network down
 

Hello all, I'm working for a company that used to have a Linux guru... he set up an IBM P75 w/16 MB Ram using RH62 as the network DHCP server, also the internet server using masquarade<sp?>. DHCP seems to have gone partially down... however that is possible??? Some users are up, some are down, and it seems others are dropping. A message is coming up saying that the DHCP lease is expiring...

When I try to use the 'man' command to learn a little about certain commands I get a message saying there is no space left on the drive... I figured out that I can use "rm" to remove a file but I have no clue as to which file to remove or even if that will help!

Help ! Please !!

<> < Michael

I'm not really sure about the DHCP problems but running out of disk space can cause all sorts of hassle so I'd look into that one first. Have you been logging a whole lot of info?? If so have a look at /var/log

You can use du -h to find out how much space is being used and where, this may help you out.

good luck, its probably something stupid.

Jamie...

following through
 

I checked /var/log and found the following:

several pages into the list I find todays date with this:
cron with apparantly 1054720 on todays date
messages shows up with a healthy 40083200

There are several other files, one of which I am comfortable deleting such as boot log taking up over 35k and several other files that don't look necessary but what do I know!

going ahead with du -h gives several lines which make no sense to me.

I read elsewhere to try df which showed me a ton of files including several called tty1, tty2 etc... once again no clue as to if I can delete them. I know that there is a file in root taking up 32k that is my fetchmail log, I was wondering if I used rm if it would remove just the file or my fear is that it is a file format and if I delete it I'll be fubar.

Thank you for your response and time !

<> <Michael

The command df should give you a list of the different partitions mounted and how much of the space that is occupied.
I have never heard of df giving a list of ttys.
Please post your info from df along with partition layout and mount point that might help someone to troubleshoot for you.

correction
 

DF did not give me the the file list as I thought, my bad.

Any advice as to whether the CRON or BOOTLOG files are able to be deleted without trouble?

I left this on the other forum b4 i saw this thread was open:
----------------------------
Ahh you closed your thread so I can't reply. Seems you have a number of issues at hand. In terms of diskspace run

df -h

This will give you a list of the filesystem devices and their mount points, their size and useage vs. free space. It sounds as if some of those volumes have filled, quite possibly because logs have become full. You may need to take a look at /var and /tmp, and do some manual pruning of files there. In particular look in /var/log

Furthermore, since this machine acts as a proxy using ipmasquerade, it probably logs all the connections. Over time this log has probably become enormous and can be safely removed. Unfortunately, i don't know what the name of the log would be or where it would be located. You would need to find the configuration file for the firewall and locate it from there.

Also, the machine may just have become squirrely, and due for a reboot. From root issue:

shutdown -r now

to reboot it.

still...
 

deleted several hundred k from /var/logs and rebooted.
Upon logging back on I df -h again and got the following
Filesystem Size Used Avail Use% Mounted on
/dev/hda6 416 415 0 100 /
/dev/hdb1 500 456 25 95 /home
/dev/hda1 25 3.1 20 13 /boot

have been looking through some docs and tried linuxconf... trouble is that everything is already setup and I'm fairly lost as to what the actual problem is... the symptoms are that my DHCP is down and my DSL and Fetchmail not working either...
On startup it gives the message that no mailserver specified for the Fetchmail app... I found how to display the IP Routes and it would seem that I have a missing route... 10.28.128.135... problem is that when I try to

route -v add 10.28.128.135 255.255.255.128 eth1

produces a bunch of route syntax including
inet_route [-vF] add {-host|-net} target [/prefix] [metric M] [network N] [mss Mss] [window W] [irtt I] [Mod] [dyn] [reinstate] [[dev] if]

and

inet_route [-vF] add {-host|-net} target [/prefix] [metric M]

(only used the Add examples)

I can't get

man route

to work because it comes up saying

gzip: stdout: no space left on device

So... is there anyone out there who knows how I can get my route rebuilt and or could tell me what all the variables mean so I can figure out the proper values please?

All this started on Saturday and I have no internal network and only this dial-up internet access to use... I would just go buy a book and RTFM but I live in rural Nebraska...

Still hopeful to resolve
<> <Michael

As can be seen from df output, your / is full, 100%.
The problem is to find where in / the space get's consumed.
Run du -h -n 1 /, not exactly sure of the -n option beeing -n.
This should give you a list with directories found in your / and their corresponding size (a summary of all sub dirs).
Post this result and we will try help you find wich one that might abnormally large.

As your harddisk seems a bit small it would be interresting to know what is actually on this machine.

Back in the saddle
 

:D
Happiness

Thanks for the help, I went in and cleared the messages file from /var/log and also the bootlog, this allowed man route to work and within a few moments I was able to determine that I was not using the -net parameter (sheesh). End result being that my network came back, though I did have a dishwashing Linux guru help with getting IPChains and Masquarade to settle down.

Thank you Dishwashing Linux Guru from North Platte, Nebraska!:D :D

I vow to buy "Running Linux" and read it cover to cover... fool me once shame on you, fool me twice...

Blessings
<> <Michael