Summary: | app-crypt/gnupg-2.1.11 fails tests | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | tka <tka> |
Component: | Current packages | Assignee: | Crypto team [DISABLED] <crypto+disabled> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | k_f, neal, toralf |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
URL: | https://bugs.gnupg.org/gnupg/issue2229 | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Bug Depends on: | |||
Bug Blocks: | 552936 | ||
Attachments: |
build.log
Logs of the failed tests info.txt app-crypt:gnupg-2.1.11:20160201-220950.log emerge-history.txt environment |
Description
tka
2016-01-30 14:19:44 UTC
Created attachment 424250 [details]
build.log
Created attachment 424252 [details]
Logs of the failed tests
The common error is "gpg: * failed: No passphrase given" except for version.test, where there is an error with the agent. The agent error is related to the passphrase though.
(In reply to tka from comment #2) > Created attachment 424252 [details] > Logs of the failed tests > > The common error is "gpg: * failed: No passphrase given" except for > version.test, where there is an error with the agent. The agent error is > related to the passphrase though. Immediately sounds like a pinentry issue, you can try export GPG_TTY=$(tty) and make sure DISPLAY is unset (although I expect it normally tries to use a pinentry loopback).. will look into it after FOSDEM.. (In reply to Kristian Fiskerstrand from comment #3) > Immediately sounds like a pinentry issue, you can try export GPG_TTY=$(tty) > and make sure DISPLAY is unset # export GPG_TTY=$(tty) # unset DISPLAY # MAKEOPTS=-j1 emerge -1 =app-crypt/gnupg-2.1.11::gentoo ==> The tests still fail. Just to rule out other packages that where updated since the last build of gnupg as the culprit, I rebuild gnupg-2.1.10-r1. Its test suite still passes; thus, the failure is likely in gnupg-2.1.11 (or its test suite) itself. (In reply to tka from comment #5) > Just to rule out other packages that where updated since the last build of > gnupg as the culprit, I rebuild gnupg-2.1.10-r1. Its test suite still > passes; thus, the failure is likely in gnupg-2.1.11 (or its test suite) > itself. fwiw, I just re-ran the test suite for .11 here and all passes. What is interesting in your logs is this gpg-preset-passphrase: problem with the agent gpg-preset-passphrase: caching passphrase failed: Invalid response One difference between the systems is that I'm not running it on a hardened host though. Do you have access to non-hardened to see if you can replicate it there? Or access to any logs or information that might explain why gpg-preset-passphrase fails to communicate with gpg-agent over the assuan channel? (In reply to Kristian Fiskerstrand from comment #6) > (In reply to tka from comment #5) > > Just to rule out other packages that where updated since the last build of > > gnupg as the culprit, I rebuild gnupg-2.1.10-r1. Its test suite still > > passes; thus, the failure is likely in gnupg-2.1.11 (or its test suite) > > itself. > > fwiw, I just re-ran the test suite for .11 here and all passes. > > What is interesting in your logs is this > gpg-preset-passphrase: problem with the agent > gpg-preset-passphrase: caching passphrase failed: Invalid response > > One difference between the systems is that I'm not running it on a hardened > host though. Do you have access to non-hardened to see if you can replicate > it there? Or access to any logs or information that might explain why > gpg-preset-passphrase fails to communicate with gpg-agent over the assuan > channel? Seems to be working on hardened system as well. And that test also included gcc 5.. Created attachment 424424 [details]
info.txt
confirmed at a hardened chroot image at a hardened host
-----------------------------------------------------------------
This is an unstable amd64 chroot image (named amd64-hardened-unstable_20160121-225008) at a hardened host acting as a tinderbox.
-----------------------------------------------------------------
make.conf: USE="mmx sse sse2 pax_kernel -cdinstall -oci8 bindist alsa compat -csc -dvd gtk gudev haptic imap javaxml kvm -llvm minizip modplug multimedia ois opengl -openmpi plasma png python qemu qt4 sqlite svg system-jpeg system-libvpx system-sqlite usbredir v4l2 xa xscreensaver xslt"
Created attachment 424426 [details]
app-crypt:gnupg-2.1.11:20160201-220950.log
Created attachment 424428 [details]
emerge-history.txt
Created attachment 424430 [details]
environment
Please try 2.1.11-r1 and see if the issue persists *** Bug 574612 has been marked as a duplicate of this bug. *** Ok, at least now we're down to only the gpgtar test failing (c.f bug 574612) All (34) tests pass now on my system. (In reply to tka from comment #15) > All (34) tests pass now on my system. yes, the gpgtar issue only shows spuriously, c.f this report on debian: http://lists.gnupg.org/pipermail/gnupg-devel/2016-January/030744.html trying to figure out a cause along with upstream and debian maintainers... (In reply to Kristian Fiskerstrand from comment #16) does https://bugs.gentoo.org/show_bug.cgi?id=574612#c9 help ? Does the issue still persist with 2.1.12? (In reply to Kristian Fiskerstrand from comment #18) no, fixed there.. (In reply to Toralf Förster from comment #19) > (In reply to Kristian Fiskerstrand from comment #18) > no, fixed there.. Thanks, closing then |