Ntopng service will not start because Redis service is not running

paftdunk

I'd love an answer for this as well. Ntop/ntopng is one of several things that ate it after a restore. It really feels like pfsense reliability has taken a hit in the last couple years.

ivor

@paftdunk:

I'd love an answer for this as well. Ntop/ntopng is one of several things that ate it after a restore. It really feels like pfsense reliability has taken a hit in the last couple years.

This has nothing to do with pfSense "reliability", it's related to the ntopng package. That being said, I run ntopng package on several boxes without issues, so perhaps start from your configuration. Lastly, I welcome you to submit bugs as everyone else does it when they have an issue.

Thanks.

paftdunk

@ivor:

@paftdunk:

I'd love an answer for this as well. Ntop/ntopng is one of several things that ate it after a restore. It really feels like pfsense reliability has taken a hit in the last couple years.

This has nothing to do with pfSense "reliability", it's related to the ntopng package. That being said, I run ntopng package on several boxes without issues, so perhaps start from your configuration. Lastly, I welcome you to submit bugs as everyone else does it when they have an issue.

Thanks.

A restore leaving multiple subsystems broken goes directly to platform quality. I've documented my related issues elsewhere.

https://forum.pfsense.org/index.php?topic=128262

Thanks.

ivor

@paftdunk:

@ivor:

@paftdunk:

I'd love an answer for this as well. Ntop/ntopng is one of several things that ate it after a restore. It really feels like pfsense reliability has taken a hit in the last couple years.

This has nothing to do with pfSense "reliability", it's related to the ntopng package. That being said, I run ntopng package on several boxes without issues, so perhaps start from your configuration. Lastly, I welcome you to submit bugs as everyone else does it when they have an issue.

Thanks.

A restore leaving multiple subsystems broken goes directly to platform quality. I've documented my related issues elsewhere.

https://forum.pfsense.org/index.php?topic=128262

Thanks.

Sorry but that's not documenting or submitting a bug report. Regarding the forum thread (unrelated to ntopng I might add), so? What's your point? You had an issue which you were unable to resolve and you decided to blame the "platform" and claim "reliability has taken a hit". If it's not clear, I don't like your attitude. That's not the right way to ask for help.

doktornotor

The answer to this problem is that you most likely have screwed redis DB, you should wipe /var/db/ntopng directory (or at least /var/db/ntopng/*.rdb) , uninstall and reinstall the package and start from scratch.

As for the rest - useless noise.

HELIA-Dar

TY Doktornotor!

Wipping the /var/db/ntopng directory solved the issue.

Andrew453

I get this error on a reboot, and found that it's related to the fact I've put temp and var on a RAM disc. Ntopng needs a reinstall after a reboot to sort it, but is otherwise stable.

yalb88

@vladimirm Thank you,
worked for you

KB8DOA

@doktornotor said in Ntopng service will not start because Redis service is not running:

wipe /var/db/ntopng directory

Can someone share the instructions on how to:
wipe the /var/db/ntopng directory
?

Thank You

yalb88

@kb8doa you should go to shell and write "redis-server /usr/local/etc/redis.conf"

KB8DOA

@yalb88 said in Ntopng service will not start because Redis service is not running:

@kb8doa you should go to shell and write "redis-server /usr/local/etc/redis.conf"

I have tried this and keep getting this:

[Redis.cpp:119] ERROR: MISCONF Redis is configured to save RDB snapshots, but it is currently not able to persist on disk. Commands that may modify the data set are disabled, because this instance is configured to report errors during writes if RDB snapshotting fails (stop-writes-on-bgsave-error option). Please check the Redis logs for details about the RDB error.

KB8DOA

This post is deleted!

KB8DOA

This post is deleted!

Paint

the same thing was happening to me. I have a script that I run at boot to check if various ntopng settings are changed in the configurations manually (aka ntopng was reinstalled so my customizations were overriden).

Anyway, to get this working again, you need to remove the ::1 address from the redis bind command in ntopng.sh.

the following sed command from CLI will do it for you. Then stop and restart your ntopng.

sed -i '' 's/::1//g' /usr/local/etc/rc.d/ntopng.sh

If you want to put this in a script like me and run it via cron / startup - here is a snippet of how I do it:

#!/bin/sh

if grep -q "::1" /usr/local/etc/rc.d/ntopng.sh
then
    # code if found
    echo "removing ::1 from ntopng.sh"
    sed -i '' 's/::1//g' /usr/local/etc/rc.d/ntopng.sh
else
    # code if not found
    echo "::1 not in ntopng.sh"
fi

trusmith

Posting this here for anyone who's about to give up on life trying to sort this. Now I'm not saying that this is the ACTUAL solution, but this is what worked for me and surprisingly no post/thread seems to be talking about it clearly.

For most people, if none of the above mentioned solutions worked for you all you need to do is the following in the exact same order:

Uninstall ntopng from the WebGUI Package Manager successfully.
Reboot
Delete the ntopng folder in /var/db/
Delete the redis conf file in /usr/local/etc/redis.conf (it will be recreated again with the defaults once you reinstall the pkg)
Reboot
From the WebGUI, install ntopng and then go to settings and enable it. It will probably fail at this point but don't worry as most people give up at this stage.
Reboot
Once the system comes up, ntop should be running.

Hardly a groundbreaking solution but missing step # 7 was the culprit for me. Hope this helps others like myself.

Now I'm not sure which hook, argument, port, sock or permission is the reason behind this issue, but let's agree that there definitley is a bug which needs squashing.

Thanks & Regards,
John

thrx1

@trusmith said in Ntopng service will not start because Redis service is not running:

Delete the ntopng folder in /var/db/
Delete the redis conf file in /usr/local/etc/redis.conf (it will be recreated again with the defaults once you reinstall the pkg)

What is the correct way to do this?

I tried going to shell and writing "redis-server /usr/local/etc/redis.conf" but keep getting the message "redis-server /usr/local/etc/redis.conf: Command not found".

thrx1

@thrx1 said in Ntopng service will not start because Redis service is not running:

What is the correct way to do this?
I tried going to shell and writing "redis-server /usr/local/etc/redis.conf" but keep getting the message "redis-server /usr/local/etc/redis.conf: Command not found".

Bump. Anyone able to help me here?

bigjohns97

@trusmith said in Ntopng service will not start because Redis service is not running:

Posting this here for anyone who's about to give up on life trying to sort this. Now I'm not saying that this is the ACTUAL solution, but this is what worked for me and surprisingly no post/thread seems to be talking about it clearly.

For most people, if none of the above mentioned solutions worked for you all you need to do is the following in the exact same order:

Uninstall ntopng from the WebGUI Package Manager successfully.

Reboot

Delete the ntopng folder in /var/db/

Delete the redis conf file in /usr/local/etc/redis.conf (it will be recreated again with the defaults once you reinstall the pkg)

Reboot

From the WebGUI, install ntopng and then go to settings and enable it. It will probably fail at this point but don't worry as most people give up at this stage.

Reboot

Once the system comes up, ntop should be running.

Hardly a groundbreaking solution but missing step # 7 was the culprit for me. Hope this helps others like myself.

Now I'm not sure which hook, argument, port, sock or permission is the reason behind this issue, but let's agree that there definitley is a bug which needs squashing.

Thanks & Regards,
John

Excellent solution, worked for me!

Thanks!