multiple logs based off how many servers its in

[silvia552]

my bot is in 3 servers and if i send a message it logs it 3 times:

[2026-02-14 10:37:10] [MESSAGE CREATE] [Guild ID: 1471455535000846349] [Channel ID: 1471836999039189093] [Message ID: 1472255356066599094] [User ID: 1466340128359321750] [Username: my_discord_username] [Content: test] 
[2026-02-14 10:37:10] [MESSAGE CREATE] [Guild ID: 1471455535000846349] [Channel ID: 1471836999039189093] [Message ID: 1472255356066599094] [User ID: 1466340128359321750] [Username: my_discord_username] [Content: test] 
[2026-02-14 10:37:13] [MESSAGE DELETE] [Guild ID: 1471455535000846349] [Channel ID: 1471836999039189093] [Message ID: 1472255356066599094] 
[2026-02-14 10:37:13] [MESSAGE DELETE] [Guild ID: 1471455535000846349] [Channel ID: 1471836999039189093] [Message ID: 1472255356066599094] 

here is the code:

#include <dpp/dpp.h>
#include <ctime>
#include <iostream>
#include <fstream>

const std::string BOT_TOKEN = "redacted";
const std::string msg_log_path = "/path/to/file/msg_log.txt";

int main() {
    dpp::cluster bot(BOT_TOKEN, dpp::i_all_intents);

    bot.on_log(dpp::utility::cout_logger());

    bot.on_slashcommand([](const dpp::slashcommand_t &event){
        if (event.command.get_command_name() == "bot") {
            event.reply("### im always watching...");
        }
    });

    bot.on_ready([&bot](const dpp::ready_t& event) {
            if (dpp::run_once<struct register_bot_commands>()) {
                bot.global_command_create(dpp::slashcommand("bot", "### im always watching...", bot.me.id));
            }
        });

    bot.on_message_create([](const dpp::message_create_t &event){

        time_t timestamp = time(NULL);
        struct tm datetime = *localtime(&timestamp);
        char output[50];
        strftime(output, 50, "[%Y-%m-%d %T] ", &datetime);

        std::ofstream msg_log(msg_log_path, std::ios::app);
        if (!msg_log) {
            std::cerr << "[MESSAGE CREATE ERROR] Could not open " << msg_log_path << std::endl;
        }

        msg_log << output <<
            "[MESSAGE CREATE] " <<
            "[Guild ID: " << event.msg.guild_id << "] " <<
            "[Channel ID: " << event.msg.channel_id << "] " <<
            "[Message ID: " << event.msg.id << "] " <<
            "[User ID: " << event.msg.author.id << "] " <<
            "[Username: " << event.msg.author.username << "] ";

        if (event.msg.content != "") {
            msg_log << "[Content: " << event.msg.content << "] ";
        }

        if (event.msg.attachments.size() > 0) {
            msg_log << "[Attachments: ";
        }

        bool comma = false;
        bool attachments = false;
        for (const auto &attachment : event.msg.attachments) {
            if (comma) {
                msg_log << ", ";
            }
            else {
                comma = true;
            }
            msg_log << attachment.url;
            attachments = true;
        }
        if (attachments) {
            msg_log << "] ";
        }

        msg_log << std::endl;
        msg_log.close();
    });

    bot.on_message_delete([](const dpp::message_delete_t &event){
        time_t timestamp = time(NULL);
        struct tm datetime = *localtime(&timestamp);
        char output[50];
        strftime(output, 50, "[%Y-%m-%d %T] ", &datetime);

        std::ofstream msg_log(msg_log_path, std::ios::app);
        if (!msg_log) {
            std::cerr << "[MESSAGE DELETE ERROR] Could not open " << msg_log_path << std::endl;
        }
        msg_log << output <<
            "[MESSAGE DELETE] " <<
            "[Guild ID: " << event.guild_id << "] " <<
            "[Channel ID: " << event.channel_id << "] " <<
            "[Message ID: " << event.id << "] ";
        msg_log << std::endl;
        msg_log.close();
    });

    bot.on_message_update([](const dpp::message_update_t &event){
        time_t timestamp = time(NULL);
        struct tm datetime = *localtime(&timestamp);
        char output[50];
        strftime(output, 50, "[%Y-%m-%d %T] ", &datetime);

        std::ofstream msg_log(msg_log_path, std::ios::app);
        if (!msg_log) {
            std::cerr << "[MESSAGE UPDATE ERROR] Could not open " << msg_log_path << std::endl;
        }
        msg_log << output <<
            "[MESSAGE UPDATE] " <<
            "[Guild ID: " << event.msg.guild_id << "] " <<
            "[Channel ID: " << event.msg.channel_id << "] " <<
            "[Message ID: " << event.msg.id << "] " <<
            "[User ID: " << event.msg.author.id << "] " <<
            "[Username: " << event.msg.author.username << "] " <<
            "[Content: " << event.msg.content << "] ";
        msg_log << std::endl;
        msg_log.close();
    });

    bot.start(dpp::st_wait);
    return 0;
}

[braindigitalis]

check you're not running 2 copies of the bot at the same time. there are 2 of each log line not 3.

[braindigitalis]

also take a look at spdlog. it will be far better than anything you are doing here.

[silvia552]

check you're not running 2 copies of the bot at the same time. there are 2 of each log line not 3.

i run it once from the terminal

also take a look at spdlog. it will be far better than anything you are doing here.

ill check it out, thanks

[braindigitalis]

the logs say otherwise, log the pid in your log line to confirm

[silvia552]

the logs say otherwise, log the pid in your log line to confirm

i cannot seem to recreate the doubled logs... ig it fixed itself? (i made another bot and stop using this one like a day after posting this issue i had)

[silvia552]

i caught it, it seems that it spawns zombie processes after being ran for days on end

and yes i am more than 100% sure that i am not running it twice...

[Ashthetik]

I would double check your deployment of the bot and log your PIDs as Brain suggested, if it's happening over a long period of time or many restart cycles you might not be terminating the process correctly

On a side note, I'd recommend keeping std::cerr to errors, since it'll be more consistent with naming conventions

[Jaskowicz1]

@silvia552 Could you try retest this on a different machine to see if the issue occurs?

[Jaskowicz1]

Closing due to stale. We can't repro this and will need more info.

Please feel free to reopen this if you can get us more information.