[Bug] Login Failed with Error: "database is locked (5) (SQLITE_BUSY)" — Caused by Disk I/O Performance Degradation

[hochenggang]

Contact Information

No response

1Panel Version

v2 beta

Problem Description

Today I was unable to log in to 1Panel, and the following error was shown:

Internal server error: database is locked (5) (SQLITE_BUSY)

After investigation, I found that this issue was caused by a significant degradation in disk write performance. The average disk write latency was around 60ms, whereas under normal conditions, SSDs typically operate below 10ms and HDDs below 100ms. Although the server remains otherwise functional, the SQLite database used by 1Panel appears to be sensitive to disk I/O delays, leading to frequent database lock contention.

Observed Symptoms:

• Unable to log in to 1Panel
• Error: database is locked (5) (SQLITE_BUSY)
• Increased disk write latency

Suggestion:
It would be helpful if 1Panel could implement better handling of SQLite database locks under high I/O latency scenarios, or provide guidance on switching to a more robust database backend for production environments.

API: /api/v2/core/auth/login
RESPONSE:

code: 500
data: null
message: "服务内部错误: database is locked (5) (SQLITE_BUSY)"

Steps to Reproduce

Just click login and error occur.

The expected correct result

No response

Related log output

Additional Information

No response

[hochenggang]

Based on the SQLITE_BUSY issue under high disk I/O latency, here are a few quick suggestions for sqlite.go:

1. Enable WAL mode
   Add this after opening the database to improve concurrency:

   db.Exec("PRAGMA journal_mode=WAL;")

2. Adjust connection pool limits
   SQLite doesn't handle high concurrency well. Set more reasonable values:

   sqlDB.SetMaxOpenConns(10)
   sqlDB.SetConnMaxIdleTime(30 * time.Second)

3. Fix incorrect duration usage
   Make sure all time-related settings use time.Duration, not raw integers.

These changes maybe help reduce lock contention and improve stability during I/O bottlenecks.

[wanghe-fit2cloud]

感谢反馈，我们将在后续版本中持续优化相关问题。
Issue 我将转为 Discussion，欢迎大家继续交流和补充建议。