Grafana Login Error

This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: grafana登录报错

| username: zhanggame1

[TiDB Usage Environment] Production Environment / Testing / PoC
[TiDB Version] 7.5.1
[Reproduction Path] Entering username and password prompts this, clearing the browser cache did not solve it

| username: TiDBer_QYr0vohO | Original post link

Take a look at the logs.

| username: zhanggame1 | Original post link

Discovered that the hard drive is full, Prometheus has written to capacity.

| username: TiDBer_QYr0vohO | Original post link

Well, I just searched on Google and found someone with the same issue as you, where the disk was full. It’s good to have identified the problem.

| username: zhaokede | Original post link

The error reported is not obvious at all.

| username: TiDBer_rvITcue9 | Original post link

Is there any information in the logs?

| username: zhanggame1 | Original post link

There are no new logs.

| username: TIDB-Learner | Original post link

I feel that Baidu

| username: Kongdom | Original post link

:call_me_hand: After finding the cause, remember to mark it as the best answer so that other members can find the solution more quickly.

| username: 呢莫不爱吃鱼 | Original post link

Is it mixed deployment? Or did Grafana fill up the disk?

| username: xfworld | Original post link

The service must be down, right? :rofl:

| username: wangkk2024 | Original post link

Restart it.

| username: 不想干活 | Original post link

If you convert the error message in the picture into text, it will be more convenient for other friends to find the correct answer.

| username: zhanggame1 | Original post link

Prometheus has occupied 100G of disk space. Currently, the data retention period is set to 10 days. I have changed it to 7 days.

| username: tony5413 | Original post link

Hahaha, hahaha.

| username: Kongdom | Original post link

:yum: This is third-party, probably doesn’t support Chinese, otherwise, it would definitely have Chinese prompt messages.

| username: 小于同学 | Original post link

The server is down, right?

| username: Kongdom | Original post link

Look at the best answer above, the hard drive is full.

| username: yytest | Original post link

Browser login prompt issue: Since you did not provide specific login prompt error information, I can only offer some general troubleshooting steps. If an error occurs on the login page, you can try the following methods:

  • Confirm whether the username and password are correct.
  • Try logging in using a different browser or incognito/private mode.
  • Clear the browser cache and cookies.
  • Ensure the browser supports WebSocket, as TiDB Dashboard and other interfaces may require WebSocket to function properly.
  • Check if browser extensions or firewalls are blocking the login request.
  • Review the server’s log files, such as Nginx or Apache, for possible error information.

Disk full, Prometheus write full issue: If Prometheus fills up the hard disk, it may cause the monitoring system to malfunction and even affect the performance of the TiDB cluster. The steps to resolve this issue are as follows:

  • Increase disk space: If possible, increase the disk space or replace it with a larger hard disk.
  • Clean up Prometheus data: Prometheus usually retains data for a certain period. You can reduce the retention time by modifying the retention_time configuration or manually clean up old data.
  • Migrate Prometheus data: If your environment allows, you can migrate the Prometheus data directory to a larger hard disk.
  • Horizontally scale Prometheus: Distribute the write pressure by running multiple Prometheus instances and using Prometheus’s federation cluster feature.
| username: 呢莫不爱吃鱼 | Original post link

I’ve learned a lot!