a way to represent rates when CAmount is used almost everywhere else only to represent quantities.

My suggestion to be less confusing / misleading would be to drop the two new CFeeRate constructors, and just add standalone RPC util functions:

CFeeRate FeeRateFromSatB(const UniValue& value); CFeeRate FeeRateFromBtcKb(const UniValue& value);

Then call these to simplify CFeeRate::FromSatB(AmountFromValue(feerate)) as FeeRateFromSatB(feerate) here and in setfeerate and to simplify CFeeRate::FromBtcKb(AmountFromValue(feerate)) as FeeRateFromBtcKb(feerate) in settxfee

Wow wow. I see that this is doing the right thing, but these constructor names seem absurdly misleading!

The current code in master isn't exactly clear right now either:

/** Constructor for a fee rate in satoshis per kvB (sat/kvB). * * Passing a num_bytes value of COIN (1e8) returns a fee rate in satoshis per vB (sat/vB), * e.g. (nFeePaid * 1e8 / 1e3) == (nFeePaid / 1e5), * where 1e5 is the ratio to convert from BTC/kvB to sat/vB. */ CFeeRate(const CAmount& nFeePaid, uint32_t num_bytes);

I think it would be good to figure out how to clear this up before introducing further features that depend on this.

Original discussion for reference: #20305 (comment)

Wow wow. I see that this is doing the right thing, but these constructor names seem absurdly misleading!

The current code in master isn't exactly clear right now either:

/** Constructor for a fee rate in satoshis per kvB (sat/kvB). * * Passing a num_bytes value of COIN (1e8) returns a fee rate in satoshis per vB (sat/vB), * e.g. (nFeePaid * 1e8 / 1e3) == (nFeePaid / 1e5), * where 1e5 is the ratio to convert from BTC/kvB to sat/vB. */ CFeeRate(const CAmount& nFeePaid, uint32_t num_bytes);

I think it would be good to figure out how to clear this up before introducing further features that depend on this.

Ok, seems best to close this then. It's almost a year old and needs to be done differently. I think using sat/vB is still a popular request from users and plan to try again a bit later.

ryanofsky

Started review (will update list below with progress). I got a little sidetracked on the cfeerate constructor names, but overall this seems like a very nice, well thought out PR

jonatack · 2021-07-13T18:08:14Z

Thanks! Huge apologies, since the merge of #21786 that resolved some of these things differently (at a lower level), the approach of the initial refactoring commits is obsolete and being changed. I need to push an update and have been intending to do so for some time. 🙏

Github-Pull: bitcoin#20391 Rebased-From: 8e863e3 (diff-minimised)

Github-Pull: bitcoin#20391 Rebased-From: 529bfc1

Github-Pull: bitcoin#20391 Rebased-From: c907f15

Was: test: update functional tests from settxfee to setfeerate Github-Pull: bitcoin#20391 Rebased-From: d87f0f3 (partial)

Github-Pull: bitcoin#20391 Rebased-From: 8e863e3 (diff-minimised)

Github-Pull: bitcoin#20391 Rebased-From: 529bfc1

Github-Pull: bitcoin#20391 Rebased-From: c907f15

Was: test: update functional tests from settxfee to setfeerate Github-Pull: bitcoin#20391 Rebased-From: d87f0f3 (partial)

jonatack mentioned this pull request Nov 14, 2020

wallet: introduce fee_rate sat/vB param/option #20305

Merged

DrahtBot added RPC/REST/ZMQ Wallet labels Nov 14, 2020

DrahtBot mentioned this pull request Nov 15, 2020

test: Implicitly sync after generate* to preempt races and intermittent test failures #20362

Closed

DrahtBot mentioned this pull request Nov 15, 2020

Descriptor Wallets BIP39 recovery #20361

Closed

jonatack force-pushed the setfeerate branch 4 times, most recently from fa492af to a96417d Compare November 15, 2020 17:56

jonatack mentioned this pull request Nov 16, 2020

Normalize fee units for RPC ("BTC/kB" and "sat/B) #19543

Closed

DrahtBot added the Needs rebase label Nov 17, 2020

luke-jr reviewed Nov 25, 2020

View reviewed changes

src/wallet/rpcwallet.cpp Outdated Show resolved Hide resolved

jonatack mentioned this pull request Nov 26, 2020

RPC: Migrate estimatesmartfee return value from "feerate" (BTC/kvB) to "fee_rate" (sat/vB) #20484

Closed

jonatack force-pushed the setfeerate branch from a96417d to e915a2a Compare November 27, 2020 17:25

jonatack mentioned this pull request Nov 27, 2020

wallet: deprecate feeRate in fundrawtransaction/walletcreatefundedpsbt #20483

Closed

DrahtBot removed the Needs rebase label Nov 27, 2020

This was referenced Nov 28, 2020

rpc: deprecate addresses and reqSigs from rpc outputs #20286

Merged

rpc: Remove duplicate name and argNames from CRPCCommand #20012

Merged

wallet: Migrate legacy wallets to descriptor wallets #19602

Merged

maflcko approved these changes Nov 28, 2020

View reviewed changes

test/functional/rpc_getblockstats.py Outdated Show resolved Hide resolved

src/wallet/rpcwallet.cpp Outdated Show resolved Hide resolved

DrahtBot mentioned this pull request Nov 28, 2020

rpc: have raw transaction decoding infer output descriptors #16795

Merged

jonatack force-pushed the setfeerate branch from e915a2a to d953407 Compare November 28, 2020 19:00

DrahtBot mentioned this pull request Nov 29, 2020

Fix -Wtautological-constant-out-of-range-compare warnings on 32-bit systems #19735

Closed

jonatack force-pushed the setfeerate branch 2 times, most recently from 14c5e50 to 2176a3a Compare November 30, 2020 10:21

jonatack added 2 commits January 28, 2021 21:24

wallet, test: upgradewallet follow-ups

1002e2d

per PR 20403 review feedback by MarcoFalke

jonatack force-pushed the setfeerate branch from 4e7aa73 to 1002e2d Compare January 28, 2021 21:21

DrahtBot removed the Needs rebase label Jan 28, 2021

S3RK reviewed Feb 12, 2021

View reviewed changes

S3RK reviewed Feb 18, 2021

View reviewed changes

DrahtBot added the Needs rebase label Mar 3, 2021

jonatack mentioned this pull request Mar 24, 2021

Add feerate histogram to getmempoolinfo #21422

Closed

jonatack mentioned this pull request Apr 21, 2021

since v21 settxfee not working correctly #21737

Closed

jonatack mentioned this pull request Apr 30, 2021

doc: Clarify that feerates are per virtual size #21752

Merged

ryanofsky reviewed Jul 13, 2021

View reviewed changes

jonatack mentioned this pull request Aug 10, 2021

rpc: Add function walletestimatefee #22673

Closed

maflcko mentioned this pull request Sep 29, 2021

Cleanup CFeeRate constructor (sat/vB vs BTC/kvB) #23129

Closed

jonatack closed this Sep 29, 2021

luke-jr pushed a commit to bitcoinknots/bitcoin that referenced this pull request Oct 10, 2021

wallet: introduce setfeerate, an improved settxfee in sat/vB

a8ae170

Github-Pull: bitcoin#20391 Rebased-From: 8e863e3 (diff-minimised)

luke-jr pushed a commit to bitcoinknots/bitcoin that referenced this pull request Oct 10, 2021

test: add setfeerate functional coverage in wallet_create_tx.py

d94518e

Github-Pull: bitcoin#20391 Rebased-From: 529bfc1

luke-jr pushed a commit to bitcoinknots/bitcoin that referenced this pull request Oct 10, 2021

test: add setfeerate functional coverage in wallet_bumpfee.py

031ebf4

Github-Pull: bitcoin#20391 Rebased-From: c907f15

luke-jr pushed a commit to bitcoinknots/bitcoin that referenced this pull request Oct 10, 2021

test: add more functional tests for setfeerate

1e6469d

Was: test: update functional tests from settxfee to setfeerate Github-Pull: bitcoin#20391 Rebased-From: d87f0f3 (partial)

luke-jr pushed a commit to bitcoinknots/bitcoin that referenced this pull request Dec 14, 2021

test: add more functional tests for setfeerate

a77000f

Was: test: update functional tests from settxfee to setfeerate Github-Pull: bitcoin#20391 Rebased-From: d87f0f3 (partial)

luke-jr pushed a commit to luke-jr/bitcoin that referenced this pull request May 21, 2022

wallet: introduce setfeerate, an improved settxfee in sat/vB

f807d13

Github-Pull: bitcoin#20391 Rebased-From: 8e863e3 (diff-minimised)

luke-jr pushed a commit to luke-jr/bitcoin that referenced this pull request May 21, 2022

test: add setfeerate functional coverage in wallet_create_tx.py

f19c78c

Github-Pull: bitcoin#20391 Rebased-From: 529bfc1

luke-jr pushed a commit to luke-jr/bitcoin that referenced this pull request May 21, 2022

test: add setfeerate functional coverage in wallet_bumpfee.py

a2293d9

Github-Pull: bitcoin#20391 Rebased-From: c907f15

luke-jr pushed a commit to luke-jr/bitcoin that referenced this pull request May 21, 2022

test: add more functional tests for setfeerate

fc4f061

Was: test: update functional tests from settxfee to setfeerate Github-Pull: bitcoin#20391 Rebased-From: d87f0f3 (partial)

bitcoin locked and limited conversation to collaborators Oct 30, 2022

		@@ -717,7 +719,7 @@ def test_option_feerate(self):

		result = node.fundrawtransaction(rawtx) # uses self.min_relay_tx_fee (set by settxfee)
		btc_kvb_to_sat_vb = 100000 # (1e5)

		fee_per_byte = Decimal('0.001') / 1000
		self.nodes[2].settxfee(fee_per_byte * 1000)

-    assert_greater_than(Decimal("0.01"), abs(fee_rate - actual_feerate))
+    bumped_txdetails = rbf_node.getrawtransaction(bumped_tx["txid"], True)
+    allow_for_bytes_offset = len(bumped_txdetails['vout']) * 2  # potentially up to 2 bytes per output
+    actual_fee = bumped_tx["fee"] * COIN
+    assert_approx(actual_fee, fee_rate * bumped_txdetails['vsize'], fee_rate * allow_for_bytes_offset)

Conversation

jonatack commented Nov 14, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jonatack commented Nov 14, 2020

Uh oh!

DrahtBot commented Nov 15, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Conflicts

Uh oh!

ghost commented Nov 15, 2020

Uh oh!

jonatack commented Nov 15, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

jonatack commented Nov 27, 2020

Uh oh!

maflcko left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

S3RK commented Feb 11, 2021

Uh oh!

S3RK left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jonatack commented Feb 12, 2021

Uh oh!

S3RK left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

DrahtBot commented Mar 3, 2021

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

jonatack commented Nov 14, 2020 •

edited

Loading

DrahtBot commented Nov 15, 2020 •

edited

Loading

jonatack commented Nov 15, 2020 •

edited

Loading