theta-testnet-001 relayers

[urban stag]

.

[amber kite]

We are running

[paper pulsar]

good

i'm running it

[molten drum]

can add to the update script any running channel or create new, whatever is needed)

[pallid wolf]

I can

[molten drum]

haha, almost everything expired after fork

[minor solstice]

I had a channel running in theory, but doing system work on indexer atm, so not sure how stable it is

[molten drum]

you just create it, it is leaving without you but can die without you 🙂 need txs or updates

[jade dust]

can add any channel to hermes and make sure it wont be expired. just tell what

set up update txs every x hours

[vivid sable]

we also can add/open channel and back up it

[minor solstice]

actually, I could set up a channel too. I'd need some naan and atom testnet tokens tho

[molten drum]

1. someone opens to don't spam with a lot
2. others configure update scripts
   the only question is in naan for updating and test atom))

[turbid remnant]

how much naan and atom is required to run for 10 days properly

i have decent atom,

[jade dust]

depends

[lament barn]

Actually working on it as we speak

Having an issue getting test atom though, seems people are spamming the faucet.

[turbid remnant]

how much do you need?

[urban stag]

amazing! thank you!

question: assuming we support a specific channel and the relayers have lots of ATOM and NAAN, what could go wrong if hundreds or thousands of participants are trying to use the channel for transfers?

we have testnet ATOM for theta-testnet-001

[jade dust]

out of balance for operators

[lament barn]

10 to 50 would be just fine

[amber kite]

Channel can be full. May be good idea to have several channels

[molten drum]

stuck txs that must be cleared from time to time I suppose

Query hermes for unreceived packets and acknowledgements with the following:

hermes query packet pending --chain <A> --port transfer --channel channel-A
hermes query packet pending --chain <B> --port transfer --channel channel-B

Clear the channel with the following:

hermes clear packets --chain <A> --port transfer --channel channel-A
hermes clear packets --chain <B> --port transfer --channel channel-B

Update clients with the following:

hermes update client --host-chain <A> --client 07-tendermint-A
hermes update client --host-chain <B> --client 07-tendermint-B

[paper pulsar]

the nam i running out quickly hahaaha

[turbid remnant]

i will try to send, what i can

[lament barn]

Gimme a sec, i'll get you my address

[urban stag]

isn't each channel considered a different asset by the recipient chain?

[jade dust]

never

[jade dust]

you are right, this will make lot of questions

[lament barn]

cosmos1h9krsew6kpg9huzcqgmgmns0n48jx9ydfvz354

[urban stag]

how much NAAN does it cost per transaction, roughly? like roughly how much NAAN would a relayer need to handle all the txs

[amber kite]

No. As i know, channel support all assets from one chain to another

[eternal osprey]

I'm on board, and will provide the channel soon if still needed tomorrow.

[jade dust]

depends on gas. main costs come from 2.5 naan fee

[paper pulsar]

based on my experiences maybe 1 or 2 , cmiiw

[urban stag]

if all relayers ran out of NAAN, would the channel expire?

[jade dust]

yes

[amber kite]

Only if channel have no transactions during some time

[jade dust]

in case no any activity at all then yes

and there wont be any updates if out of balance

[molten drum]

and it will not coz update will not be done as no naan/atom 😄

[paper pulsar]

yes

[paper pulsar]

because the fees of ibc-client update is really expensive tbh hahaha

[urban stag]

on the Namada side they are?

[paper pulsar]

yeah

[jade dust]

they are not expensive. expensive tx on namada

[urban stag]

how are the ibc client update fees determined / set?

[vivid sable]

yes

[urban stag]

ah!

[paper pulsar]

not on another chain. only on namada

[jade dust]

they will be 2.501 NAAN

[urban stag]

so Namada tx fees are high, making client update fees expensive

[molten drum]

you can limit max gas in config, but risk that then tx will be failed if gas will be high all the time

[urban stag]

how are Namada gas fees set? validator operators? in their client

[amber kite]

Chain activity and validators settings

[paper pulsar]

i've already try that method and the result is the same as what you said

[jade dust]

wrong

[paper pulsar]

it's not corelated with the validator settings i think

[urban stag]

in Cosmos chains i recall that validators set a minimum gas fee in their config file

[amber kite]

That's right

[molten drum]

++, but that is minimum

and max depends on the activity in the chain, afaik

[urban stag]

oh well, probably not worth pursuing lower tx fees rn
will just need to calculate how much relayers will each need in NAAN, assuming.. we happy assuming 2.5 NAAN per tx?

[turbid remnant]

a governances proposal for running a relayer for x naan would be cool!

[jade dust]

just add extra dust like 0.01 naan

[turbid remnant]

network down at the moment , query rpc or some error not sure

[urban stag]

wdym?

looks like there was an upgrade to theta-test-001 chain

[jade dust]

tx wont be really expensive. main issue is 2.5NAAN per tx. All other stuff like chain activity and so on wont cost much

[paper pulsar]

that's the real problem i think

[urban stag]

to clarify, is the issue that any tx is ~2.5 NAAN? or specifically it will cost the relayer 2.5 NAAN per IBC tx?

[minor solstice]

thinking it might be more

[cunning finch]

Not necessarily, non operators can also run update command in order to update ibc clients on both sides

[minor solstice]

won't relayer wallets still pay for relay?

[urban stag]

for which? relaying IBC txs vs any Namada tx

[lament barn]

https://rpc.sentry-01.theta-testnet.polypore.xyz/ this one works for me

[cunning finch]

Yep, but if others help in order to update clients, relayer operators won't spend too much

[minor solstice]

ibc. thought my relayer wallets were drained at higher rates

[jade dust]

any tx 2.5 naan

[minor solstice]

sure

[lament barn]

Ouch

[jade dust]

i am sure this won't be coordinated ever)

[cunning finch]

We have decided the canonical channels already?

[urban stag]

not yet

[cunning finch]

For theta-testnet-001

[lament barn]

And just like that, NAAN became valuable hah

[minor solstice]

I'm sure there are people with some hundreds of k naan

[cunning finch]

If anyone already has one, I will kindly help funding the relayer address as well as updating both clients regularly

[lament barn]

I have a ton of NAAN, thats why I spoke up but still working on getting my relayer working. Once I'm confident I'd be willing to do this @urban stag

[median island]

😁

Well have no problem to fund relayer

[paper pulsar]

hermes created by informal systems, right?

[turbid remnant]

check dm

[minor solstice]

going offline for an hour or two, ping me if y'all need anything from me on this

[lament barn]

Yep 🙂 but not by me ha

[urban stag]

can you comment in your addresses? https://docs.google.com/spreadsheets/d/13tnM-o8Cmcpmu7Q9_7xGRZHmGWCUxT4i2UIdlpF3KII/edit?usp=sharing

[lament barn]

I just need some test atoms and i'll be up

[molten drum]

comment only, no edit perm, sent req

[urban stag]

can't you suggest edit? and then i can accept and it will be permanent

[molten drum]

ah, as a comment, I guess yes

[median island]

Add me too

[pallid wolf]

add Mike

[paper pulsar]

already request, "i'm baconvalley"

[urban stag]

okay i've dropped some usernames into the sheet
we should probably limit it to 8 relayers, because we'll need to keep all of these funded

[molten drum]

do you want multiple channels?

[cunning finch]

Multiple operators in one canonical channel

[vivid sable]

Sent edit request

[crude timber]

Seem like 8 is enough, so I won't request

[molten drum]

like it

[urban stag]

when you get the chance, pls comment in:

1. tnam address
2. theta-testnet-001 address

[jade dust]

left comment

[urban stag]

added you

[jade dust]

any chance to add me?

[minor solstice]

don't have google-access to edit. if someone can fill in this I'd appreciate: (assuming it's the relayer addresses)
cosmos185h858752697h2ut6q04eascjy5vknr3emt8pj
tnam1qz7zlsyt780rzdga5cdda7epddu4g7thxv5vmt3j
(back in a lil bit)

[urban stag]

i'll add you

[frosty valve]

I also can if needed

[urban stag]

but that's it, cut-off is 9

[brazen bear]

If there are still openings, please add me

[urban stag]

thanks SO much for volunteering!

[paper pulsar]

just don't forget us in the future 🙂

[urban stag]

there's a lot to remember 😅
but i would like to try a more focused incentivized testnet, similar to #🏠🔥-housefire-canary-net

really this is about the chance to work together a bit to advance this shielded expedition to its finality, haha

[torpid gull]

We have been capturing all open clients. We are happy to keep running.

[urban stag]

not incentivizing relaying here, just to be clear

[median island]

Being in top 100 is enough 😁

[molten drum]

you'll be in top 100 relayers 😄

[solar fog]

Just made a new one after hardfork

`Namada

            "channel-1137",

  Theta 
        
            "channel-4117",

`

[jade dust]

correct trusting period?

[paper pulsar]

what number do you think is correct?

[solar fog]

I use script to update tendermint user every 5 minutes

[jade dust]

no more questions

each chain has its own tursting period. for example if in axelar you create it with wrong , your channel won't be accepted
creating channel is not only run hermes with guide from chat

[inland nebula]

how can I check if my relayer is healthy?

[paper pulsar]

no no, i've created my own ibc channel, i only asked

[jade dust]

anyway this is testnet so we are free to do everything)

[paper pulsar]

shielded-expedition.88f17d1d14 client_id: ClientId : ClientId("07-tendermint-3540") , connection_id: ConnectionId("connection-1738")
theta-testnet-001 client_id: ClientId : ClientId("07-tendermint-3516") , connection_id: ConnectionId("connection-3622")```

[molten drum]

it should not be more than unbonding period and you can't start relayers if the config is wrong, actually, even config test is enough

[jade dust]

i know it. fact that you can relay doesnt mean your channel is correct for counterparty

[gusty lintel]

yes we were running it before fork, we can enable it again

[molten drum]

I am still wondering about the trusting period in Namada, using now smth like a bit more than 1 hour, but didn't try to find a border value

[molten drum]

I doubt, if it is dead, only through gov

[urban stag]

does it cost NAAN to relay an asset into Namada, out of Namada, or both?
if we put like 2500 NAAN on each relayer, that's like 1000 txs per relayer

[molten drum]

from namada to external - yes, apmk

[median island]

Yes

[paper pulsar]

maybe, i think so, have thoughts guys?

[inland nebula]

how to fix this?client state is not valid: latest height is outside of trusting period!

[molten drum]

and update costs as well

[jade dust]

that is like 1k tx, but this 1k include update clients and so on
I would suggest relaying in other way

You need just 2-3 operators for just updating channel

[molten drum]

dead 😄 restore through gov (almost impossible in testnets) or re-create new

[jade dust]

without relaying

anyway the easiest way to track balance is just to run it ad see

[molten drum]

we should not overspend on updates, no need to have 8 updaters for 8 relayers, even one is enough, but if it is stable

[urban stag]

ohh interesting
how do you limit a relayer to just updating vs relaying?

[jade dust]

easy

you are not running relayer. You running just one bash script in cron)

[inland nebula]

what is the best value here?trusting_period = '4752s'

[molten drum]

#1222959128629215262 message 🤓

[median island]

You are in wrong place

[molten drum]

I recall in one of the projects there was a discord bot with notifications about clients close to expiration

[urban stag]

how often do updates happen?
how often must they happen to prevent the channel from expiring?

[jade dust]

/root/.hermes/bin/hermes update client --host-chain namada --client 00--tendermin--

[jade dust]

in case of current testnet 2 times a day more then enough

[molten drum]

#1222959128629215262 message ideally)

[urban stag]

must relayers always be updating? or can we just relegate that to the bash script

[jade dust]

missed it

[paper pulsar]

solutive

[jade dust]

they should be manually updated only if there is no any activity for a day for example

[molten drum]

I know the project, but not sure they can share, need to ask )

[paper pulsar]

quasar use it too

i saw it in their official server

[urban stag]

does an update happen when an IBC tx is relayed?

[molten drum]

didn't see in discord, maybe closed

[jade dust]

yes

[urban stag]

this is the ~2.5 NAAN, yeah? or is it 2.5 NAAN for the update and 2.5 NAAN for the relay

[jade dust]

2.5 total

[cunning finch]

Guys, take into account that in Namada we should use pipeline_length as a reference to set the trusting period instead of the unbonding_length

[paper pulsar]

check their "ibc-monitor" channel

[cunning finch]

2/3 of pipeline_length

[gusty lintel]

just when you relay out of namada you use naan

[urban stag]

is the NAAN cost for every IBC tx, or just relaying transactions away from Namada?

[jade dust]

2 ways

[urban stag]

ah! thanks (great timing, haha)

[urban stag]

because the update happens regardless, and that's the cost?

[jade dust]

no, when assets come from other chain, they should be relyed at distant chain too.

if i sent tx from namada to cosmos there will be tx at namada, and at cosmos) aand someone should pay for them)

[gusty lintel]

sry, i might be wrong, i think it used both ways

[cunning finch]

Relayer would pay NAAN for recv tx (receiving in) and ack txs (sending out)

[gusty lintel]

yes you are correct

[urban stag]

how is this for a plan?

1. four of us run an updater script and not a relayer

• each set to update like 12x per day; each funded with 1000 NAAN

2. five of us run relayers

• updates set to manual; each funded with 5000 NAAN

time horizon being 7 days

[paper pulsar]

why don't we create our Namada <> Cosmos configurations similarly for each other? the only difference is in the memo

[jade dust]

this is a good plan
2nd group is the main in fact, and the 1st is just to make sure nothing gonna broke

[cunning finch]

It looks good to me

[paper pulsar]

seems good

[paper pulsar]

do u mean in hermes config.toml?

[urban stag]

great! thanks for all this support

[cunning finch]

Yep

[urban stag]

is there an example of a bash script for cron?

[jade dust]

you need config.toml

and channel frist of all

[paper pulsar]

can i dm u

[jade dust]

so command will be after we create channel; AFK for a couple of hours

[cunning finch]

Ofc

[toxic cave]

We will, as we talked before 🙌🏼

[cunning finch]

We can take Min. epoch duration: 43200 to get an estimate in seconds of an epoch and then: 43200 * 2 = 86400 * 0.66 = 57024s -> this is the "ideal" trusting period on Namada's side

[molten drum]

amazingly amazing

[urban stag]

great! @cunning finch i made you an editor, are there other setting that relayers will need? and/or will be needed to open the channel

@quaint arch is going to take over coordination
i think he'll start with sending some tokens (NAAN & ATOM), and then discuss opening the channel

for reference, here's the coordination sheet: https://docs.google.com/spreadsheets/d/13tnM-o8Cmcpmu7Q9_7xGRZHmGWCUxT4i2UIdlpF3KII/edit#gid=0

@here take note your role

[inland nebula]

сosmos side trusting period?

[pallid wolf]

upd addrs relayer

[paper pulsar]

hey @minor solstice don't forget to comment your address in the spreadsheet

[paper pulsar]

check the spreadsheet

[jade dust]

supporting IBC channel [tbd], set updates to manual- this can cause a little confusion ) ( would be better if u can remove this)

[cunning finch]

Just added the trusting period as well for Cosmos testnet. Aside from that, I think default values can work fine. Will add gas settings as well just in case (unsure if they are ideal on Cosmos side, but it worked to me previously)

[quaint arch]

sent naan and testnet atom to everyone
(@minor solstice once you add your addresses i'll send you some too)

[silk totem]

don't forget to use filter packet to only relay your own channel messages

[quaint arch]

so now we just have to agree on a canonical channel-id, which bengt needs to compile the wasm tomorrow and make the governance proposal
and then keep it active during and after the proposal period or we have to start all over again

[fathom bloom]

I have atom test if needed and many naan (reward)

[paper pulsar]

@cunning finch

[cunning finch]

Yeah, but that's after channel creation. Will add it afterwards 🙂

[pallid wolf]

I'm all set, I just need cosmos coins (naan is available).
ready

[cunning finch]

So now we have to agree on who will create the canonical channel?

[quaint arch]

sure, is anyone set up and ready to make a channel?

[elder cipher]

hey im running IBC relayer between Theta and Namada, but just have 2 ATom and 650 naan in my wallet, so maybe i will need more tokens

do i must change my hermes config to the same setting on google doc ?

oh i saw a cutoff

[cunning finch]

Ideally yes, especially the trusting period, even though if you already set a value that's notably lower than the recommended one when you created the channel, we'll need to create a new one with those

[elder cipher]

anyway if u need more volunteer, im ready

[minor solstice]

asked if someone could input my values as per message above, but nobody seems to have taken the task

[minor solstice]

someone do me a favor and input these? I don't have easy access to edit google docs

[minor solstice]

@quaint arch my addresses are here

[minor solstice]

do we all relay the same channel or?

[pallid wolf]

You don't need access, post a comment and they'll add it all up

[molten drum]

Daniel adding atm)

[pallid wolf]

right click, comment, later it will be entered as main text

[cunning finch]

Yep, added

[minor solstice]

bro, I meant what I said. thanks to whoever put it in there 🙂

[quaint arch]

got a channel:

SUCCESS Channel {
    ordering: Unordered,
    a_side: ChannelSide {
        chain: BaseChainHandle {
            chain_id: ChainId {
                id: "shielded-expedition.88f17d1d14",
                version: 0,
            },
            runtime_sender: Sender { .. },
        },
        client_id: ClientId(
            "07-tendermint-3546",
        ),
        connection_id: ConnectionId(
            "connection-1743",
        ),
        port_id: PortId(
            "transfer",
        ),
        channel_id: Some(
            ChannelId(
                "channel-1142",
            ),
        ),
        version: None,
    },
    b_side: ChannelSide {
        chain: BaseChainHandle {
            chain_id: ChainId {
                id: "theta-testnet-001",
                version: 0,
            },
            runtime_sender: Sender { .. },
        },
        client_id: ClientId(
            "07-tendermint-3518",
        ),
        connection_id: ConnectionId(
            "connection-3624",
        ),
        port_id: PortId(
            "transfer",
        ),
        channel_id: Some(
            ChannelId(
                "channel-4118",
            ),
        ),
        version: None,
    },
    connection_delay: 0ns,
}

[cunning finch]

whoop, just made one as well haha

[cunning finch]

changing to that one

[quaint arch]

for the 'updaters', does this look right? we just have to run this in a cron job?

hermes --config ~/.hermes/config.toml update client --host-chain theta-testnet-001 --client 07-tendermint-3518
hermes --config ~/.hermes/config.toml update client --host-chain shielded-expedition.88f17d1d14 --client 07-tendermint-3546

[cunning finch]

Yeah, a loop in a bash script will do the trick as well ```#!/bin/bash

while true
do
hermes update client --host-chain shielded-expedition.88f17d1d14 --client 07-tendermint-3546
hermes update client --host-chain theta-testnet-001 --client 07-tendermint-3518
sleep 6000
done```

[quaint arch]

@minor solstice sent you funds
@pallid wolf @elder cipher did i get your addresses yet?

[pallid wolf]

mikhail88888

[elder cipher]

not yet, wait me a minute

namada : tnam1qqd3z5qeg0pu0s5h24szxq4pe4dgd3naxsztstjw
cosmos : cosmos1tqzktq6ukyvu83r2sd8g3mjsjhp2pf9q08nxzc

[minor solstice]

got funds on nam. were you sending on atom as well?

[quaint arch]

yes should have come on both

oh hang on, one sec

[minor solstice]

received on atom

as well

got my relayer set up with my pubkey in memo, assuming it's best if we remove that before going live?

[elder cipher]

okay , 2 IBC channels are healthy

[minor solstice]

do we need to do anything special to relay this channel, or just configure permissions so this is what we relay?

[cunning finch]

Transfers from/to working fine

transfer/channel-1142/uatom: 160```

```gaiad q bank balances cosmos1ay642tntyt36kpgu7x2ynkxptq0gar4qenf7kt
balances:
- amount: "16"
  denom: ibc/F04FF8C3379F78D47F8839AE06C6EA217176BE98D0FA0550914195B4BF501E23```

[elder cipher]

i will comeback next 4 hours, its 4AM rn, sry guys 😦

[quaint arch]

ty, tokens should be there when you get back 😉

[quaint arch]

nothing special, but you might want to set filters to only those channels so you don't run out of gas early

[minor solstice]

exactly. I'll have to create a channel of my own to create the client I think but guessing that's not an issue. will get on it asap

[minor solstice]

wyt @quaint arch remove this?

[quaint arch]

no reason to but up to you

[minor solstice]

alright I'll just leave it there then.

can track what goes through mine

[cunning finch]

In order to register it for shielded set rewards, do we need the tnam address-like denom of the (shielded?) testnet atoms, right?

If I'm not wrong

[quaint arch]

yep which depends on the channel (though the code will convert it from human readable)

const ATOM_BASE_TOKEN: &str = "uatom";
let ibc_atom_denom = format!("transfer/{ATOM_CHANNEL_ID}/{ATOM_BASE_TOKEN}");
let ibc_atom_token = ibc::ibc_token(&ibc_atom_denom);
# then use this to change various parameter values

[warped holly]

Omw home, just seeing this. Is the issue with clients randomly freezing resolved btw? Relayers here who had this happening? Or was this related to a specific hermes version?

[minor solstice]

trying to create client but atm got severe issues with my rpc. thought only validator. some widespread issues with timeouts idk why. but everything complex on rpc fails atm (anything with lots of calls). I'll try later and see if it calms down. if more practical to have someone else run it lmk, think it'll come through eventually though

[molten drum]

why do you need a client? for the sake of testing by itself?

[minor solstice]

was under the impression a client needed to be created in order to start the relayer. that's not the case?

[molten drum]

nope, you can use an existing one, that is the target here: have one canonical channel which already created

[minor solstice]

I can boot up hermes fine ofc but have a feel it'll stall on relays for same reason. but definitely willing to give it a go

[minor solstice]

ofc, just thought I had to go the other route

let me see how it behaves then..

[molten drum]

the other route is what everyone did during last month and how my page appeared: >1k not working clients 😄

[minor solstice]

aye

[minor solstice]

alright started relaying, let's see how this goes

[molten drum]

those channels are not active atm... actually, their clients have expired

[knotty jacinth]

Weird. Thanks for pointing that out, I should have monitored more closely. I take it back then 🙂

[mighty cloud]

can you guys add me as well?

[elder cipher]

we must change to exact tendermint id and channel id in google sheet right ?
btw i received naan tokens but atom still low on funds

[jade dust]

relayer set up, updates and tx checked. working!

[quaint arch]

which issue was that? hadn't heard about it yet

[quaint arch]

i'm also finding hermes times out on namada rpc calls about ~30% of the time, even just a hermes health-check fails randomly. maybe just system specs issue, if the node is busy with masp stuff?

is anyone else getting that?

[warped holly]

Will link to it in the other chat

[cunning finch]

Facing the same time to time

[minor solstice]

think it's a more general rpc performance issue

created a gh issue on this, but so far hasn't gotten much attention (the more general issue)

[quaint arch]

are people still running v1.7.4-namada-beta7 or should we switch to v1.7.4-namada-beta8-rc from last week?

[minor solstice]

I'm running beta-7

the long form memo branch

[cunning finch]

Same

Beta7

[minor solstice]

imma be out for a few hours, will check back here later if any developments

[warped holly]

beta-7 as well, haven't set up my relayer yet though. Are there enough people already? Saw that gavin preferred a small set.

[cunning finch]

I think I'm currently myself relaying, pretoro, and Gylijov from Nodiums

Sheet updated. @molten drum hey Liver, do you have Hermes running already?

[warped holly]

Aah, aah I read passed that, thank you!

[cunning finch]

Afaik the only active updater is Spork for now

I'm also running update commands though, just in case

[quaint arch]

only in the sense that we don't have unlimited tokens to keep everyone topped up, anyone is welcome to relay ofc if they already have funds for gas

[warped holly]

Got it! I'll just turn one on, no need to place me in the sheet. If there comes a moment where we run short on relayers, some have difficulties or have to bail out then I'm here!

Still got like 10 testnet ATOM

Curious to see how quick that depletes

[quaint arch]

ty for the offer! we're not adding more to the sheet but only because we don't want to add too many and run out of gas funds

[warped holly]

Hehe yw and yes that's for sure the right choicee

Think it's on now. Channel-1142 <=> channel-4118.

How often do you update btw?

Could set a script up for that as well just in case

Ah 12x a day, so every 2hrs, got it.

[elder cipher]

synced client id and host chain with sheet, do we just wait now ?

[warped holly]

Oh lol, I just realized that response wasn't towards me 🤣!

Awkwarddddd

[molten drum]

++ from yesterday yet
just do not see real activity atm, yesterday tried 2 transfers, and both were done through your relayer

[quaint arch]

cosmos tx didn't go through last time but should have it now

[quaint arch]

yep

[elder cipher]

should i do this loop scripts ? @quaint arch

[quaint arch]

today we'll hopefully test the shielded rewards proposal code on campfire and then post a proposal on shielded expedition for voting next week (we don't want voting period to be during the long weekend)

[quaint arch]

yep that works

[elder cipher]

ThreadId(01) connection handshake already finished for Connection { delay_period: 0ns, a_side: ConnectionSide { chain: BaseChainHandle { chain_id: shielded-expedition.88f17d1d14 }, client_id: 07-tendermint-3546, connection_id: connection-1760 }, b_side: ConnectionSide { chain: BaseChainHandle { chain_id: theta-testnet-001 }, client_id: 07-tendermint-3518, connection_id: connection-3631 } }

i updated hermes relayers to the same cliend id as google sheet, but the connection id is difference.
its still okay right ?

[quaint arch]

hmm, i'm not sure why... it should look like this

~$ hermes update client --host-chain shielded-expedition.88f17d1d14 --client 07-tendermint-3546
2024-03-29T14:32:35.107647Z  INFO ThreadId(01) using default configuration from '/home/hermes/.hermes/config.toml'
2024-03-29T14:32:35.108255Z  INFO ThreadId(01) running Hermes v1.7.4+38f41c62
SUCCESS [
    UpdateClient(
        UpdateClient {
            common: Attributes {
                client_id: ClientId(
                    "07-tendermint-3546",
                ),
                client_type: Tendermint,
                consensus_height: Height {
                    revision: 0,
                    height: 20964457,
                },
            },
            header: Some(
                Tendermint(
                     Header {...},
                ),
            ),
        },
    ),
]

[minor solstice]

you know if proposal 447 is you guys or pretendoors?

[elder cipher]

Because my relayers had difference channel id previously , so i created another one with the client id on sheet.
My update client log still the same as you

[quaint arch]

proposal 447 is 'proposals swarm', do you mean a different id?
either way, we didn't make any proposals recently nor did heliax

[minor solstice]

sorry typo. 477

[molten drum]

you need the channel id in the hermes config and not the client
and in the update script yes, client )

[molten drum]

hm, anyway funny and strange here how do you get diff connection with the same clients, only if someone created manually an additional connection....

lol

2024-03-29T14:59:00.840454Z  INFO ThreadId(01) using default configuration from '/home/namada_rpc/.hermes/config.toml'
2024-03-29T14:59:00.841186Z  INFO ThreadId(01) running Hermes v1.7.4+1ede457f
SUCCESS ConnectionEnd {
    state: Open,
    client_id: ClientId(
        "07-tendermint-3546",
    ),
    counterparty: Counterparty {
        client_id: ClientId(
            "07-tendermint-3518",
        ),
        connection_id: Some(
            ConnectionId(
                "connection-3624",
            ),
        ),
        prefix: ibc,
    },
    versions: [
        Version {
            identifier: "1",
            features: [
                "ORDER_ORDERED",
                "ORDER_UNORDERED",
            ],
        },
    ],
    delay_period: 0ns,
}

2024-03-29T14:58:28.581937Z  INFO ThreadId(01) using default configuration from '/home/namada_rpc/.hermes/config.toml'
2024-03-29T14:58:28.582374Z  INFO ThreadId(01) running Hermes v1.7.4+1ede457f
SUCCESS ConnectionEnd {
    state: Open,
    client_id: ClientId(
        "07-tendermint-3546",
    ),
    counterparty: Counterparty {
        client_id: ClientId(
            "07-tendermint-3518",
        ),
        connection_id: Some(
            ConnectionId(
                "connection-3631",
            ),
        ),
        prefix: ibc,
    },
    versions: [
        Version {
            identifier: "1",
            features: [
                "ORDER_ORDERED",
                "ORDER_UNORDERED",
            ],
        },
    ],
    delay_period: 0ns,
}```

we have two connections on the same client 07-tendermint-3518 (cosmos side)

[paper pulsar]

lol

[cunning finch]

If I'm not understanding wrong Hermes docs, this is possible actually? 🤔
https://hermes.informal.systems/documentation/commands/path-setup/connections.html#new-connection-over-existing-clients

[molten drum]

yes, but why someone created them

and this makes me thinking I am missing something 😄

2024-03-29T15:01:59.428249Z  INFO ThreadId(01) using default configuration from '/home/namada_rpc/.hermes/config.toml'
2024-03-29T15:01:59.428729Z  INFO ThreadId(01) running Hermes v1.7.4+1ede457f
SUCCESS [
    PortChannelId {
        channel_id: ChannelId(
            "channel-1140",
        ),
        port_id: PortId(
            "transfer",
        ),
    },
    PortChannelId {
        channel_id: ChannelId(
            "channel-1141",
        ),
        port_id: PortId(
            "transfer",
        ),
    },
    PortChannelId {
        channel_id: ChannelId(
            "channel-1142",
        ),
        port_id: PortId(
            "transfer",
        ),
    },
]```
```hermes query connection channels --chain shielded-expedition.88f17d1d14 --connection connection-1760
2024-03-29T15:03:24.786621Z  INFO ThreadId(01) using default configuration from '/home/namada_rpc/.hermes/config.toml'
2024-03-29T15:03:24.787084Z  INFO ThreadId(01) running Hermes v1.7.4+1ede457f
SUCCESS []```

[minor solstice]

do I need to update client to the client listed in the canonical channels json?

[molten drum]

update script? yes, you need to update specifically those clients

[atomic tendon]

Created a channel below.

tnam1qq7nfjqrsg8x9vssf87wamav883dw6eargagd763
cosmos1980eh8cnppjs6yn6wre7tgj9ywe6swwc75hn6e

SUCCESS Channel {
    ordering: Unordered,
    a_side: ChannelSide {
        chain: BaseChainHandle {
            chain_id: ChainId {
                id: "shielded-expedition.88f17d1d14",
                version: 0,
            },
            runtime_sender: Sender { .. },
        },
        client_id: ClientId(
            "07-tendermint-3589",
        ),
        connection_id: ConnectionId(
            "connection-1761",
        ),
        port_id: PortId(
            "transfer",
        ),
        channel_id: Some(
            ChannelId(
                "channel-1154",
            ),
        ),
        version: None,
    },
    b_side: ChannelSide {
        chain: BaseChainHandle {
            chain_id: ChainId {
                id: "theta-testnet-001",
                version: 0,
            },
            runtime_sender: Sender { .. },
        },
        client_id: ClientId(
            "07-tendermint-3529",
        ),
        connection_id: ConnectionId(
            "connection-3632",
        ),
        port_id: PortId(
            "transfer",
        ),
        channel_id: Some(
            ChannelId(
                "channel-4125",
            ),
        ),
        version: None,
    },
    connection_delay: 0ns,
}

[molten drum]

for what?

[atomic tendon]

Client will be updated everyday.

[woven sky]

i need chanel cosmos and se working for ibc dapp build

Who is doing well with it?

[paper pulsar]

its confusing

pleae just stick to the canonical channel

[elder cipher]

Hermes guide, you can create new connection with existing client id

[molten drum]

I know, but for what? 😄

[elder cipher]

my channel id is diff to the sheet, i dont know how to change it to

[molten drum]

what do you mean your is different? did you create a new channel and it is different?
you only need to take that canonical and put it into hermes config in filters if you want to stick to it

[elder cipher]

Oh okie got it, thank you

[urban stag]

we got shielded set rewards working on a local testnet 🎉
this was super handy! #1222959128629215262 message

[urban stag]

we're standing by for word from Brent to check our wasm code, discuss parameters (like inflation amount, target), and overall if we're good to go

i think we'll want the voting to begin mid-day Monday (i know it's a holiday for some) and have the voting period be active for validators all of Tuesday as well

ideally we'd have shielded set rewards active by Friday

[lament barn]

Sorry been mostly out of commission today but will get the update script going and settings updated tomorrow

[urban stag]

okay, Spork is the one leading this

[minor solstice]

are we in scoring v2 already post fork, or only when shielded set rewards are active?

[cunning finch]

Guys, please don't create more channels. We already supporting the canonical one - check spreadsheet on pinned messages for further reference

[minor solstice]

read the channel above..

[lament barn]

@quaint arch using update script now and updating canonical channels

[molten drum]

totally wrong channel...

[minor solstice]

#1220430590315204660

[swift grove]

I have setup my relayer between selected channel shielded-expedition.88f17d1d14 channel-1142 and theta-testnet-001 channel-4118
My address
cosmos1l2hfmy3uengp2pnec9qhuz504j0m86y8slkgst
tnam1qqevylt0tatda2wzjz5v9r4u8gmm3tkptq9n4q7g

Can I get some gas ?
https://explorer.polypore.xyz/theta-testnet-001/tx/DD7E62867B44E3EEAE015327868FF88A6B6966AF940F43C21ABECB2944C97BC5

[urban stag]

omg there are sooo many Cosmos Hub testnet IBC channels 😂 https://services.liveraven.net/cosmos-testnets/namada/ibc-channels

okay how is the canonical channel (the only channel we care about) doing? target Channel 4118

[warped holly]

Still working, just tested via site. Implementing cosmos atm.

Just refactoring the code still and abstracting out logic so I can continue working on it with a clear mind.

I also relay, and I still got like 10.28 ATOM compared to 3 days ago 😮 (which was also around 10)! Though not many are probably using the channel.

(and also update), so it might be quite cheap. Though yeah the real deal starts when everyone will make use of it.

Edited: forgot to mention how much I had 3 days ago.

[urban stag]

great! this is what we expected. the constraint will be NAAN, we think

[urban stag]

interface looks awesome

[warped holly]

Nice!

[warped holly]

Tysm man!

[minor solstice]

let me check my wallets and see how they look

not sure mine has relayed a lot

looks like it did a few only

[quaint arch]

big thanks to everyone here for helping out on this! everything looks good ahead of the ssr proposal, hopefully coming later today
if anyone needs anything or has questions, lmk

[cunning finch]

Let's go 🤟

[minor solstice]

not sure why my relayer isn't picking up much of these

[cunning finch]

It's like the far west, so the fastest ones (it may related to hermes config, server specs, etc.) are likely to relay packets more frequently - anyway, if those start to have latency, the rest of operators are there to keep relaying packet as normal

[minor solstice]

alright, I'll be the backstop then 🙂

[cedar otter]

Has anyone manage to send shielded ATOM through the channel ? Or is it still not possible because shielded transfer are broken ?

[wet obsidian]

My submission made two days back is not showing.

[molten drum]

the last update of the file is from March 29 if I recall correctly
you can check 2nd tab in the spreadsheet

[quaint arch]

anyone else seeing a frozen client?

ERROR foreign client error: client 07-tendermint-3546 on chain id shielded-expedition.88f17d1d14 is frozen: client state reports that client is frozen

[urban stag]

@here

[warped holly]

Ah damn it happened?

Just a sec, lemme check

Yeah crap, seeing it

[molten drum]

yep ++

[molten drum]

btw, I recall that I used to update once per hour and the trusting period was something about a bit more than 1 hour, especially as saw that the client has expired in 2-3 days all the time
thought it was fixed somehow long time ago

ppl continue to create their own channels instead of using canonical...

[silk totem]

frozen != expired

[warped holly]

Freezing of client is connected to misbehavior last I checked out this problem.

Dunno how exactly it happens though.

[molten drum]

updated 11 minutes ago

[molten drum]

ok, wrong wording, nvm

[silk totem]

i never experienced a frozen client on Noble, but happened to my osmosis client 3 times

[misty grove]

liver23 has been warned

Reason: Duplicated text

[warped holly]

Did the freezing of clients start to happen ever since the memo size was increased? I don't remember this to happen before that long memo branch. But unsure, cause back then I had an unfiltered relayer on steroids relaying on all channels.

[warped holly]

Damn, saw it live. Rip.

[molten drum]

crap)

[warped holly]

😂

[molten drum]

just before frozen error

nothing unusual

[warped holly]

Lemme check mine, see if something odd's in there.

[molten drum]

and those messages are from Cosmos side, looking to the height

https://www.mintscan.io/cosmoshub-testnet/tx/4028D1A5ABB9B4D0FAAB4CBFB5F36E31E353E93FC38538480494960CBE330604?height=21003472
1st ibc timeout that I see on my wallet from cosmos side

[fathom bloom]

when i make ibc to cosmos osmo noble stargaze and celestia i found that client frozen after ~ trusted period and cosmos not living more 2 days

update not helping for me

[molten drum]

In uncommon situations, a highly valued client may become frozen due to uncontrollable circumstances 😄

[warped holly]

Well. We're being too high class now.

[molten drum]

I didn't see any misbehaviour evidence in the log also
When submitting evidence of Misbehaviour to freeze a malicious client

but updating it once per hour helped to keep it alive

If the one third of the validator set of the chain the client represents decides to collude, they can sign off on two valid but conflicting headers each signed by the other one third of the honest validator set. The light client can now be updated with two valid, but conflicting headers at the same height. The light client cannot know which header is trustworthy and therefore evidence of such misbehaviour is likely to be submitted resulting in a frozen light client.```

[cunning finch]

What was the trusting period on Namada's side?

The one I suggested in docs?

[molten drum]

yes, but can't answer for everyone
I used 4k+ smth and updated once per hour before, and didn't see frozen channels
but recall that the client had two connections as well, what I wrote on the first day, just that should not be the problem usually

[cunning finch]

Ofc it's the same (technically speaking)

[molten drum]

statuses are not by nature, but the result is the same, yes))

[cunning finch]

Yep

[molten drum]

dunno, can two connections lead to that inconsistency? #1222959128629215262 message

it is like apphash in the chain seems

if we missed update it should be Expired and not Frozen

[cunning finch]

I think I already reported internally something related to frozen height on ibc clients on Namada's side one month ago or more

[warped holly]

Yeah we spoke bout it a month or so ago

[jade dust]

those with updating scripts, when was your last update from namada's side?

any chance to get tx?

hmm, weird tbh. something seems to be wrong on namada side

[silk totem]

what height your client was frozen on ? i mean output of this command

hermes query client state

[cunning finch]

I assume that frozen height can't still be fetched on Namada, and I think it's somehow related to the issue we are facing

[silk totem]

correct, it always shows height 1

[warped holly]

Yeah, for some reason I'm seeing height 1, revision 0.

[warped holly]

Could make sense indeed.

[cunning finch]

Wanna open the issue, but I don't know in which repo would fit better, Hermes or Namada? Had the same doubt when I reported it a while ago

[warped holly]

I used to think it's hermes, but now I'm leaning more towards namada itself.

But Hermes might be better, cause it's related to relaying

We dunno what the actual problem is

[jade dust]

what if we try rly instead of hermes

[warped holly]

Might as well be something in the fork of hermes that's not taken care of well

[warped holly]

I did see someone (from the rly team?) asking for guidance to get namada integrated in rly

[cunning finch]

Yeah, but is it not supposed that chain should send the frozen height, in order to be fetched from Hermes? 🤔

Unsure tho

[warped holly]

Yeah, I'm not sure how to see if the namada protocol supports this already. If it doesn't, then yes it's likely that it fits the namada repo better

[jade dust]

cant it be related to hardfork?

[warped holly]

Tho we do isolate the issue then

It might not be the actual prob I mean to say

[jade dust]

i mean hermes could try to update height that is lower then earliest because we dont have history

[cunning finch]

Gonna open it once I get back to home, and let's see what team discovers

[silk totem]

no it's an old issue existing from early days of SE

[warped holly]

Cool! Thank you :)!

[warped holly]

Or be an outlaw and copy paste the issue into both repos 🤭

[molten drum]

wondering then why in the config with the trusting period around 4k+ sec and update once per 1 hour I didn't get my own clients frozen even one time?

still have a suspicion it can be related to the multiple connections... dunno why 🧐

[warped holly]

Hmm, I sadly had all my channels frozen

I do still have an osmosis one open I think

3rd day now.

[molten drum]

as per that explanation when you get two correct hashes and can't decide which to use, then it can be frozen

[pallid wolf]

++

ERROR foreign client error: client 07-tendermint-3546 on chain id shielded-expedition.88f17d1d14 is frozen: client state reports that client is frozen```

[cunning finch]

But that would be in case of hash fork? Or maybe I'm missing smth

[molten drum]

dunno 🤔 👻

[quaint arch]

last successful client update for me was apr 1, 19:00 utc

[cunning finch]

Okay, just opened the issues on both Heliax Hermes fork and Namada repositories
https://github.com/anoma/namada/issues/2985
https://github.com/heliaxdev/hermes/issues/27

[molten drum]

as said before, why then this #1222959128629215262 message 🤔 I really didn't get clients frozen for a long time
we need to check 2 things as for me, with the new channel

1. be sure that the client is unique, no multiple connections on it and still stick to the current trusting period from the file
2. if it will be frozen again, reduce the update and period to ~1 hour and check again
   we can use two new channels for that to check in parallel

also if it is Frozen, it is not due to missing the trusting period, it is some non-deterministic misbehaviour

[cunning finch]

I'd wait for team to dig into it

[silk totem]

i had my osmosis channel updated every hour but eventually it got frozen too
i also set misbehaviour=false in hermes configs to disable misbehaviour detection
so i guess this issue has nothing to do with trusting period or update frequency

[warped holly]

Yes, I remember this, also same for me when I tried it afterwards

Though atm I still have an osmosis channel open, not sure when it will freeze.

[silk totem]

🤔 one of my channels frozen just after 1 day, another one lived for a week or so
i'll open a new one tomorrow to examine it more carefully

[warped holly]

Yeah it's really weird, it might be due to a tx?

Keep thinking that it might be something with a wrong shielded tx, with an incorrect memo, but this is a leap.

[silk totem]

btw which hermes branch /version should we use now ?

[warped holly]

I'm using 1.7.4 beta 7 still

the long-memo version tho

Think most here do (either that branch or clean branch)

[cunning finch]

Same git status On branch 1.7.4-namada-long-memo Your branch is up to date with 'origin/1.7.4-namada-long-memo'.

[warped holly]

Btw just for a quick check: this one translates as a -dirty-version?

hermes 1.7.4+dc498bbf-dirty

[silk totem]

i used same version too

[cunning finch]

I recompiled it to double check and getting this one./hermes -V hermes 1.7.4+1ede457f

[warped holly]

Ah I think I did something dirty then

I'll recompile, cause that commit version I gave is totally off btw lol, wth I do.

[cedar otter]

@warped holly checkout using the tag version (from the docs) to be sure. Lankou also had that kind of issue at some point because he compiled at the wrong commit

[warped holly]

Yeah mine was cause of own custom changes, I was merging two separate branches together at some point in the past and forgot to replace the binary again with a clean one 🙈

[cedar otter]

I feel you pain

[minor solstice]

may be on an older version here..

but thought I checked out the long-memo branch

[minor solstice]

made some infra changes, so assume my relayer should be more stable now

[minor solstice]

do we have agreement on which is the right commit?

[cedar otter]

https://docs.namada.net/operators/ibc#from-binaries
or
https://docs.namada.net/operators/ibc#from-source
👇
export TAG="v1.7.4-namada-beta7"

[minor solstice]

I believe the faq lists another version?

can't check atm but thought it prefered the long-memo for SE

[cedar otter]

2. Is the latest hermes branch v1.7.4-namada-beta7 compatible with Namada +v0.31.0?
   It's not up to date with +0.31.0 but it's based on a version that is compatible with +0.31.0.

oh 😮 I didnt know there was that long memo version !

I've done all my test with the version from the docs. got no trouble with it

[molten drum]

do we have a new canonical channel pair?

[minor solstice]

did the old one expire?

[molten drum]

frozen, the whole discussion above ))

[cedar otter]

Is there a way to unfreeze the thing?
I've since governance proposal on some IBC chains to restore channels, is this it? It frozen=dead= create new one ?

[molten drum]

it's impossible to unfreeze something in testnets, you need to get consensus on voting 😄

[cunning finch]

Also, there's no such proposal type in Namada yet. It wouldn't solve the issue though

[molten drum]

are you going to create a new channel(s)? or someone else

[cunning finch]

Nope, we haven't informed yet and don't think that issue is going to get resolved soon. Shielded set rewards can still be tested fortunately. In my case I sent a few testnet atom to Namada

[molten drum]

ok, I'll spinup my own to check if it is a persistent problem

[quaint arch]

not yet but a heliax dev is going to suggest some config changes and then we'll try again

does anyone have anything in their logs that would give an idea of what height the client froze at? even approximate?

[molten drum]

nope 😦 looked through and nothing suspicious
the only thing worrying me is that we had 2 connections up for that client, it should not be a problem, but who knows here
as per that explanation how the client can be frozen it can be that the same tx came through two channels simultaneously and therefore due to that a misbehaviour appeared

[pallid wolf]

looked at the logs yesterday and nothing...

[warped holly]

Lemme check, I had relayed something 10 min before it froze. That gap might be too big though.

This one didn't freeze

2024-04-02T06:48:42.499482Z  INFO ThreadId(1157) worker.batch{chain=theta-testne
t-001}:supervisor.handle_batch{chain=theta-testnet-001}:supervisor.process_batch
{chain=theta-testnet-001}:worker.packet.cmd{src_chain=theta-testnet-001 src_port
=transfer src_channel=channel-4118 dst_chain=shielded-expedition.88f17d1d14}:rel
ay{odata=cleared/78a458bd ->Destination @0-21020899; len=1}: response(s): 2; Ok:
A8CF032D7E60AC2DACC1AB3990763852B095E0562F583848539A0DB8625FB43F; Ok:7C297721334
37DFC30F09159FD1D30ABD58253AB57FF2C9E4CD2D0021D06F8C0 target_chain=shielded-expe
dition.88f17d1d14 

This one did freeze:

2024-04-02T06:58:04.460934Z  INFO ThreadId(1157) worker.batch{chain=theta-testne
t-001}:supervisor.handle_batch{chain=theta-testnet-001}:supervisor.process_batch
{chain=theta-testnet-001}:worker.packet.cmd{src_chain=theta-testnet-001 src_port
=transfer src_channel=channel-4118 dst_chain=shielded-expedition.88f17d1d14}:rel
ay{odata=cleared/2cda6347 ->Source @0-281973; len=1}: response(s): 1; Ok:E589293
A28347ECBB4685D38375CAC7977AE237EA51B721871B0878A01A03ADB target_chain=theta-tes
tnet-001

I'm on my phone, think the copy paste is screwed up.

I'm comparing two different directions here tho if I check the target_chain value.

Actually the last one before the frozen one that had target chain theta testnet 001 had an error. But I've seen that error appear more often

4-04-02T06:47:52.878993Z ERROR ThreadId(18) worker.batch{chain=shielded-exped
ition.88f17d1d14}:supervisor.handle_batch{chain=shielded-expedition.88f17d1d14}:
supervisor.process_batch{chain=shielded-expedition.88f17d1d14}:worker.packet.cmd
{src_chain=shielded-expedition.88f17d1d14 src_port=transfer src_channel=channel-
1142 dst_chain=theta-testnet-001}:relay{odata=2bd66fb8 ->Destination @0-281918;
len=1}:send_messages_and_wait_check_tx{chain=theta-testnet-001 tracking_id=2bd66
fb8}:send_tx_with_account_sequence_retry{chain=theta-testnet-001 account.sequenc
e=190}: failed to broadcast tx with unrecoverable error response=Response { code
: Err(22), data: b"", log: "packet messages are redundant", hash: Hash::Sha256(7
8B9529F856E4CF0A6154457068669D45BCAEF680F354F0FE833CA87B14C6C00) } diagnostic=un
known TX sync response error: 22
2024-04-02T06:47:52.879074Z  INFO ThreadId(1177) worker.batch{chain=shielded-exp
edition.88f17d1d14}:supervisor.handle_batch{chain=shielded-expedition.88f17d1d14
}:supervisor.process_batch{chain=shielded-expedition.88f17d1d14}:worker.packet.c
md{src_chain=shielded-expedition.88f17d1d14 src_port=transfer src_channel=channe
l-1142 dst_chain=theta-testnet-001}:relay{odata=2bd66fb8 ->Destination @0-281918
; len=1}: response(s): 1; Error with code 22:78B9529F856E4CF0A6154457068669D45BC
AEF680F354F0FE833CA87B14C6C00 target_chain=theta-testnet-001

Ima go behind the pc and cut a segment of my logs, this copy pasting is probably not looking okay on a desktop.

[jade dust]

this just mean someone relyed before you. nothng unusual

[warped holly]

Ah right!

Here's a log snippet around where it froze, might narrow the height down a bit more @quaint arch.

Made sure not to narrow it down too much, the 'surrounding' txs before and after might be useful.

[quaint arch]

amazing, thanks! are the timestamps in utc?

[jade dust]

nothing special at logs. have tons of such at all relayers(

[warped holly]

Yes UTC

[warped holly]

It might be useful to narrow down, perhaps if everyone does this we can figure out roughly at which height it froze

I'll fix switching between chains today, then redeploy my ibc app with osmo-test-5/theta-testnet-001 => namada-se shielded fix.

Will then continue to experiment with the SDK, see what's possible for namada-se => the other chains in a shielded manner.

[cunning finch]

Honestly don't think it has to do with config as frozen height isn't being emitted - this prob has to do with how ibc-rs is integrated in Namada

But who knows

[jovial compass]

is everything working with channel-1142/channel-4118? I sent some uatom (both to a shielded and non-shielded addr) but not sure if I did it correctly and nothing appeared on the Namada side. Sending from Namada gets rejected by VPs

[molten drum]

the whole discussion above is about the client was frozen...

[jovial compass]

Thanks, I saw that but wasn't sure whether that was just one relayer or everybody's. Also not sure why my Namada tx was rejected by VPs, that's probably just something I'm doing wrong.

[molten drum]

I recall someone got it rejected when used non-working channel (expired/frozen client)

https://services.liveraven.net/cosmos-testnets/namada/ibc-channels you can find working here (updated once per 2h)
there are few problems with getting osmosis data but cosmos is actual

[jovial compass]

just discovered https://github.com/anoma/namada/issues/2730

[warped holly]

@quaint arch I'm not sure if this is a coincidence, but my osmosis client froze the moment I accidentally did a non-shielded IBC transfer to a znam address. This simply means I left the IBC memo empty in the transfer.

[warped holly]

Doesn't seem coincidental, my suspicion was already this, see the message I reply to

Hmm, I should create a new channel and try to replicate this.

Brb.

BINGO

2024-04-03T19:33:47.708799Z ERROR ThreadId(20) spawn:chain{chain=osmo-test-5}:client{client=07-tendermint-3191}:connection{connection=connection-2869}:channel{channel=channel-6731}:worker.packet.cmd{src_chain=osmo-test-5 src_port=transfer src_channel=channel-6731 dst_chain=shielded-expedition.88f17d1d14}: task aborting after encountering fatal error: link error: failed during a client operation: client 07-tendermint-3704 on chain id shielded-expedition.88f17d1d14 is frozen: client state reports that client is frozen

It immediately froze

So okay, uhm I can prevent it in the frontend, but this requires fixing on-chain.

This is what I did, as you can see it's set to UNSHIELDED, but I'm sending the tokens to a znam address.

Unshielded simply means it leaves the ICS-20 memo field blank, so this whole freezing thing is related to either the memo being empty/faulty or the address format being off. Or ofc both.

[cunning finch]

Great, hope that this save time and helps devs to identify and resolve the reported issue 🤟

[warped holly]

I wanted to fix it in my frontend already, but now think it's better to leave it like this for now. That way the edge case can be tested.

Deployed latest version of the IBC app btw, shielded from osmo or theta-testnet => namada working.

Also assets get auto-converted to known denoms + shows which channel they come from.

[jovial compass]

Testing now. I still can't see assets on the Namada side, but maybe takes a while for IBC to complete?

[warped holly]

The channels are frozen, just a sec haha.

I'm creating new channels for both chains atm, takes a bit.

But you can use other people's channels aswell. My default channels are just not working now.

[warped holly]

If anyone is going to try and freeze a channel with the edge case above, please don't use a channel of mine 😅.

It's okay if you accidentally do, cause I did the same by accident. Just make sure to toggle it to SHIELDED if you need to do a shielded tx.

[jovial compass]

the edge case is just leaving the memo empty?

[warped holly]

It's either:

• sending to a znam address and having the memo empty.
• sending to a znam and having an invalid memo.
• sending to a foreign address.

So it's either the address field causing issues or the memo field.

The latter is actually something I have to try still, a foreign random address.

But if you toggle to SHIELDED, the IBC memo generation is taken care of automatically.

[jovial compass]

on a related note, I'm wondering how to import the extension's shielded address into the CLI. no worries if you don't know, I can just set it manually

[warped holly]

viewing key or the pyment address? (payment address you can see in the extension under view keys)

Ah I get what you mean

No, I haven't gotten that solved, I keep getting a different Viewing Key if I import it

[ionic ravine]

if i am not wrong the memo is generated by keplr> or namada side

or it use a api to generate it somewhere

i guess it is a remote api

[warped holly]

I for now have that done via API, but next step is to get that shifted to SDK. Will deep-dive into the SDK tomorrow, cause I wanna tackle most stuff concerning namada => external chains.

Hope it's possible tho

[ionic ravine]

gotchu

just want to confirm that

nice work anyway

[warped holly]

It's no biggy to have this generated via an api, but I really, like really dislike CLI usage whatsoever lol.

[warped holly]

Thanks chad

Lemme redeploy with different channels.

Ah wait it already shifted

ah phew, glad to have made the default channels inside of an .env file. That way I can just change those in vercel's env settings and click 'Redeploy' on the latest deployment.

Might need to CTRL+F5, channel should say 6732 for osmosis.

[ionic ravine]

so the api, generate memo using the cli on the vm,with get request ? something like namadac ibc-gen-shielded --output-folder-path . --token x --amount x --port-id x--channel-id x --target x

[jovial compass]

Yes, finally! It worked! Thanks, you're a legend

[warped holly]

Yes something like that indeed! Will soon hop off that method tho, hopefully tomorrow already :)!

[warped holly]

Awesome!! Haha thank youu

[median island]

Great it worked l..

🫡🫡 salute you zen.

[warped holly]

Glad to hear! Saluteeee back brother!

[vivid sable]

Hey guys! Did someone manage to do shielded-ibc (using memo via ibc-gen-shielded) after fork and receive NAAN or UATOM tokens on shielded address?

[urban stag]

whoa @warped holly i'm just seeing all this now 👀

[warped holly]

Haha, that's okay!

[urban stag]

looks pretty exciting! @forest condor check this out

[warped holly]

So many threads/channels 😥!

I do have to say, please only replicate the freezing of channels on own channels 🤣...else if @quaint arch's going to test shielded tx's it might fail due to channels not working anymore.

The channel field is open tho, so it's also possible to type a different channel

[unique aspen]

I tried your app yesterday. I can see transactions on Mintscan that include a memo that I have set to my public key. But I did not see them show up in Namadexer.

[warped holly]

Hmm not getting decoded by namadexer?

Though it does get over to the other side?

[unique aspen]

I will try again

Yesterday I was not sure if the app was supposed to be functional.

Because of the discussions.

[warped holly]

Depends btw which direction you're trying. Some transactions will of course be on the other chain.

[warped holly]

Cool! Tell me if it's actin' up!

I periodically check if the channels I set as default are still working.

[unique aspen]

Some things were freezing up, like the Vault.

[warped holly]

Namada => other chain shielded is not implemented btw.

[unique aspen]

Now I see I have NAAN on different channels

[warped holly]

Ah yes, yes, this is the chaos and mess we created after letting so many create channels 😂.

I am guilty of it as well.

[unique aspen]

That was to be expected when the docs say you have to create channels to run a relayer, but is not clear about reusing existing channels.

[warped holly]

True man, it did allow me (and I believe many others) to understand relaying more in-depth now

And testnets are there to be made a mess out of

So yeah, this eventually is all good.

[unique aspen]

And to prove the S Task it is also safer and easier to create new channels, although the format document says you can also submit an update transaction

I did too, just to be sure

[warped holly]

Ah interesting, I did not know! That's indeed way simpler

[unique aspen]

But it is kind of ambiguous if that could be considered Operating IBC infrastructure

So I totally get why people are creating new channels

[unique aspen]

Is that also why setting From to Namada and enabling Shielded throws the error No results have been found on Vault?

[warped holly]

Yeah the extension doesn't have this implemented, it tries to get the balance for the shielded address, but this doesn't work (at least in my experience atm). That's where the No results have been found on Vault comes from.

I'm playing around with the SDK now, see if I can figure something out. Though I don't think sending actually from Shielded namada to an external chain will work with the extension.

So I'll either come up with something temporary, but this will likely mean having to handle the keys outside of the extension...

[unique aspen]

Ah I was wondering how it would have worked considering the resources necessary to shielded-sync

[warped holly]

Yeah, so right now:

• All things unshielded works, but who cares.
• Shielded works from external to namada.
• Balance fetching for osmosis and theta works (all assets). And for transparent namada addresses, though this is limited to only those you added and NAAN ofc.

Though hearing the latest in the validator circle, I'm not sure whether we'll be able to get something working properly since they stopped the SE development.

Shielded => shielded or shielded => transparent not working for me either, so I'm not even sure if that direction is doable now.

😥...

I'm ripping the SDK apart as we speak lol, so I'll see what I can get out of it.

[quaint arch]

sorry zen, not sure i understand...

are you talking about something like this?

gaiad tx ibc-transfer transfer transfer channel-4152 znam1qrylpn9848mz9v06sqefs9nekgqrj9hxa5v7p4e3l46j3q02sxsxymyjzyw9wt0fggtpn6qw2gpe3 100000uatom --from hub-relayer --keyring-backend test --node http://localhost:46657 --fees 7000uatom --chain-id theta-testnet-001 --memo ""

because if so, i haven't been able to replicate. or is it something specific to your app?

[warped holly]

Strange, this is exactly what I meant yes, let me try again.

Forgive me, I'll possibly destroy my osmo channel now.

Okay, very interesting

It froze just now when I used my relayer address

to transact with

[minor solstice]

I wonder if using a wrong version could cause someth like this?

[warped holly]

This, so that translates to:

osmosisd tx ibc-transfer transfer transfer channel-6732 znam1qqtnyzu97zknt3sff74aj8fnxxx6lr5gke2vqp8w8pmxz7dp25rt7jsmq46dn2g2t0al2ps2paryv 100000uosmo --from osmorelayer --chain-id osmo-test-5

I was trying all sorts of combi's with another address and it didn't freeze.

But now that I tried it with my relayer address that's relaying the osmo-test-5 channel it froze.

[warped holly]

Is hub-relayer also the one relayng?

Or is that --from another address

[warped holly]

This can also be the culprit yes

But aint that messed up if someone could freeze a whole channel just by being on an incorrect version?

I'm creating a new channel, will try all sorts of combinations again.

[minor solstice]

sure but wouldn't be the most surprising thing that could happen

we were all using at least three different versions I think

[quaint arch]

yes, the same key the relayer is using

[warped holly]

I changed my version to long-memo branch, my hermes was dirty...

I'm not getting the issue anymore

Though, ain't that a big problem? If my hermes caused this, then just one bad actor can destroy a channel.

If it's related to hermes version

(btw channel-6735 can be used for osmosis)

[quaint arch]

yeah, don't think it's supposed to work that way
do you know what was different about your hermes version?

[warped holly]

I unfortunately only have the commit hash but it said -dirty so I probably did something in there.

Though I find it weird, others had the same problem in the past, so it may have been related to that commit

hermes 1.7.4+dc498bbf-dirty

I'm btw not sure if it's related to this, but atm I'm not getting issues with the channel freezing

Should I again switch to that one and do the same tx and see if it freezes?

Wait let me try finding out what was different with that one. I can't really figure it out precisely since I only have the binary.

[minor solstice]

commit hash you're using now?

or app version I guess

[warped holly]

Ah I remember, so back then I grabbed branch: yuji/1.7.4-namada-shielded
and changed the part in the code where the memo limit was upped.

[minor solstice]

sus haha

[warped holly]

This is all speculation still though

Let me go back to that version and try the tx again

No wait, my client froze agan

again*

Okay this is confusing now lol.

I'm on hermes 1.7.4+1ede457f long memo and it froze ._.

recreating channels :)...! Gotta love bugtesting huh, Spork.

[minor solstice]

think that's the one I'm on too if memory serves

[warped holly]

Redeployed my site btw, think it will auto show the channel I just made now

Will be short-lived tho lol.

Ah wait it's still building.

[quaint arch]

Are you freezing using the cli or your app? Could it be something your app is doing (no offense)

[warped holly]

Hahahaha

None taken

Let me do it via CLI

And this issue happened before the app was around, but yes let me keep the env we test in consistent.

[warped holly]

This made me laugh

If it is, then I definitely need to be labeled the destroyer

[quaint arch]

Solution is clear, we just need to keep zen away from our stuff and everything will be fine 😂

There's only one relayer running on these channels, right?

[warped holly]

Yes, don't think anyone else joined

Okay I did the same CLI command 5 times

t froze

It

AAAAAAAAAHH

Perhaps there's something here

osmosisd tx ibc-transfer transfer transfer channel-6736 znam1qqtnyzu97zknt3sff74aj8fnxxx6lr5gke2vqp8w8pmxz7dp25rt7jsmq46dn2g2t0al2ps2paryv 500000uosmo --from osmrel --chain-id osmo-test-5 --node https://osmosis-testnet-rpc.polkachu.com:443 --fees 2000uosmo

[quaint arch]

Do you have misbehaviour set to true or false in your config?

[warped holly]

Set to false

[mode.clients]
enabled = false
refresh = false
misbehaviour = false

Got a cronjob open with update client commands

[quaint arch]

Oh and can you please set log level to debug if you haven't already?

[warped holly]

Oh, is it too late to do this?

[quaint arch]

For next time
You switched hermes versions, is it possible the cronjob commands are still pointing to the dirty version?

[warped holly]

(At times I'm a noob, forgive me)

[quaint arch]

No worries, me too, this is advice from yuji

[warped holly]

This is how my .sh file looks like:

#!/bin/bash
/root/.hermes/bin/hermes --config ~/.hermes/config.toml update client --client 07-tendermint-3705 --host-chain "shielded-expedition.88f17d1d14"
/root/.hermes/bin/hermes  --config ~/.hermes/config.toml update client --client 07-tendermint-3706 --host-chain "shielded-expedition.88f17d1d14"
/root/.hermes/bin/hermes  --config ~/.hermes/config.toml update client --client 07-tendermint-3737 --host-chain "shielded-expedition.88f17d1d14"
/root/.hermes/bin/hermes  --config ~/.hermes/config.toml update client --client 07-tendermint-3587 --host-chain "theta-testnet-001"
/root/.hermes/bin/hermes --config ~/.hermes/config.toml update client --client 07-tendermint-3196 --host-chain "osmo-test-5"

cronjob:
0 */2 * * * /bin/bash /root/update-hermes.sh >> /root/update-hermes.txt 2>&1

my service file also points to:

ExecStart=/root/.hermes/bin/hermes --config /root/.hermes/config.toml start

I even daemon-reloaded btw.

/root/.hermes/bin/hermes version
2024-04-04T22:15:56.140305Z  INFO ThreadId(01) running Hermes v1.7.4+1ede457f
hermes 1.7.4+1ede457f

So in conclusion everything points to 1.7.4 long memo branch.

Unless I need to do something bout that cronjob? Does it have caching or something?

Hmm, perhaps I should make a change to the crontab -e, save and then change it back, save again.

Don't think it caches the scripts it's supposed to run? That would actually be annoying...

Uhm how would I set debug on btw? Do I do this in the service file?

[cunning finch]

At the beginning of hermes config file [global] log_level = 'info'

[warped holly]

facepalm

Thanks 😂.

Alright, step by step, how fitting: channel-1234 just got made.

Okay everything sorted, now let's spam till it freezes or not.

Hmm okay frozen in 3 min roughly.

Here's the log from start to end

Again I kept doing this to keep it consistent:

osmosisd tx ibc-transfer transfer transfer channel-6737 znam1qqtnyzu97zknt3sff74aj8fnxxx6lr5gke2vqp8w8pmxz7dp25rt7jsmq46dn2g2t0al2ps2paryv 500000uosmo --from osmrel --chain-id osmo-test-5 --node https://osmosis-testnet-rpc.polkachu.com:443 --fees 2000uosmo

Perhaps it might be related to refunding tokens?

The log's okay now right @quaint arch?

Ah btw, this might be useful as well. All my 9 tx's.

2024-04-04T22:31:34.397114Z DEBUG ThreadId(32) worker.batch{chain=osmo-test-5}:supervisor.handle_batch{chain=osmo-test-5}:supervisor.process_batch{chain=osmo-test-5}:worker.packet.cmd{src_chain=osmo-test-5 src_port=transfer src_channel=channel-6737 dst_chain=shielded-expedition.88f17d1d14}:update_schedule{batch.tracking_id=fa47b0c4 batch.height=5-6475416}:schedule{odata=fa47b0c4 ->Destination @5-6475416; len=1}: connection delay need not be taken into account: client update message will be prepended later
2024-04-04T22:31:34.482246Z DEBUG ThreadId(32) worker.batch{chain=osmo-test-5}:supervisor.handle_batch{chain=osmo-test-5}:supervisor.process_batch{chain=osmo-test-5}:worker.packet.cmd{src_chain=osmo-test-5 src_port=transfer src_channel=channel-6737 dst_chain=shielded-expedition.88f17d1d14}:relay{odata=fa47b0c4 ->Destination @5-6475416; len=1}: retrying retry.current=1 retry.max=5
2024-04-04T22:31:34.482264Z DEBUG ThreadId(32) worker.batch{chain=osmo-test-5}:supervisor.handle_batch{chain=osmo-test-5}:supervisor.process_batch{chain=osmo-test-5}:worker.packet.cmd{src_chain=osmo-test-5 src_port=transfer src_channel=channel-6737 dst_chain=shielded-expedition.88f17d1d14}:relay{odata=fa47b0c4 ->Destination @5-6475416; len=1}: prepending Destination client update at height 5-6475417

These messages look suspicious to me. Like some queue is building up.

But yeah, I will have to sleep now, lemme get some valid channels up and running again on my site though.

[quaint arch]

Well it's Greek to me but I'll forward it to yuji to see if it makes sense to them... tysm zen, you've been super helpful as always

[warped holly]

THANK YOU for saying that, I was feeling the exact same way when those debug messages appeared XD!

https://tenor.com/view/krule-john-c-reilly-what-face-reaction-gif-16881707

And you're very welcome man!

Thank you for testing it out, it's just weird that I managed to instantly freeze my own channel back when I tried to recreate the problem. It made me think it was definitely caused by this transaction, but it isn't 'exactly' the case? I dunno. It's weird.

Yuji and others will speak some ancient greek and do some magic

Okay, redeploying now. Once Osmosis says channel-6738 on the site, it's updated.

[warped holly]

Nah btw, this I also see in normal transactions.

[minor solstice]

same

[unique aspen]

Just received this with an unshielded IBC transaction Osmosis -> Namada through your app:
transfer/channel-1235/uosmo: 1000000

[warped holly]

I wonder how one can rewrite this to a tnam address

[unique aspen]

Just received this with a shielded IBC transaction Osmosis -> Namada through your app:
transfer/channel-1235/uosmo : 1000000

[warped holly]

For unshielded, if you know the tnam address instead of the 'transfer/channel-1235/uosmo' denom, then you could add it. Only for unshielded though.

Still need to figure out if there's a quick way to do this

Best case would be that it auto fetches all denoms.

[unique aspen]

Not sure I follow, this is what the CLI returns when querying balance.

Just to confirm that the transaction went through.

[warped holly]

Yeah, transfer/channel-1235/uosmo actually has a tnam... address

Just like naan has a tnam...zee address

[unique aspen]

Oh

[warped holly]

It ends with zee I believe? I'm not sure.

[unique aspen]

zcee

[warped holly]

Ah yes, that one

[unique aspen]

I think that is a low priority UX thing

[warped holly]

I could be really gay and just let balances get queried via CLI 😂.

[unique aspen]

I think the MASP algorithm has to be improved so it can run in a browser client

Starting with scanning backwards

[warped holly]

Yeah, really hope to see that, will definitely implement it.

[unique aspen]

That is what ZCash light wallets do

[minor solstice]

you have a week to make protocol improvements:)

[warped holly]

Interesting, I didn't know! I did write an issue on Github about starting the scan from the end of the chain.

But reading it back I just notice how ignorant I am to it all.

[unique aspen]

https://forum.zcashcommunity.com/t/zecwallets-blazesync-sync-entire-chain-in-60s/39447

[warped holly]

That's genius.

[unique aspen]

There is another algorithm that I wanted to implement but I fear too short before the end: WarpSync

[unique aspen]

I already have points for protocol improvements so now I need to focus on security 😉

[warped holly]

Good luck Rigorous!

[unique aspen]

I will need to summon my inner stampede

But to be honest I think implementing a fast sync would be more satisfying

[warped holly]

I realized this really late, but the outro's of that anime had gems for life lessons

The English version also did a great job at this

[unique aspen]

Oh, I never realized

[warped holly]

https://www.youtube.com/watch?v=X-tuU43QFyo

[unique aspen]

I tried reading the manga too, the art strains my eyes trying to figure out what happens

Blame kind of the same, but I guess that is the charm

[warped holly]

Hmm I never had this, but I may get it now 😂, I'm older now compared to back then.

[unique aspen]

Wow, I just always skipped to the next episode!

[warped holly]

Me too lol

[unique aspen]

I never watch the spoilers

[warped holly]

A friend of mine put this on once while he was smoking his weed and nodded like 'uhuh' 'uhuh', you didn't know this huh?

[unique aspen]

I wonder if BlazeSync can be translated to Namada though, because ZCash has a UXTO model while Namada has an account-based model

[warped holly]

I'll have to lookup UXTO, not so well versed with these terms. Will check it out more tomorrow!

[quaint arch]

the team tracked down the cause to an issue in ibc-rs: https://github.com/cosmos/ibc-rs/issues/1080
so it should be resolved when ibc-rs is upgraded in Namada. (the ability to unfreeze clients should be possible in the newer version also)

hermes retries transactions when they're rejected by the destination chain (eg a znam address with no memo) and before each retry, a client update is requested. so it could be related to many client updates being requested... the consensus state of the latest client gets overwritten by an earlier one due to the bug, and on the next update the incongruency is detected as misbehaviour and the client is frozen.

for now, working with what we have, they suggest a partial workaround: set trusted_node true in the hermes config for the counterparty chain... this setting ensures only one update message is submitted per request. it will work for a single hermes, but does not perfectly avoid the issue when using multiple hermes for a single channel

[warped holly]

Nice! Interesting. Glad it can be fixed in Namada itself btw eventually!

Just creates a bit of a problem here with the channels this temp workaround.

Means that if multiple relay the channel and this edge case gets done and a relayer without the workaround keeps relaying it, the chance is there the client could still freeze?

[cunning finch]

Well, my suspicions were right in regard to the ibc-rs 🤔, great!

[warped holly]

The counterparty chain being osmo/theta/any other chain btw?

[cunning finch]

Glad we stressed it out before mainnet, because we def need to have canonical channels with multiple operators supporting them

[warped holly]

For sure man, really important. I got PTSD with expiring clients in another chain.

[cunning finch]

Hahaha we felt that way on Evmos a long time ago, almost all highly active ibc clients expired after a halt

But using ibc-go, ofc, ibc-rs has not been battle tested enough like the 1st one I guess

[warped holly]

Damn, mine may have been related XD. The chain was a fork of evmos.

[cunning finch]

They really have a cool tech, it's really sad to see that some of its previous founders harmed the project like that. With Dymension launch, they have eaten almost all the market share of their tech, along with Kava, CRO, Canto, etc.

[unique aspen]

I got rugged on Evmos

[warped holly]

Hm, they used to be Ethermint right?

[unique aspen]

Nomad exploit

[cunning finch]

Yes, that's right

They took measures to protect their work with licenses after all the drama

[cunning finch]

That was one of the things that slowed down and basically ruined its traction and trustness

[warped holly]

That's a shame

[warped holly]

Set my settings for osmo-test-5 and theta-testnet-001 to trusted_node = true for now.

If we still going to do the theta relaying over here, perhaps my open channel could be used. (channel-4157 - channel-1217).

[warped holly]

#shielded-expedition message

[unique aspen]

I transferred back and forth between Cosmos and Nama with my public key as memo. The transactions show up on Mintscan, but they do not register on the extended NEBB.

[warped holly]

via CLI?

[unique aspen]

Yes

Actually it would not register with your app either, since Namada as source does not work, right?

[warped holly]

I think the relayer who transacts the package gets registered

You used my channel?

[unique aspen]

Yes

[warped holly]

Can you see if it gets on my profile

[unique aspen]

Good idea

[warped holly]

(only if I'm the only one relaying the channel tho)

Something weird going on with my shielded sync though...it keeps fetching like a minimum of 500 blocks.

[warped holly]

This does work btw but only unshielded

Oh oops, I've also now done 2 unshielded transactions to double-check ._.

Dunno if you're able to distinguish. Apologies lol. Won't do anything now.

[unique aspen]

Could not find it

Are you using memo_overwrite in Hermes config.toml?

[warped holly]

No memo_prefix, which does seem to register on extended nebb

[molten drum]

looking at the nebb from Kintsugi, yes it is, coz I have ibc txs done today without doing actually them on my own

[warped holly]

Or has Kintsugi switched over to only strictly 'memo_overwrite'?

Think this memo_overwrite is a new field in 1.8 ish version right?

[unique aspen]

Thanks. memo_overwrite throws an error even though it is described in the documentation

[molten drum]

sitting on long-memo versions

[warped holly]

Yeah coincidentally I was reading about that one for another chain, but think I saw that one being introduced in hermes v1.8.2

So for now we still just use memo_prefix

And yes like LiveRaveN said, long-memo branch

You'll set up own channels?

[unique aspen]

I am trying something else

[unique aspen]

invalid length 66, expected a string length of at most 50 I installed Heliax Hermes binary, at least I thought

[warped holly]

Yes so you need the branch

[molten drum]

you need long-memo branch

[unique aspen]

Dang

[warped holly]

Hol up, what's it called exactly

[unique aspen]

That is not obvious at all

[warped holly]

git status
On branch 1.7.4-namada-long-memo
Your branch is up to date with 'origin/1.7.4-namada-long-memo'.

hahahaha this all is not supposed to be obvious 👁👁

[molten drum]

https://github.com/heliaxdev/hermes/tree/1.7.4-namada-long-memo

[unique aspen]

Yeah, found it.

"One can download the latest binary release"

Yeah, no

[warped holly]

Hahahaha, yeah

Might save you time:

cargo build --release --bin hermes

If you're a dummy like me forgetting simple commands

It will then be in the target/release folder.

[unique aspen]

You are a psychic!

I actually already got the ROIDs for the Operating IBC task, but I want to play around.

[warped holly]

Yes, yes, dance with it

[unique aspen]

Because it bugs me out that the transactions do not register

[warped holly]

I likely need to create a new shielded address

Yeah I feel ya

Had the same

[unique aspen]

Curiosity is a powerful driver

[warped holly]

Though actually unfair if that's the case, then a relayer could farm points made by others

Or perhaps the treat for providing such service 👿

[unique aspen]

I thought that was your plan 😂

[warped holly]

My great masta plan unraveled by Rigorousss

[unique aspen]

I knew there was a reason why you made that website so nice

[warped holly]

ahahahah, you want some candy? Come hereee, comee heereeeeee

Thank you though ❤

Oh that's a different heart. Ah well, all hearts matter.

[unique aspen]

Actually I have a similar plan for a different purpose, but not sure I will be able to pull it off.

[warped holly]

Keep at itt, there's a ton you can still do

[unique aspen]

I know, I have a huge to-do list, but not enough time

[warped holly]

I just uploaded a fix so that you can't type a wrongly formatted address in the address field on the site

I have to create a new shielded address btw, just a moment. Let me see if I can.

Cause of this weird bug now:
Asset threshold of selected conversion for asset type bfe4922fe3da46887eaeedfc37dfd286c11f961cc37dbe285dd08dc241539020 is 0, this is a bug, please report it.

But you already saw it, saw your response on github!

The log made me laugh.

This is a bug, please report it. - I imagined a robot saying this. AI being self-conscious.

[unique aspen]

What happens when you do balance --owner YourSpendingKey?

[warped holly]

Oh that's interesting, think last I tried this it asked for my viewing key

[unique aspen]

I never had that

[warped holly]

Like literally gave a prompt to fill in my viewing key

It used to say something with not consumed or something right?

[unique aspen]

That is super weird, I never had to do anything with a Viewing Key to do the point I wonder what its purpose is

[unique aspen]

I think that is why you get the divide by 0 and I cannot replicate it

[warped holly]

I dunno man, this divide by 0 just appeared outta nowhere

[unique aspen]

You have not shielded-synced that shielded pair?

Let see if I can replicate it with a new pair

[warped holly]

It used to just give my balance, like 4 hours ago it worked

[unique aspen]

Maybe shielded.dat is corrupt, you can delete it but have to resync

[warped holly]

What I do is perhaps really odd tho, I shielded sync without specifying viewing key, then right afterwards I do it with a --viewing-key flag

Yeah though odd if it got corrupted in two servers

[unique aspen]

Ah

[warped holly]

I can find balance for another viewing key tho

Try querying balance of my viewing key

namada client balance --owner zvknam1qynmxyf5qqqqqq983j0nwgsc8zf0jhglrq3x49y05vz3vdhlwcw9taldce6apv73e3g8mpn57afyqdnkh8dpmp7v962l24m69f8aynp7sa02jsf93ftaehmlwtj9ak7hply4c93862sahqtprluapxda5p5x3gx4k6wzqedjuu8yhtyk5paxedw6gpfhfsatr96jeqh9tjh2258xtrnyz0f2pddhsr3ms6lwkywaphy687uvcnwc06jd32zu375437xe6p2yx6ets6c07r988

[unique aspen]

New shielded pair = context should contain viewing key

[warped holly]

I wonder about something btw, you may perhaps know.

What's the difference between calling shielded sync and calling it with --viewing-keys and specifying one? And what happens if one does these two back to back.

My assumption was that shielded sync without --viewing-keys would look at all viewing keys you got in your wallet.toml and sync those.

[unique aspen]

That seems a reasonable assumption

[warped holly]

So I tested a viewing key I didn't import (namely the one in the extension cause I still can't figure out how to import that one)

[unique aspen]

After doing a little scanning:
balance MyViewingKey
Last committed epoch: 79
No shielded balance found for given key

[warped holly]

Ah so you do have to implicitly shielded-sync that one by specifying it in --viewing-keys

Don't though lol.

[unique aspen]

Why not?

[warped holly]

You don't want your shielded sync to crap out now.

Or get corrupt or whatnot

Oh

You're talking bout your own?

[unique aspen]

It does not matter, the shielded actions besides unshielded->shielded are not working for me anyway.

[warped holly]

Thought you tried my VK, nvm hehe.

[unique aspen]

Oh

[warped holly]

This is your new shielded

Uhm what's the payment addr? can I try sending to you?

Or wait, let me do my own stuff first.

[unique aspen]

I have not created a payment address yet. Only created a new spending-key viewing-key pair, queried balance, synced a few hundred blocks, queried again.

[warped holly]

You specified the from-height?

[unique aspen]

Yes

[warped holly]

Like one really close I assume?

[unique aspen]

--from-height 310600

[warped holly]

Did this cause trouble with your other shielded addresses?

(if you have any)

[unique aspen]

I imagine

I mean I just skipped over a couple thousand blocks

[warped holly]

Yeah, ah well as long as you have one working addr

[unique aspen]

About to find out what the consequences are

[warped holly]

Hehehe, anything to bugtest this

[unique aspen]

Weeeee

When you're a kid and you want to go weeeee

[warped holly]

When you're an adult it's when you're on a rooftop and just want to go weeee

😂 hahahaha, too morbid.

[unique aspen]

https://www.youtube.com/watch?v=op0Kk2aCdJY

[warped holly]

Hahhahahaha weleleleel weeeee

Whenever I hear that I remember giving my lil niece math lessons via video calls

She kept doing this with all her other lil brothers and sisters

She picked up her phone and she just said that with her face all up in the camera

Oh shit someone just got knocked out by a tree

🤣 that one caught me off-guard

[unique aspen]

Straight from Looney Tunes. What are the odds?

[warped holly]

This world can't be taken serious at times

[unique aspen]

transfer --source MySpendingKey --amount 1 --token naan
The balance of the source tnam1pcqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqzmefah is lower than the amount to be transferred.

balance --owner MySpendingKey
naan : 42

[warped holly]

I spammed shielded sync after I got that

Though VP will probably reject anyways

[unique aspen]

You are right.

After syncing:
naan : 1042

Transfer:
The balance of the source tnam1pcqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqzmefah is lower than the amount to be transferred.

Can we fix the MASP storage or is that part of the buggy behavior that is not going to be fixed this SE?

Maybe resync from the fork

[warped holly]

What type of transfer you're trying to do?

[unique aspen]

spending-key to transparent

[warped holly]

Ah so unshielding

Don't think that works

[unique aspen]

Oh yeah, that was broken 🤦

[warped holly]

Yeah, also shielded -> shielded didn't work.

[unique aspen]

Maybe call it a day and get some sleep

[warped holly]

Hehe, I'll also do that after I implemented copy to clipboard buttons, wanna add for viewing key as well.

Since the extension doesn't show this

[unique aspen]

I read the extension deliberately lacks features.

[warped holly]

deliberately huh, hm.

More and more think it would have been better if I sticked to CLI hehe.

But nah, I liked to have made these applications. Can be used eventually, just need to keep up and keep adding on to it.

[unique aspen]

Yes, somewhere in specs > browser-extension

```The architecture of the extension and extension API somewhat resembles that of `@keplr-wallet/extension

There are a few key differences in functionality however. Our extension will only handle the creation of a seed and derived keypairs, as well as signing transactions. It will not be responsible for allowing the user to send transfer transactions, but should be responsible for submitting account initialization transactions```

[warped holly]

Ah I see

[unique aspen]

The web app has to do that

[warped holly]

Though I can let it do a transaction

Unshielded that is.

Hmm, things may have changed last this has been updated

Ah I perhaps understand what they mean

So the extension itself, just opening it won't offer much functionality

In Keplr you could do transactions, do IBC transfers, all that from within the screen.

Ah okay yeah, got it.

[unique aspen]

Yes, but I wonder if that is a design choice due to time constraints, technical limitations or expected performance issues.

Or just out of principle.

[warped holly]

Hmm, asking a question thurrrr

I do not know

Could be a 'for now we keep it simple'

[unique aspen]

Say you want to make your own Namada wallet that behaves more fully featured like Keplr, would the organization try to talk you out of that?

[warped holly]

Think you'll get thrown in a bonfire

[unique aspen]

Haha

Ouch

[warped holly]

😂

[unique aspen]

"YOU DIED"

[warped holly]

😂, why do I think of that video of that dude singing.

"Before I was born, my motha died"

[unique aspen]

No I was thinking of emotional damage

[warped holly]

hahaha 😂

EEEEEEEMOOOOOOOTIONALL DAAAAAMAAGEEEE

wullullululuppp weeeeeee

[unique aspen]

https://www.youtube.com/watch?v=miD_TWmdGIY

[warped holly]

Hahahahahahahah

[unique aspen]

Why why why why

[warped holly]

My god, finally got those copy to clipboard buttons implemented. I hate styling + making everything responsive/mobile-friendly arghh...hope it's not too cluttered now tho. I actually just wanted to add a button to copy viewing key to clipboard but ended up adding all addresses.

5AM 💤...peace.

[cunning finch]

Shielded set rewards have been activated in theory at start of epoch 80. Unluckily I messed up keys and I can't find the one I used to receive the uatom from the frozen channel. Is there anyone here that managed to get uatom transfer/channel-1142/uatom? If so, please make a shielding transfer to test out the feature.

Hm, it looks like the two selected assets to receive shielded set rewards aren't appearing while running:

The following tokens may earn MASP rewards:```

[cunning finch]

@quaint arch

Ah wait, grace epoch is 78, so they should have been activated at start of epoch 79 in theory 🧐

[paper pulsar]

the channel is still live?

[median island]

Do i need uatom too or just NAAN can give the rewards

[cunning finch]

It's frozen - the issue is related to ibc-rs. Follow the above conversation to further reference

[cunning finch]

NAAN as well

[median island]

means both tokens need to have in same shielded add

[paper pulsar]

ok

[cunning finch]

One of them could serve as well. Do you see any increase in the NAAN shielded balances in case of having shielded NAAN or uatom (the one that belongs to the frozen channel!)?

The rewards are in form of NAAN

In theory 🙄

Perhaps it had to do with proposal wasm code?

[warped holly]

I did sent that specific uatom (0.1 atom) to a shielded address, but I did it before this started I think

[median island]

when im trying to check my shielded balance i just getting this

namadac balance --owner zvknam1qw0ufd22qqqqpqqn5pp03dkflh6w7m3rh8ut6mk9jefw2fy86v0hnphljl0n82t7x84hw02tumqeeqsa688rzw63f0cel739d6g2gy9g0gfrl2spcepv6yzyp0sllrajefj56nse0p60vdqut07m9qcvrj2gfza8pfrz3fxx4c6yd33swl3aqntn7d7x4jn7phth6mx03gfjkywhu07yxyex03wsfvmfn76356lc3pe594kku6x78apklmf5a0dwust9usx8fweng4s08hwmu
Last committed epoch: 80
converting current asset type to latest asset type...
converting current asset type to latest asset type...
converting current asset type to latest asset type...
converting current asset type to latest asset type...
converting current asset type to latest asset type...
converting current asset type to latest asset type...
converting current asset type to latest asset type...
converting current asset type to latest asset type...

and its not genrating any result

[warped holly]

This seems an issue with not being shielded synced I believe

Not sure tho

[cunning finch]

There's no increase in the NAAN shielded balance?

[median island]

just did 1 min before

[warped holly]

Let me check this

[cunning finch]

Guys. Shielded to shielded transfers and unshielding transfers are broke for you as well?

[median island]

didnt try yet

[carmine trail]

I can't get it to work now. It did before hardfork

[cunning finch]

The shielded sync thing seems to work but only to fetch balances and for shielding transfers

[cunning finch]

For me too...

[warped holly]

Here's a current screenshot, I'll wait a bit and see if anything changes.

Doesn't look like my NAAN changed, think I already had this yesterday

It's auto shielded syncing every couple mins on my server.

[cunning finch]

Let's pray is just a display bug, but I'm afraid that code hasn't beene executed properly or something was missing/wrong

[warped holly]

How long should I wait? You know how often these rewards come?

[warped holly]

Yeah I read above you couldn't see the pair in the shielded reward set hm?