How to restart crashed HTTPD server and Transmission?

[Cubytus]

Hi there,

this may sound as a silly question, but how can one restart daemons when they crash? Is there a documentation page for that?

The usual, Linux way would use either

Code: Select all

systemctl

or

Code: Select all

service

commands, but they aren't available in fvdw, so these have to be called directly.

When running, Transmission daemon uses command line

Code: Select all

/usr/local/bin/transmission-daemon -g /share/1000/tr-daemon -w /share/1000/tr-downloads

under user rd

However when I call httpd from the command line with

/usr/bin/httpd -f httpd.conf -d /etc -k start

it runs as user root.

Is that the proper way to restart?

[fvdw]

we don't use systemctl as it increade firmware file a lot.
So yes you should start from the command line using the commands as shown in ps -ef overview.
For apache it is

Code: Select all

/usr/bin/httpd -f httpd.conf -d /etc

There will be then two lines the script to launch httpd is run by root but the apache server is running under user nobdy. This is set in the httpd.conf file

Code: Select all

root      1326     1  0 15:40 ?        00:00:00 /usr/bin/httpd -f httpd.conf -d /etc
nobody    1330  1326  0 15:40 ?        00:00:00 /usr/bin/httpd -f httpd.conf -d /etc


[Cubytus]

Good to know!

Any specific reason Apache is the web server and not another, lighter or faster? I thought it wasn't the most efficient on smaller devices likes the ones fvdw is designed to power.

As for Transmission: is the command correct?

As a suggestion, even if fvdw is GUI-driven, these command lines should be in the documentation, just in case a process crashes.

[fvdw]

all daemons except apache can be stopped and started from the webinterface.

To start or stop a daemon requires more then only running the start or stop command. Therefor it is not recommended to start daemons from the command line but to use the webinterface

[Jocko]

Hi Cubytus

Hi there,

this may sound as a silly question, but how can one restart daemons when they crash? Is there a documentation page for that?

The usual, Linux way would use either

Code: Select all

systemctl

or

Code: Select all

service

commands, but they aren't available in fvdw, so these have to be called directly.

When running, Transmission daemon uses command line

Code: Select all

/usr/local/bin/transmission-daemon -g /share/1000/tr-daemon -w /share/1000/tr-downloads

under user rd

However when I call httpd from the command line with

/usr/bin/httpd -f httpd.conf -d /etc -k start

it runs as user root.

Is that the proper way to restart?

As mentioned by Fvdw, you have to use the firmware web interface to cleanly restart Transmission.
Transmission must run as rd to protect the firmware partitions. With transmission you can easily change the downloading location. So if you select a location from a firmware partition, there will be quickly no free space for the firmware.

The user rd can not write on the firmware partition instead of the user root.

Some other environment variables are also set by the firmware before starting Transmission

[Cubytus]

It solved the issue for a while, but now it has returned…

For some reason, Apache stopped accepting connections. I restarted it from the CLI with command line
/usr/bin/httpd -f httpd.conf -d /etc -k start

The Web interface does load, but trying to get to another page just never loads completely:

Captura de Pantalla 2025-03-31 a la(s) 13.45.46 .png

Since it can't load pages, I can't restart daemons from there.

Any way to know what's causing Apache to hang?

[Jocko]

Hi

Currently I think the fimware fails to daemonize httpd or all threads are using with some no-end loops for example and then no new request can be handled by the daemon
Can you post

Code: Select all

ps axf|grep httpd

and

Code: Select all

 netstat -ap|grep httpd

You can also check if there is no error with conf file

Code: Select all

 /usr/bin/httpd -f httpd.conf -d /etc -t

[Cubytus]

Here it is:

Code: Select all

12914 pts/0    R+     0:00          \_ grep httpd
11730 ?        Ss     0:05 /usr/bin/httpd -f /etc/httpd.conf -d /etc -k start
11735 ?        S      0:11  \_ /usr/bin/httpd -f /etc/httpd.conf -d /etc -k start
11742 ?        S      0:02  \_ /usr/bin/httpd -f /etc/httpd.conf -d /etc -k start

and:

Code: Select all

tcp        6      0 :::http                 :::*                    LISTEN      11730/httpd
tcp       95      0 :::4343                 :::*                    LISTEN      11730/httpd


No reported error in httpd.conf.

[Jocko]

Ok
So httpd is daemonized and ports are opened. That would mean the httpd threads are hanged.

Then do

Code: Select all

killall httpd

and check if all processes are stopped

Code: Select all

ps axf|grep httpd

Start again httpd

Code: Select all

/usr/bin/httpd -f httpd.conf -d /etc

and try to load the firmware interface

[Cubytus]

Killed httpd process.

Indeed they're stopped:

Code: Select all

13191 pts/0    S+     0:00          \_ grep httpd

Restarting httpd:

Code: Select all

13204 pts/0    S+     0:00          \_ grep httpd
13198 ?        Ss     0:00 /usr/bin/httpd -f httpd.conf -d /etc
13202 ?        S      0:00  \_ /usr/bin/httpd -f httpd.conf -d /etc


Web interface loads on the home page, but refuses to do anything else.

Captura de Pantalla 2025-04-01 a la(s) 20.04.46 .png

On hovering, none of the tabs are clickable.