Everything installs properly in a mock chroot, which is probably all that I can test. Installing into an unconfigured chroot does result in an error from the %posttrans scriptlet of pesign that involves siguldry_pkcs11:
>>> Finished %posttrans scriptlet: pesign-0:116-7.fc42.x86_64
>>> Scriptlet output:
>>> INFO siguldry_pkcs11: return=0
>>> INFO C_Initialize{pInitArgs=0x5619b00be620}: siguldry_pkcs11: new
>>> INFO C_Initialize{pInitArgs=0x5619b00be620}:new{socket_path="/run/siguldry-client-proxy/siguldry-client-proxy.socket"}: siguldry::client::proxy: new
>>> INFO C_Initialize{pInitArgs=0x5619b00be620}:new{socket_path="/run/siguldry-client-proxy/siguldry-client-proxy.socket"}: siguldry::client::proxy: close
>>> INFO C_Initialize{pInitArgs=0x5619b00be620}:list_keys: siguldry::client::proxy: new
>>> INFO C_Initialize{pInitArgs=0x5619b00be620}:list_keys: siguldry::client::proxy: close
>>> ERROR C_Initialize{pInitArgs=0x5619b00be620}: siguldry_pkcs11: Failed to retrieve a list of keys from the server via the proxy error=channel closed
>>> INFO C_Initialize{pInitArgs=0x5619b00be620}: siguldry_pkcs11: return=6
>>> INFO C_Initialize{pInitArgs=0x5619b00be620}: siguldry_pkcs11: close
I’m not convinced this is a serious problem, and the RPM transaction does complete successfully.
This update can be pushed to stable now if the maintainer wishes
Thanks for testing. You're right, that's expected (although not ideal behavior). I think I should probably reduce the default log verbosity, or at least not emit to stderr by default, and it should report no slots rather than failing to initialize if there's no proxy available.
This update's test gating status has been changed to 'waiting'.
This update's test gating status has been changed to 'ignored'.
This update has been submitted for testing by bodhi.
This update has been pushed to testing.
Everything installs properly in a mock chroot, which is probably all that I can test. Installing into an unconfigured chroot does result in an error from the
%posttransscriptlet ofpesignthat involvessiguldry_pkcs11:I’m not convinced this is a serious problem, and the RPM transaction does complete successfully.
This update can be pushed to stable now if the maintainer wishes
Thanks for testing. You're right, that's expected (although not ideal behavior). I think I should probably reduce the default log verbosity, or at least not emit to stderr by default, and it should report no slots rather than failing to initialize if there's no proxy available.
I've filed https://github.com/fedora-infra/siguldry/issues/172 and https://github.com/fedora-infra/siguldry/issues/173
This update has been submitted for stable by jcline.
There is an ongoing freeze; this will be pushed to stable after the freeze is over.
This update has been pushed to stable.