A.Adverse's questions

Ubuntu 16.04LTS -> 18.04LTS upgrade failure

Yes it was a long in the tooth install. But it finally got a release upgrade. Since then I haven't been able to get apt to install anything much looking for dosemu but has no targets, and update spews errors about InRelease not containing bionic.

W: Skipping acquire of configured file 'bionic/cnf/Commands-amd64' as repository 'http://archive.ubuntu.com/ubuntu focal InRelease' doesn't have the component '

#------------------------------------------------------------------------------#
#                            OFFICIAL UBUNTU REPOS                             #
#------------------------------------------------------------------------------#


###### Ubuntu Main Repos
deb http://au.archive.ubuntu.com/ubuntu/ focal bionic main restricted universe multiverse
deb-src http://au.archive.ubuntu.com/ubuntu/ focal bionic main restricted universe

###### Ubuntu Partner Repo
deb http://archive.canonical.com/ubuntu focal bionic partner
deb-src http://archive.canonical.com/ubuntu focal bionic partner
deb http://archive.ubuntu.com/ubuntu focal bionic main # auto generated by ubuntu-release-upgrader

deb http://in.archive.ubuntu.com/ubuntu/ focal-updates main restricted universe
deb-src http://in.archive.ubuntu.com/ubuntu/ focal-updates main restricted universe

deb http://in.archive.ubuntu.com/ubuntu/ focal-security main restricted universe
deb-src http://in.archive.ubuntu.com/ubuntu/ focal-security main restricted universe

deb http://in.archive.ubuntu.com/ubuntu/ focal-backports main restricted universe
deb-src http://in.archive.ubuntu.com/ubuntu/ focal-backports main restricted universe

The google-fu gives all sorts of strange answers, but nothing I can find fixes this issue.

Regards, A

I have used this setup on other systems before with no problem,so I don't understand it now, but I need an assist to sort it out.

TightVncServer with xfce4 and xfce4-goodies, launched from vncserver starts just fine, but if a window passes into the background I lose all interactivity with it. Current xstartup follows.

#!/bin/bash
xrdb $HOME/.Xresources
startxfce &

A

Is there a way to have fail2ban ban the entire class C network from which a ban address comes from?

A

I have an image created by a CGI script.. if you call the script directly it displays quite happily with.

Content-type: image/jpeg

However I need to embed this inside a page, and no matter what I try the browsers just display raw data. I've tried octet-stream, binary and a few other things to no avail.

#!/bin/bash 
echo "Content-type: image/jpeg"
echo ""
/usr/bin/xearth -pos fixed/10/144.96 -shade -size 320/320 -gif |convert gif: jpeg:-

.

<!--#include file="head.shtml" -->
<!--#exec cgi="/cgi-bin/xearth.cgi" -->
<!--#include file="tail.shtml" -->

A

I know I know, don't put images in the DB. But I keep small thumbnails of book covers in a a small DB. Now this worked fine up until some moron decided to steal my db and ask for bitcoin.

Ignoring the request for bitcoin, I got out the back up and restored it. So far so good, data is intact searches run beautifully. Now when I finally got around to adding a new item, I find it will not let add images. It throws a wobbly with an error

Notice in ./tbl_replace.php#246
Undefined variable: message

The field is called "cover" and its a blob, with the default value of null, but even the null doesn't appear when you add an entry.

I see the field in the sql query is blank where the image files name ought to be, but that's all I know. Can anyone help me solve this one?

I've recently upgraded my install of VirtualBox to 5.2.28, and while trying to reconfigure the VM's I find data entered into any field turns into gibberish. For arguments sake, I type in 23 and the field becomes 0- which of course isn't valid. The rest of the keyboard is similarly affected.

VirtualBox is the only thing affected by this problem. The original 14.04LTS install worked without a problem, and upgrade to 5.2.x broke it. The default install for 18.04LTS is also broken the same way at 5.2.18

Never seen anything like it before, any ideas?

A

In case this helps identify the current mapping, qwerty comes out as 4c.gvn

EDIT: Well its still going on, not only that, but a fresh install of 18LTS actually gives me the same result.

I've had this working before, but I can't get it right on a new machine. I have the following setup on a ubuntu install as a router.

 192.168.2.0/24 p34p1 wifi1 -------------|
 192.168.1.0/24 p33p1 wired -------------|------Internet p32p1 
 192.168.3.0/24 p34p1 wifi2 -------------|      192.168.0.0/24     

This might look overly complicated but it keeps the kids gaming traffic off my regular network, which can't use it on a console anyway. At present all three segments can talk to the internet, but cannot talk to each other in different segments as required.

routing

 Kernel IP Routing table
 Destination     Gateway       Genmask         Iface
 0.0.0.0         192.168.0.1   0.0.0.0         p32p1
 192.168.0.0     *             255.255.255.0   p32p1
 192.168.0.1     *             255.255.255.255 p32p1
 192.168.1.0     *             255.255.255.0   p33p1
 192.168.2.0     *             255.255.255.0   p34p1
 192.168.3.0     *             255.255.255.0   p35p1    

iptables -t nat -v -x -n -L

Chain PREROUTING (policy ACCEPT 998 packets, 64328 bytes)
pkts      bytes target     prot opt in     out     source               destination

Chain INPUT (policy ACCEPT 115 packets, 9157 bytes)
pkts      bytes target     prot opt in     out     source               destination

Chain OUTPUT (policy ACCEPT 30271 packets, 1701429 bytes)
pkts      bytes target     prot opt in     out     source               destination

Chain POSTROUTING (policy ACCEPT 172 packets, 17314 bytes)
pkts      bytes target     prot opt in     out     source               destination
30982  1739286 MASQUERADE  all  --  *      p32p1   0.0.0.0/0            0.0.0.0/0
  102     8264            all  --  *      p33p1   0.0.0.0/0            0.0.0.0/0
   61     7327            all  --  *      p34p1   0.0.0.0/0            0.0.0.0/0
    7     1603            all  --  *      p35p1   0.0.0.0/0            0.0.0.0/0

my iptables rules are turning into mush.

Any Assistance appreciated, thank-you

I'm running Virtual box to run some MS-DOS machines, I find if I tag them all to launch at startup Virtual Box seems to trip over its feet, i end up having to kill it, restart, and launch the VM's manually.

So I need to either postpone all the launches until VB is fully up and running, or stagger the start of individual VM's to one at a time maybe.

A

Is there a mechanic to let me limit SSH and telnet logins to the root account based on IP address? I have 3 nets used for different things locally all in the 192.168.x ranges, I need to login from those and banned from anywhere else. I know having root active is frowned on... but like a bad admin I live in my root account.

This is only an issue because some antique software has Y2K issues, I need to move the file creation dates from 1990's to 2000 but for files across 21 subdirectories. A way to manage it recursively is what I'm looking for.

This ends up as an NFS share back to an MSDOS client and it appears to be using the creation date. Not the modified date, as I've tried the touch mod already with no success.

I have a strange process on Ubuntu 14.04 that fires up fairly frequently. It appears as a bash shell in top, I haven't actually located it in ps xfa.

When it runs, top says it's using 215% CPU and my load avg shoots up to 9. There's nothing in the root crontab and I haven't been able to locate anything else. Suggestions?

Found it is PS it has this

23266 ?   Sl   106:29 sh    -a cryptonight -o stratum+tcp://0x6BBF635F:3333 -u 47EaZKYrX8NBBCE4N1H5zhXn3FR3tFmGYSBQWPPt2Vg8GmCwiGuwKdMbXKHTCCKpAUjU9AJ6Se4PU7G7uDLDoE4h3dtjkck -p x -t 3

That doesn't help me much locate it, or stop it...

#*/30 * * * * /var/tmp/". "/c

was active in a user crontab, looks normal except for the ". " any idea what that does?

I don't know whats happening here, if I cp -Rfv /a /b the file names in the destination are suddenly getting mangled to 8.3 is gibberish characters. This only started after I installed NFS and I do have old dos clients accessing it, but I don't understand why cp is mangling names that already comply to 8.3 even and not in exported directories. Copy single files seems to work more reliably.

On further investigation this appears to be a DOS client issue. Only files in lowercase on the server appear correctly to the client, regardless of 8.3 qualifications. No idea how you fix that, I can't find anything substantive on NFS filename mangling.

TIA, A

Anyone know how to stop xhost running at login for cli logins?

all it gives is "xhost: unable to open display """ error because theres no Xserver running on this system.

Regards, A

Everything I can find with "google" says delete account, I need the account, but I need to prevent it being able to send mail. The account is public and doesn't need mail at all.

Ideally I need it not to have access to mail, alternatively I need to drop all mail to/from this account.

A

One of these things is making my HD appear full while running. Virtual Box is running 6 msdos boxes, and vnc is only there to serve virtualbox. Over time it persistantly fills the HD to 100% in a roughly 24hr period. Its not the /tmp directory it remains pretty empty.

HELP!!!

this is normal..15% usage, I've just had to restart so I'll post a full hd soon

9.3M    /bin
280K    /home
300M    /lib
16M     /etc
37M     /boot
13M     /sbin
4.0K    /mnt
6.4G    /root
16K     /lost+found
4.0K    /usb
20K     /tmp
4.0K    /opt
2.4G    /usr
928M    /var
4.0K    /srv
10G     /

Its full again... du is having a hard time. I see VirtualBox has filled /root with logs again.

9.3M    /bin
496K    /home
304M    /lib
16M     /etc
37M     /boot
13M     /sbin
4.0K    /mnt
65G     /root 
16K     /lost+found
4.0K    /usb
20K     /tmp
4.0K    /opt
2.4G    /usr
1.1G    /var
4.0K    /srv
69G     /

I delete all the logs, and they are massive. I get..

9.3M    /bin
496K    /home
304M    /lib
16M     /etc
37M     /boot
13M     /sbin
4.0K    /mnt
514M    /root
16K     /lost+found
4.0K    /usb
20K     /tmp
4.0K    /opt
2.4G    /usr
1.1G    /var
4.0K    /srv
4.4G    /

but df says 100% usage still.

/dev/sda1        71960912   71944524          4 100% /

I'm looking for a way to create a daily internet limit for my Ubuntu Server. All outbound traffic leaves Eth2, a usb dongle, this is prepaid and the kids eat it quickly. I need to divide my prepaid amount usually 12Gb into daily allotments and stop traffic for the day after this amount is reached. Maybe with a web page thrown in saying daily limit exceeded.

Preferably something from the CLI. It's a headless beast with only SSH access.

VNSTAT seems to do what I need, I just don't have the scripting skills to have it drive an ifdown command.

Thanks.