Linux Corner

[Avtrician]

Because programs (Virii) cant install and run at user level (not with out a lot of effort), Av is not as needed in Linux. There is usualy a firewall that is installed, its just not as intrusive or recsource hungry as in Windoze.

Helped a workmate install Ubuntu on an old troublesome (under XP) Dell laptop. everything including the webcam worked without effort.

[rgbrock1]

BOAC:

As others have so correctly stated: I wouldn't worry about av on linux.
It is a rarity when any Linux distro suffers from virus infection. Don't give it another thought.

As for a firewall. Following this link and behold the joys of configuring iptables, if you so desire, or other methods of installing a firewall on your ubuntu distro. (Ubuntu is one of the few Linux distros which do not come with a default firewall installation.)

Basic Ubuntu Linux Firewall Configuration - Techotopia

[rans6andrew]

I spent the week stealing components from another, reliable, working machine and proved that the crashing is not power supply, wireless card, graphics card or ram module related. It is not the software as it has been known to spontaneously restart before it has completed the POST. I also updated to Ubuntu 10.04 to see if that made any difference. I called Scan to ask about returning the motherboard and the CPU and they suggested I try updating the bios. This I have done and, fingers crossed, it seem to be more stable. 22 hours and no restart.

One thing that I have noticed is that the scroll wheel on the mouse doesn't always work. This feature has been hit and miss since the 10.04 upgrade. Any idea why? or how to sort it?

Rans6....

[Mac the Knife]

"One thing that I have noticed is that the scroll wheel on the mouse doesn't always work."

Firefox does this on my Windows XP machine (yes, I have one...) - I think its a Firefox thing - Opera doesn't.

Mac

[rans6andrew]

'tis not a FF thing in this case, it doesn't work in gedit either.

[The Nr Fairy]

Sorta Linux question, sorta Mac (VMware Fusion) question.

My VM running Ubuntu 10 LTS worked ok for a few hours. Now I can't click in the VM's window and get it to focus - clicking / cmd-G, nothing.

Googling has been little help, I suspect because I'm not using the right terms

Anyone else seen this ?

[Hawkeye79]

@rgbrock1
Firestarter is a good graphical firewall on ubuntu. It can be installed via Synaptic or apt-get.

[rgbrock1]

Good point Hawkeye. Firestarter is indeed a nice GUI firewall front-end.

[N727NC]

Linux Server Problem

The new-to-us family server is a secondhand Dell Poweredge 64 bit machine, 4GB, running Suse 11.2 (64bit) with a 180 GB drive with the operating system only and a Linux Raid 1 of 2 1 TB drives mounted as /home. The Raid is about half full of family data. You might think that the 180 GB drive should last forever before it filled, but only a month after commissioning the server, the drive is full.

What is likely to have caused this and where do I start to look for junk to delete?

[rgbrock1]

First place to look is in /var/log. Can be lots of useless crap in there.
Second place, /tmp. (Although most Linux distros delete the contents of /tmp on reboot)

And if you really want to see what's chewing up your hard drive space issue the following command:

$ cd /
$ du -h | more

This will give you a rather lengthy listing of ALL your files on the entire hard drive.
You'll have to hit the <return> key to scroll through the listing.

Hope this helps.

[MacBoero]

It might be worth looking into setting up some crontab entries to clear out the log directory at regular intervals automatically.

[N727NC]

Thanks

Thank you for the suggestion RGB1 - I cleared out /var/log and /tmp (the wonders of shift+del!) and now the drive has 3.7 GB on it and the server is happy again. There is no way that there were over 175Gb of messages, so I'm guessing somewhere a file was incorrectly allocating space - perhaps following a less-than-orthodox shut-down.

I'll look into crontab to see if I can set up a routine deletion of old logs.

[Gertrude the Wombat]

"Family" machine I think you said ...

Maybe some of those "log files" were in fact not log files but the kids' cunningly hidden stolen porno movies.

[aerobelly]

Before deleting logs check just how much space they take. On this 'ere system, Kubuntu 10.04 installed just under 6 months ago, all the logs since then take all of 54MiB. If there are problems the logs can help a lot in isolating them to hardware/software/circumstances. Logs are in plaintext and very repetitive, so they compress very well (that's the .log.N.gz files). They shouldn't take much space, but can save your ass. One place I'd look is .xsessionerrors in users' home directories. They can get huge, and because they're out of the normal log hierarchy are frequently overlooked. If you're going to scan the whole disk for space usage try sudo du / | sort -nr | more This will list directories in order of the space they take. But it's slooooooow in GiB territory, what it's like in TiB areas I hate to think. (The "sudo" is needed to see into other users' usage, if you haven't used it before.) There are also graphical tools that do much the same job, if looking at columns of numbers doesn't float your boat. 'b

[N727NC]

Gertie - that sort of thing is hidden on pendrives - they know Dad is too much of a geek for them to get away with it on the server.
AB - too late now, all are gone, but you confirm what I saw, which is that logs aren't large files. there must have been something in the /tmp area which either was genuinely huge or was erroneously reporting itself as such.

Grateful to all for the very valuable help.

[LH2]

The quick and dirty way, preferably as root:

du -csh /* 2>/dev/null

Example output:

8.3M /bin
23M /boot
264K /dev
41M /etc
73G /home
156M /lib
23M /lib64
16K /lost+found
4.0K /media
4.0K /mnt
4.0K /opt
0 /proc
1.6M /root
12M /sbin
4.0K /selinux
214G /srv
0 /sys
728M /tftpboot
6.0M /tmp
3.4G /usr
20G /var
311G total


Then repeat as needed on any subdirectories of interest, e.g.,:

du -csh /var/* 2>/dev/null
61M /var/adm
50M /var/cache
4.0K /var/crash
4.0K /var/games
20G /var/lib
32K /var/lock
79M /var/log
0 /var/mail
4.0K /var/opt
372K /var/run
99M /var/spool
17M /var/tmp
4.0K /var/X11R6
12K /var/yp
20G total


An alternative syntax to have the output appear in order of decreasing size:

du -s /var/* 2>/dev/null |sort -nr
20403992 /var/lib
100660 /var/spool
80352 /var/log
61476 /var/adm
51000 /var/cache
17360 /var/tmp
372 /var/run
32 /var/lock
12 /var/yp
4 /var/X11R6
4 /var/opt
4 /var/games
4 /var/crash
0 /var/mail


Various graphical tools are also available, but I always use the above as it's guaranteed to be on any Linux installation (some embedded platforms excepted).

[txdmy1]

archiving logs

copy them daily and compress to an oldlogs directory with a cron script. Delete any over 30 days old in oldlogs within same script, simples.
Copied this from what we did @ work when I moved onto nix systems support and set up my own sever as a VM image. Don't use it much now though, thinking about getting rid

[LH2]

Originally Posted by txdmy1

copy them daily and compress to an oldlogs directory with a cron script. Delete any over 30 days old in oldlogs within same script, simples.

man logrotate

As the OP is using OpenSUSE, this will do the trick: zypper install logrotate

[BOAC]

Linux IS vulnerable

'Fix' due out shortly to close down a bug in the OS that enables someone to attain superuser rights on a system.

Just when you thought you were safe.....................

[mad_jock]

Any link to what the bug is?