MantisBT - Issues

0011713: GNS takes ages to resolv

Thu, 06 Aug 2026 11:17:51 +0200

during bootstrap issue investigation https://bugs.gnunet.org/view.php?id=11711
@schanzen did create a TXT dns record in `info.higepi.gnunet.gns.alt' that expire every 30 min

when trying to get this records : $ gnunet-gns -t TXT -u info.higepi.gnunet.gns.alt -L debug

2026-08-05T21:25:27.527848+0200 gnunet-gns-35901 DEBUG Loading block plugin `libgnunet_plugin_gnsrecord_gns'
>>> Looking for `TXT' records under `info.higepi.gnunet.gns.alt'
...
2026-08-06T02:02:18.690705+0200 gnsrecord-35901 DEBUG Determined expiration time for block with 2 records to be Thu Aug 06 02:32:04 2026
<<< 2 record(s) found:

TXT: `Thu Aug 06 02:02:04 2026: Connected (R) M0J8 (timeout in 248 s)'
TXT: `Thu Aug 06 02:02:04 2026: Awaiting initiation (R) E2R1 (timeout in 0 ms)'

Record set expires in 29 m.

5h...

0011712: gnunet-identity never return and endless loop

Thu, 06 Aug 2026 11:16:17 +0200

tried
# gnunet-identity -C Jllx -L Debug
(see attached log)

it never create the identity
gnunet services are up

# gnunet-arm -I
Services (excluding stopped services):
(started: 19 / stopped: 16)
cadet (binary='gnunet-service-cadet', status=started)
core (binary='gnunet-service-core', status=started)
dht (binary='gnunet-service-dht', status=started)
gns (binary='gnunet-service-gns', status=started)
hostlist (binary='gnunet-daemon-hostlist', status=started)
namecache (binary='gnunet-service-namecache', status=started)
nat (binary='gnunet-service-nat', status=started)
nse (binary='gnunet-service-nse', status=started)
peerstore (binary='gnunet-service-peerstore', status=started)
pils (binary='gnunet-service-pils', status=started)
rest (binary='gnunet-rest-server', status=started)
revocation (binary='gnunet-service-revocation', status=started)
rps (binary='gnunet-service-rps', status=started)
setu (binary='gnunet-service-setu', status=started)
statistics (binary='gnunet-service-statistics', status=started)
topology (binary='gnunet-daemon-topology', status=started)
transport (binary='gnunet-service-transport', status=started)
communicator-tcp (binary='gnunet-communicator-tcp', status=started)
communicator-udp (binary='gnunet-communicator-udp', status=started)

0011714: gnunet-arm -I never returns if no service are running

Thu, 06 Aug 2026 11:11:25 +0200

when trying to know if the services are running I do a "gnunet-arm -I" which never returns if there is no service running (have to kill the process)

0011711: can't bootstrap (again :( )

Thu, 06 Aug 2026 10:39:43 +0200

linked to https://bugs.gnunet.org/view.php?id=11629
so I tried today again and now it don't connect...

after gnunet-arm -s I monitor connections states using
# gnunet-transport -a
PEER ROUTE VIA CORE-WIN STALLED PENDING RTT
DXYV direct tcp,udp 4 0 0 1857 ms

so it seems to be connected to DXYV
and
# gnunet-pils -1
82D13WPCJXQ4ZPYKS3XNAEKSSYCM0KG7Z1GKNTK99A391CXJ2MD0
(but not sure it's relevant, since it also output the same result later)

# gnunet-core -s
Mon Aug 03 00:14:29 2026: Awaiting initiation (R) DXYV (timeout in 0 ms)
so it's waiting initiation

I do this several time and after some time it ends like this :
# gnunet-transport -a
No virtual links.
# gnunet-core -s
nothing displayed
# gnunet-pils -1
82D13WPCJXQ4ZPYKS3XNAEKSSYCM0KG7Z1GKNTK99A391CXJ2MD0
still display something

# gnunet-arm -I
Services (excluding stopped services):
(started: 19 / stopped: 16)
cadet (binary='gnunet-service-cadet', status=started)
core (binary='gnunet-service-core', status=started)
dht (binary='gnunet-service-dht', status=started)
gns (binary='gnunet-service-gns', status=started)
hostlist (binary='gnunet-daemon-hostlist', status=started)
namecache (binary='gnunet-service-namecache', status=started)
nat (binary='gnunet-service-nat', status=started)
nse (binary='gnunet-service-nse', status=started)
peerstore (binary='gnunet-service-peerstore', status=started)
pils (binary='gnunet-service-pils', status=started)
rest (binary='gnunet-rest-server', status=started)
revocation (binary='gnunet-service-revocation', status=started)
rps (binary='gnunet-service-rps', status=started)
setu (binary='gnunet-service-setu', status=started)
statistics (binary='gnunet-service-statistics', status=started)
topology (binary='gnunet-daemon-topology', status=started)
transport (binary='gnunet-service-transport', status=started)
communicator-tcp (binary='gnunet-communicator-tcp', status=started)
communicator-udp (binary='gnunet-communicator-udp', status=started)

done it several time today and it always end like this, is RTT Round Trip Time ? if yes the value varies slightly during my polling from 1857ms to 1859ms so it don't seems to be a "ghost" message (previous message not updated and still displayed, as for gnunet-pils -1 output)

0011410: first purchase on blog experience [QC]

Thu, 06 Aug 2026 08:12:14 +0200

consider the following scenario, very simple from the first demo experience

1) to to demo.taler.net using your desktop, from the you see you can install the wallet (android or ios, same)
2) install the mobile version & withdraw demo money from the stating button / onboarding
3) go to the shop __ in your desktop __
4) pay the article with your mobile wallet

now you can read the article in your desktop and the wallet shows you the fulfillment url, all fine BUT

After clicking on the fulfillment URL on the recipe you experience the following

1) the wallet shows "you need to pay" for a second time (bad UX since you come from the wallet, the same you used to pay)
2) click the pay link, you go back to the wallet, completes without interaction (good!) BUT you are not redirected to the blog (bad UX, I'm coming from the blog then repurchase was triggered so the wallet should be fast and redirect back)
3) void
4) you still see the the article on desktop for on refresh it ask to pay, weird (bad UX it should poll and change automatically)

Nothing is broken but I think we have a big gap for improvement:

for (1) clicking the fulfillment URL from the wallet should always "steal" the session. Maybe the wallet can include the sessionID of the last purchase or something similar?

for (2) I think is an UI problem (maybe present in all wallets): when the wallet is triggered by a taler:// from external source then the UI should redirect automatically to the fulfillmentURL when done. Not when it was QR or triggering behavior.
I think that the repurchase detection could be faster.

for (3) well, the blog should poll the backend. This is more are details but seeing fast and responsive it work is very nice first experience. Maybe also tell the user that the session lost access to the order if this is intended. I'm not entirely sure that the mutual exclusive access is needed in the example, but if it is it can be better.

0009315: need tests for auditor KYC support [2d]

Wed, 05 Aug 2026 23:05:31 +0200

The src/auditor/test-kyc.sh is a starting point, but inadequate.

0011706: avoid latex dependency in TOPS postal auth

Wed, 05 Aug 2026 15:59:37 +0200

We already have typst installed, there's little reason to also require the very heavy-weight latex tooling. Furthermore, the letter generation has potential issues with escaping.

0011574: Token "Scope" section is incomplete: scopes are missing

Wed, 05 Aug 2026 15:13:57 +0200

18.4.4.1. Scopes does not contain all available token scopes

0011682: trying to add SEPA bank account in iOS wallet leads to internal core error

Wed, 05 Aug 2026 12:36:46 +0200

Steps:
* Start with empty wallet
* Go to Settings => Bank Accounts
* Tap "plus"
* Enter a label
* Tap "Account holder" field

Now an "internal core error" message is shown.

App version: nightly 1.6.4 (1.6.14).

0011631: Text overlap bug when paying

Wed, 05 Aug 2026 06:51:12 +0200

Text overlap issue seen in trying Taler on iOS Nightly (app v1.6.3; wallet core: v1.6.10); cf. at the bottom.

0011166: Coin Demominations are confused

Wed, 05 Aug 2026 01:26:06 +0200

```
# taler-exchange-offline upload < sig-response.json
# taler-exchange-offline upload < acct-response.json
# taler-exchange-offline upload < fee-response.json
# taler-exchange-offline upload < global-response.json
```

BUG: the last step gives an error:

> 2026-02-27T14:05:42.020965+0000 taler-exchange-offline-427284 WARNING External protocol violation detected at json_helper.c:101.
> 2026-02-27T14:05:42.021036+0000 taler-exchange-offline-427284 WARNING Expected currency `BOOKZ', but amount used currency `BKZ' in field `history_fee'
> 2026-02-27T14:05:42.021049+0000 taler-exchange-offline-427284 ERROR Invalid input to set wire fee: history_fee#0 at 0 (skipping)
```

0010878: Help user filling their payto

Wed, 05 Aug 2026 01:25:44 +0200

Cyclos UI never show the user ID that is necessary to fill the payto URI.
We need a new Taler API for this.

0006696: discuss better structure for LibEuFin docs

Wed, 05 Aug 2026 01:23:43 +0200

The LibEuFin docs are currently pretty chaotic. We should come up with a better structure.

I guess there are three different target audiences for the docs:
* Operators of LibEuFin (who want to use it with Taler)
* Developers who want to use the LibEuFin API
* Core Developers of LibEuFin ("internal stuff, documented in public")

0010463: KYC auth transfer not understood

Wed, 05 Aug 2026 01:23:08 +0200

User reported that they expected the KYC auth transfer they made to authorize a deposit into their bank account to result in a withdraw into the Taler wallet. We need to make it clear(er) in the UI that the KYC auth wire transfer is just for authentication of the bank account owner and that the money will NOT be withdrawn into the wallet.

0011588: manual withdrawing partially inform non-withdrawn part as fee, even with 0 fee configured in exchange

Wed, 05 Aug 2026 01:17:59 +0200

create a manual withdraw of X
send a wire transfer of X - Y
wallet will show Y as fee and wallet core won't look for new funds

some other issue will tackle the fact that wallet-core might keep looking for new funds for uncompleted manual-withdrawal
but the UI shows Y as fee because fee is calculated as effective - raw based on https://docs.taler.net/design-documents/041-wallet-balance-amount-definitions.html

we also see this happening on other transactions, for example when p2p transaction is aborted mobile shows instructed X and fee -X

=====
i propose change this https://docs.taler.net/design-documents/041-wallet-balance-amount-definitions.html
and remove the "raw" definition

raw: The raw amount always refers to the amount with fees applied. The exact interpretation of that depends on the transaction type.

change it for list of fee

fees: list of operation reference and fee amount

The main problem is the raw is based on assumption that are not always a satisfied and when this happen the problem raised in the UI.
Instead wallet-core should inform transactions with the list of fee and this list is mutable over the time.
We may have an initial calculation of expected fee (informed when the user prepare the transaction) and it allows to add additional based on (abort, refresh coins, etc..)
Old raw can be calculated adding-up or subtracting-from instructed amount.
Operation reference can be shown in the UI in case the user wants to inspect the reason of that number. (it can be a taskId or the name + id of the operation like refresh#1)

in the withdrawal case fee will be always 0 since all operation are free of charge (the list will have N elements with amount = 0)

0011651: wallet-core should tag currencies with scopeIdx

Wed, 05 Aug 2026 01:17:38 +0200

This allows us to concisely reference a scopeInfo and the corresponding currencySpec.

0011665: local storage sync

Wed, 05 Aug 2026 01:17:21 +0200

local storage is used in very few specific case, to save the withdrawal operation id or in merchant spa to save the current personalization. this allows save preference that survives reload

the impl that keep sync the updated items is onAnyUpdate which iterates over all keys on every update of local storage
this is bad for performance

but we dont have more than 1 or 2 keys in local storage, so this is not urgent by any mean.
The API is generic and can be used in places that this is a problem.

0006243: sandbox filtering on EBICS date range is missing in C53 requests

Wed, 05 Aug 2026 01:15:55 +0200

The sandbox is ignoring the date range from C53 requests. To be fixed.

0006962: CAMT reports need more structure to specify negative balances.

Wed, 05 Aug 2026 01:15:44 +0200

CAMT numbers cannot be negative, therefore whenever a bank account enters the debit status, the CAMT generation must include more structure to reflect it.

0007344: withdraw target unclear

Wed, 05 Aug 2026 01:15:31 +0200

One user was confused where the money would go when withdrawing money from the bank. The idea that it would go into the Wallet/Webextension in the browser (or on a Mobile phone) was completely alien. He would have liked to FIRST specify the destination before selecting an amount.

Idea: maybe do it more parallel to what we do in the wallet, where we can now specify "send money" and then pick "to wallet" or "to bank account" -- and then ask for the amount?

0006400: consider doing faster re-tries for some scheduled operations

Wed, 05 Aug 2026 01:14:58 +0200

When the daily bank statement download fails, should LibEuFin retry before the next scheduled execution?

For most tasks, a faster re-try doesn't make sense though.

0007506: Add helpers to generate test data.

Wed, 05 Aug 2026 01:14:46 +0200

database data used in tests should be generated by convenient helper functions, instead of
being entered by verbose blocks of bare database instructions copied in several parts.

0009857: SPA displays default currency, not product currency

Wed, 05 Aug 2026 01:14:33 +0200

https://backend.demo.taler.net/instances/sandbox/webui/#/inventory
for a product with currency EUR in the inventory page, we see EUR, but if we go to "Quick edit" or "Update", there is KUDOS. See screenshot below

0005962: find out details about EBICS case sensitivity

Wed, 05 Aug 2026 01:14:16 +0200

Apparently some EBICS clients treat data in a case-insensitive way and upper case even when they got data (e.g. transaction IDs) in lower case.

0006399: review and document date/time/timezone handling

Wed, 05 Aug 2026 01:14:04 +0200

Especially for EBICS order params.