fix: kill stale GPG agents before key generation by nogeenharrie · Pull Request #892 · IJHack/QtPass

nogeenharrie · 2026-04-02T14:15:29Z

Summary

Fixes issue #815 - GPG key generation fails due to a timeout while waiting on a locked key database.

Problem

When another GPG process (from another application) holds a lock on the key database, key generation fails with:

gpg: Note: database_open 134217901 waiting for lock (held by 9485) ...
gpg: keydb_search failed: Connection timed out
gpg: key generation failed: SQL library used incorrectly

Solution

Before generating GPG keys, kill any stale gpg-agent processes that might be holding locks:

Executor::executeBlocking(gpgPath, {"--kill", "gpg-agent"});
Executor::executeBlocking(gpgPath, {"gpgconf", "--kill", "gpg-agent"});

This releases any stale locks on the GPG key database before attempting key generation.

Testing

Build passes with Qt 6.8.2 and Qt 5.15
All unit tests pass

Summary by CodeRabbit

Bug Fixes
- Improved GPG key generation reliability by ensuring any stale GPG agent is terminated before generating new keys.
- Reduces failures and hangs during key creation, leading to a smoother, more predictable key generation experience for users.

coderabbitai

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@src/pass.cpp`:
- Around line 207-213: The two Executor::executeBlocking calls are invoking gpg
incorrectly and calling gpgconf as an argument; fix by deriving the gpgconf
executable path from gpgPath (use QFileInfo to get gpgPath's directory) then
call Executor::executeBlocking(gpgPath, {"--kill", "gpg-agent"}) should be
removed or replaced with the proper gpgconf invocation: build gpgconfPath from
QFileInfo(gpgPath).absolutePath() + QDir::separator() + "gpgconf" and call
Executor::executeBlocking(gpgconfPath, {"--kill", "gpg-agent"}); add `#include`
<QFileInfo> to the includes.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

Push a commit to this branch (recommended)
Create a new PR with the fixes

ℹ️ Review info

⚙️ Run configuration

Configuration used: Repository UI (base), Organization UI (inherited)

Review profile: ASSERTIVE

Plan: Pro

Run ID: 882a7afc-c438-4aab-be45-ca4c6ad895fa

📥 Commits

Reviewing files that changed from the base of the PR and between 7a9fe4c and 86cc302.

📒 Files selected for processing (1)

src/pass.cpp

coderabbitai · 2026-04-02T14:19:17Z

src/pass.cpp

+  // Kill any stale GPG agents that might be holding locks on the key database
+  // This helps avoid "database locked" timeouts during key generation
+  QString gpgPath = QtPassSettings::getGpgExecutable();
+  if (!gpgPath.isEmpty()) {
+    Executor::executeBlocking(gpgPath, {"--kill", "gpg-agent"});
+    Executor::executeBlocking(gpgPath, {"gpgconf", "--kill", "gpg-agent"});
+  }


⚠️ Potential issue | 🔴 Critical

🧩 Analysis chain

🌐 Web query:

What is the correct command to kill gpg-agent using gpgconf?

💡 Result:

The correct command to kill gpg-agent using gpgconf is gpgconf --kill gpg-agent.

Citations:

1: https://www.gnupg.org/documentation/manuals/gnupg/Invoking-GPG_002dAGENT.html

2: https://www.gnupg.org/(en)/documentation/manuals/gnupg24/gpgconf.1.html

3: https://www.gnupg.org/(en)/documentation/manuals/gnupg24/gpg-agent.1.html

4: https://www.gnupg.org/documentation/manuals/gnupg26/gpgconf.1.html

🏁 Script executed:

# Check for existing includes in src/pass.cpp around the code area head -50 src/pass.cpp | grep -E "^#include"

Repository: IJHack/QtPass

Length of output: 259

🏁 Script executed:

# Search for gpgconf references in the codebase rg "gpgconf" --type cpp --type h

Repository: IJHack/QtPass

Length of output: 145

🏁 Script executed:

# Search for QFileInfo usage in the codebase rg "QFileInfo" --type cpp --type h -A 2 | head -30

Repository: IJHack/QtPass

Length of output: 1429

🏁 Script executed:

# Search for QtPassSettings methods to see if there's a pattern for executable paths rg "class QtPassSettings|getGpg|Executable" --type cpp --type h -A 1 | head -50

Repository: IJHack/QtPass

Length of output: 2967

Both agent-kill commands are incorrect and will fail silently.

Line 211 executes gpg --kill gpg-agent, which is invalid—GPG does not support the --kill option. Line 212 executes gpg gpgconf --kill gpg-agent (with "gpgconf" as an argument to the gpg binary) instead of running the gpgconf executable directly. The correct command is gpgconf --kill gpg-agent.

Derive the gpgconf path from gpgPath by extracting its directory. This pattern is already used elsewhere in the codebase (e.g., tst_util.cpp). Add #include <QFileInfo> to the includes and update the code:

Fix: correctly invoke gpgconf to kill the agent

// Kill any stale GPG agents that might be holding locks on the key database // This helps avoid "database locked" timeouts during key generation QString gpgPath = QtPassSettings::getGpgExecutable(); if (!gpgPath.isEmpty()) { - Executor::executeBlocking(gpgPath, {"--kill", "gpg-agent"}); - Executor::executeBlocking(gpgPath, {"gpgconf", "--kill", "gpg-agent"}); + // Derive gpgconf path from gpg path (they are typically in the same directory) + QFileInfo gpgInfo(gpgPath); + QString gpgconfPath = gpgInfo.absolutePath() + QDir::separator() + "gpgconf"; +#ifdef Q_OS_WIN + gpgconfPath += ".exe"; +#endif + if (QFile::exists(gpgconfPath)) { + Executor::executeBlocking(gpgconfPath, {"--kill", "gpg-agent"}); + } }

Add to includes section:

`#include` <QFileInfo>

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed. In `@src/pass.cpp` around lines 207 - 213, The two Executor::executeBlocking calls are invoking gpg incorrectly and calling gpgconf as an argument; fix by deriving the gpgconf executable path from gpgPath (use QFileInfo to get gpgPath's directory) then call Executor::executeBlocking(gpgPath, {"--kill", "gpg-agent"}) should be removed or replaced with the proper gpgconf invocation: build gpgconfPath from QFileInfo(gpgPath).absolutePath() + QDir::separator() + "gpgconf" and call Executor::executeBlocking(gpgconfPath, {"--kill", "gpg-agent"}); add `#include` <QFileInfo> to the includes.

Before generating GPG keys, kill any stale gpg-agent processes that might be holding locks on the key database. This helps prevent 'timeout while waiting on locked key database' errors. The issue occurs when another GPG process (possibly from another application) holds a lock on the key database, causing key generation to fail with timeout errors.

coderabbitai

♻️ Duplicate comments (1)

src/pass.cpp (1)

209-212: ⚠️ Potential issue | 🔴 Critical

Invoke gpgconf directly; current command line is malformed.

Line 211 currently runs gpg with "gpgconf" as an argument, which is not the same as executing gpgconf --kill gpg-agent. This means stale-agent cleanup may not happen and the lock-timeout failure can persist.

Proposed fix

   QString gpgPath = QtPassSettings::getGpgExecutable();
   if (!gpgPath.isEmpty()) {
-    Executor::executeBlocking(gpgPath, {"gpgconf", "--kill", "gpg-agent"});
+    QFileInfo gpgInfo(gpgPath);
+    QString gpgconfPath = gpgInfo.absolutePath() + QDir::separator() + "gpgconf";
+#ifdef Q_OS_WIN
+    gpgconfPath += ".exe";
+#endif
+    const int rc = Executor::executeBlocking(gpgconfPath, {"--kill", "gpg-agent"});
+#ifdef QT_DEBUG
+    if (rc != 0) {
+      dbg() << "Failed to kill gpg-agent:" << gpgconfPath << rc;
+    }
+#endif
   }

What is the correct GnuPG command to terminate gpg-agent, and must it be executed via the gpgconf executable (not as an argument to gpg)?

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@src/pass.cpp` around lines 209 - 212, The code is invoking the gpg binary
with "gpgconf" as an argument (gpgPath + {"gpgconf","--kill","gpg-agent"}) which
is wrong; instead execute the gpgconf program directly. Replace the call that
uses gpgPath and args {"gpgconf","--kill","gpg-agent"} so it invokes the gpgconf
executable (e.g. use a QtPassSettings::getGpgConfExecutable() if available or
the literal "gpgconf") and call Executor::executeBlocking(gpgconfPath,
{"--kill","gpg-agent"}) so gpg-agent is terminated properly.

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Duplicate comments:
In `@src/pass.cpp`:
- Around line 209-212: The code is invoking the gpg binary with "gpgconf" as an
argument (gpgPath + {"gpgconf","--kill","gpg-agent"}) which is wrong; instead
execute the gpgconf program directly. Replace the call that uses gpgPath and
args {"gpgconf","--kill","gpg-agent"} so it invokes the gpgconf executable (e.g.
use a QtPassSettings::getGpgConfExecutable() if available or the literal
"gpgconf") and call Executor::executeBlocking(gpgconfPath,
{"--kill","gpg-agent"}) so gpg-agent is terminated properly.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Repository UI (base), Organization UI (inherited)

Review profile: ASSERTIVE

Plan: Pro

Run ID: 1cab2960-c85d-4025-bdf8-85c9af7c919a

📥 Commits

Reviewing files that changed from the base of the PR and between 86cc302 and b9d8446.

📒 Files selected for processing (1)

src/pass.cpp

This comment has been minimized.

Sign in to view

coderabbitai bot reviewed Apr 2, 2026

View reviewed changes

This comment has been minimized.

Sign in to view

nogeenhenk approved these changes Apr 2, 2026

View reviewed changes

nogeenharrie force-pushed the fix/gpg-keygen-timeout branch from 86cc302 to b9d8446 Compare April 2, 2026 14:47

This comment has been minimized.

Sign in to view

coderabbitai bot reviewed Apr 2, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix: kill stale GPG agents before key generation#892

fix: kill stale GPG agents before key generation#892
nogeenharrie wants to merge 1 commit intomainfrom
fix/gpg-keygen-timeout

nogeenharrie commented Apr 2, 2026 •

edited by coderabbitai bot

Loading

Uh oh!

This comment has been minimized.

coderabbitai bot left a comment

Uh oh!

coderabbitai bot Apr 2, 2026

Uh oh!

This comment has been minimized.

This comment has been minimized.

This comment has been minimized.

coderabbitai bot left a comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Uh oh!

Conversation

nogeenharrie commented Apr 2, 2026 • edited by coderabbitai bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Problem

Solution

Testing

Summary by CodeRabbit

Uh oh!

This comment has been minimized.

coderabbitai bot left a comment

Choose a reason for hiding this comment

Uh oh!

coderabbitai bot Apr 2, 2026

Choose a reason for hiding this comment

Uh oh!

This comment has been minimized.

This comment has been minimized.

This comment has been minimized.

coderabbitai bot left a comment

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

nogeenharrie commented Apr 2, 2026 •

edited by coderabbitai bot

Loading