But only if you can figure out in a transaction or locked sytate whether the card needs a verify. Otherwise we have a race between changing the PIN and verifying a PIN.

We need to see whether we can use the gpgconf.ctl feature also for Windows here. Registry entries for gpd and vsd also require a change.

It could have been discussed whether this makes sense. However, we can't change it anymore because it would change the behaviour. Consider a cron job which looks into a directory with keyids and imports them from a keyserver. It is totally fine if the script returns success if no keys are available.

Your above excerpt for the log is not a bounce. Can you please give me an example from a rejected bounce? Noet that BATV is also in use.

That is on purpose. Please add an SPF record to your site. If there is really really a problem for you with that, write me off tracker.

See T7034

Sorry, this is not a help line but a bug tracker. If you lost or forgot your password you are screwed up.

I try it. See T7026

See also rG40b85d8e8cecadf35e51e84b30de4fac820d714b for gnupg 2.4.

Sorry, for not following up earlier. Can you please do me a favor and run the last tests again, this time adding -v and --debug 1 to the invocation? Feel free to forward the output to my private address is that is easier (wk at gnupg.org).

We migrated to another box and it might be the case that we planned to also support https. I need to see whetehr I can find notes in the etckeeper.

See also: https://gnupg.org/blog/20240125-smartcard-backup-key.html

How to test:

Thanks Ingo.

In 2.4 we have rG1383aa475 which does

Although I don't think this is the case here one should be aware that tests mail fail due to global configuration of GnuPG (/etc/gnupg/*.conf). There is no easy way so solve this except for running a per-test local installation of GnuPG using the gpgconf.ctl feature.

Arghh, a GPGME_DEBUG=3 which shows basic I/O preparation does not exhibit the bug.

Fixing gpg is easy but there is some bug lingering in gpgme which might be a recent regression. An strace shows

Those options where originally intended for debugging but your suggestion makes sense. I also add this to most other tools.

At the failed attempt I notice a DEVINFO_STATUS removal soon after the SWITCHCARD. This is related to the reader status file and triggered by ccid_slot_status. --debug ipc.app,reader might be a better selection of debug flags here.