App crashes after some days, probably memory leak?

[leondape]

What happened?

I had the app crashing after 1-2 weeks. Restart fixes it. Today i could retrieve a log for the first time.
I can hardly remember, but I thought I catched up memory leaks mentioned somewhere in the terminal, but I am not sure in what context I read about a memory leak warning.

Steps to Reproduce

1. App is running for a prolonged time period
2. crashed and not reachable, docker container is still up and running.

What browsers are you seeing the problem on?

No response

Relevant log output

<--- Last few GCs --->

[32:0x7f6ed03dd020] 76545407 ms: Scavenge 4043.2 (4122.6) -> 4042.2 (4143.4) MB, 15.28 / 0.00 ms  (average mu = 0.217, current mu = 0.139) allocation failure; 
[32:0x7f6ed03dd020] 76547091 ms: Mark-Compact 4048.6 (4143.9) -> 4043.7 (4149.3) MB, 1680.14 / 0.00 ms  (average mu = 0.166, current mu = 0.050) allocation failure; scavenge might not succeed


<--- JS stacktrace --->

FATAL ERROR: Reached heap limit Allocation failed - JavaScript heap out of memory
----- Native stack trace -----

Screenshots

No response

Code of Conduct

• I agree to follow this project's Code of Conduct

[danny-avila]

Review the logs on app startup, user sessions can lead to memory leaks.

It does use in-memory caching for a lot of operations. The best to do in production is to use Redis, which is already integrated with LibreChat and solves this.

Disabling terminal logging also helps.

However, I'll review how the memory allotment can be improved.

In general, minimum is 1GB of RAM but it's recommend to use 2 GB of RAM. I use 2gb and never run into this issue using vanilla docker. i can go months with it running but I usually update once or twice a week.

[DenisPalnitsky]

In documentation I see that Note: Redis support is experimental, and you may encounter some problems when using it.
Is it still in experimental phase?

[AlliotTech]

``` > cross-env NODE_ENV=production node api/server/index.js

2024-12-26T07:15:48.126Z info: [Optional] Redis not initialized. Note: Redis support is experimental.
2024-12-26T07:15:50.372Z info: Connected to MongoDB
2024-12-26T07:15:52.930Z debug: [indexSync] There are 1317329 messages and 1315925 indexed
2024-12-26T07:15:52.930Z debug: [indexSync] There are 136749 convos and 136749 indexed
2024-12-26T07:15:52.931Z debug: [indexSync] Messages out of sync, indexing
2024-12-26T07:15:53.010Z info: Custom config file loaded:
2024-12-26T07:15:53.011Z info: {

This memory issue occurs frequently when the message volume is relatively large.

NODE_OPTIONS: "--max-old-space-size=4096"