and even though everything works, you get spammed with errors like.

# lvdisplay
Couldn’t find device with uuid ‘dEiMau-Kytn-mEb8-FXpW-MCTQ-yEAf-tFIWmD’.
Couldn’t find all physical volumes for volume group data.
Couldn’t find device with uuid ‘dEiMau-Kytn-mEb8-FXpW-MCTQ-yEAf-tFIWmD’.
Couldn’t find all physical volumes for volume group data.
Volume group “data” not found

# pvscan
Couldn’t find device with uuid ‘dEiMau-Kytn-mEb8-FXpW-MCTQ-yEAf-tFIWmD’.
Couldn’t find device with uuid ‘p9fqIr-7F7O-TR3L-sODE-MMKp-3doT-yeNjZ6′.
Couldn’t find device with uuid ‘dEiMau-Kytn-mEb8-FXpW-MCTQ-yEAf-tFIWmD’.
Couldn’t find device with uuid ‘p9fqIr-7F7O-TR3L-sODE-MMKp-3doT-yeNjZ6′.

PV /dev/sda1        VG data   lvm2 [232.88 GB / 0    free]
PV unknown device   VG data   lvm2 [233.75 GB / 0    free]
PV unknown device   VG data   lvm2 [232.88 GB / 0    free]

most of us are lazy when it comes to even thinking about reading the man pages, using –help or simply checking out google. so one have might tried everything from reinstalling lvm, filtering out the disks in /etc/lvm configs to god knows what.

this simple command will check for missing disks and do everything for you. Just remember that this will screw you up if your planning to re-add your disks.

# vgreduce –removemissing data
Couldn’t find device with uuid ‘dEiMau-Kytn-mEb8-FXpW-MCTQ-yEAf-tFIWmD’.
Couldn’t find all physical volumes for volume group data.
Couldn’t find device with uuid ‘dEiMau-Kytn-mEb8-FXpW-MCTQ-yEAf-tFIWmD’.
Couldn’t find all physical volumes for volume group data.
Couldn’t find device with uuid ‘dEiMau-Kytn-mEb8-FXpW-MCTQ-yEAf-tFIWmD’.
Couldn’t find device with uuid ‘p9fqIr-7F7O-TR3L-sODE-MMKp-3doT-yeNjZ6′.
Couldn’t find device with uuid ‘dEiMau-Kytn-mEb8-FXpW-MCTQ-yEAf-tFIWmD’.
Couldn’t find device with uuid ‘p9fqIr-7F7O-TR3L-sODE-MMKp-3doT-yeNjZ6′.

Wrote out consistent volume group data
#

# pvscan
PV /dev/sda1   VG data   lvm2 [232.88 GB / 232.88 GB free]

however, after setting up various programs, i kept getting *** buffer overflow detected *** errors,

on different programs, however, only the ones i had compiled myself.

it seems like its an issue with the gcc 4.3 with FORTIFY. and disabling it worked for me.

(remember to make clean first)

export CFLAGS=-D_FORTIFY_SOURCE=0; ./configure

and then do make;make install again.

horay!

user@server:/# ftp localhost 12345
Connected to localhost.
421 Service not available, remote server has closed connection

(various errors can produce this message)

checking /glftpd/ftp-data/logs/error.log you might see something like

Tue Sep 23 12:40:20 2008 [7183Â Â Â ] Couldn’t canonicalize local hostname (1)

usually i would check the /etc/hosts but in this case, it was a typo i made in /etc/glftpd.conf

ifip 10.0.0.*
pasv_addr 10.0.0.2
elseip
pasv_addr some.fancy-dyndns.tld 1
endifip

the pasv_addr had a typo which made it impossible to do a lookup, and produced this error (thanks to glftpd for the fantastic precise detailed error message pointing me directly to glftpd.conf on line xxx)

and yes, im being sarcastic..

i had ensured that my exports list was correctly set up, actually, i had it tested from my tvix (was able to browse the share) but from another nix server, i got this message.

there’s probably alot of different solutions for this, since its anything else than accurate error description.

in my case, the other server i tried to access my server from, didnt have the nfs-common package installed.

after installing it, everything went just smooth..

what i forgot all about, was that of course OpenVPN was affected by this aswell, so after the update and a reboot, all my vpn connections died suddenly.

as far as i know, theres no other way to fix this, other than generating new keys, (ca, dh, all of it) since the openvpn server itself wont work anymore.

however, a temp solution until one gets time to generate new keys and certificates could be….

sudo mv /usr/sbin/openssl-vulnkey /usr/sbin/openssl-vulnkey.bak

$ sudo mv /usr/sbin/openvpn-vulnkey /usr/sbin/openvpn-vulnkey.bak

$ sudo ln -s /bin/true /usr/sbin/openssl-vulnkey

OpenVPN will now start with the naughty keys and you can at least update the keys.

usually i just upgrade the kernel, but hey, time for some cleanup!

i save the .config file for the running kernel, as lazy i am, no need to spend more time than necessary.

since i decided to make it my router, i had to compile iptables support. which i did, and compiled rest of the netfilter/iptables related stuff as modules. so the weird thing was, while i were trying out an old iptables rules script

i got an error like

iptables -t nat -A POSTROUTING -o ppp0 -j MASQUERADE

FATAL: Module ip_tables not found.

iptables v1.3.8: can’t initialize iptables table `nat’: Table does not exist (do you need to insmod?)
Perhaps iptables or your kernel needs to be upgraded.

First i reinstalled iptables which made no difference, then the kernel again (included a few useless modules)

and i still got the error.

anyway, it seems like it can be rather unhealthy to use your current kernel config everytime you compile a new kernel, since there are some options which arent updated(no questions asked)

anyway, i did a make menuconfig, removed the support for networking (Networking – [*]Networking Support)

save the config file, opened it again, and enabled the support. all the original settings were removed, and i had to config it from scratch, but to my big surprise, i suddenly had alot more options to play with. and after compiling the kernel, rebooting – the problem was fixed

but since i recently got fiber installed at my place, and my isp got a thing for PPPoE

and also got multiple access concentrators (servers you can connect to, providing redundancy) i got a small problem since i hooked up the link directly to my linux gw.

 the problem is that they only delivery client software for windows. usually its not a problem since there’s almost always a client for every other os, if not build in.

but the “smart” thing about this client is that you can select which server you want to connect to. currently my isp (fiberbredbaand.dk) got two servers in different locations.

a normal client (windows,linux,mac) and the one in a router, will automatically just jump on the first servers that answers you, so it can be random from time to time. which is the whole point.

 now, the problem occurs if the server got problems. from bandwidth issues to not even being able to browse websites, or similar stuff. your normal client will most likely still use that server, and your fucked.

 now, if we take a look at your current config file, probably located at

/etc/ppp/peers/dsl-provider

there should be a uncommented like looking like this

pty “/usr/sbin/pppoe -I eth1 -T 80 -m 1412″

mtu size and interface might different.

doing a man pppoe will reveal

-C ac_name
Specifies the desired access concentrator name. pppoe will only initiate sessions with the specified access concentrator. In most cases,
you should not specify this option. Use it only if you know that there are multiple access concentrators. If both the -S and -C options
are specified, they must both match for pppoe to initiate a session.

so adding -C ser-sil01 to the line, so it looks like

pty “/usr/sbin/pppoe -I eth1 -T 80 -m 1412 -C ser-sil01″

and then doing a poff && pon dsl-provider

should force it to connect to the server ser-sil01 (another choice might be ser-ski01in my case) remember this is only 2 servers which is available at my isp, in your case its something else.

anyway, remember that if the selected server dies completely, you need to either remove the -C option or use another server, since it wont shift to the other server automatically

VMware Server Console

The Process exited with an error:

End of error message.

And tailing the log file might output stuff similar to this

foo# tail -f /var/log/vmware/vmware-serverd.log

Nov 05 20:34:38: app| Attempting to launch vmx : /var/lib/vmware/Virtual Machines/Other Linux 2.6.x kernel/Other Linux 2.6.x kernel.vmx
Nov 05 20:34:38: app| Error during launch: 11, The process exited with an error:
Nov 05 20:34:38: app| End of error message
Nov 05 20:34:38: app| VmsdVmStatePendingCmdFailed: cmd status is already set
Nov 05 20:34:41: app| Msg_Post: Error
Nov 05 20:34:41: app| [msg.vmmonPosix.badVersion] Version mismatch with vmmon module: expecting 138.0, got 137.0.
Nov 05 20:34:41: app| [msg.vmmonPosix.badDriver] You have an incorrect version of the `vmmon’ kernel module.
Nov 05 20:34:41: app| Try reinstalling VMware Server.
Nov 05 20:34:41: app| [localized] Version mismatch with vmmon module: expecting 138.0, got 137.0.
Nov 05 20:34:41: app| You have an incorrect version of the `vmmon’ kernel module.
Nov 05 20:34:41: app| Try reinstalling VMware Server.
Nov 05 20:34:41: app| —————————————-
Nov 05 20:34:41: app| Msg_Post: Version mismatch with vmmon module: expecting 138.0, got 137.0.
Nov 05 20:34:41: app| You have an incorrect version of the `vmmon’ kernel module.
Nov 05 20:34:41: app| Try reinstalling VMware Server.
Nov 05 20:34:41: app|
Nov 05 20:34:41: app| Msg_Post: Error
Nov 05 20:34:41: app| [msg.vmmonPosix.initFailed] Failed to initialize monitor device.
Nov 05 20:34:41: app| [localized] Failed to initialize monitor device.
Nov 05 20:34:41: app| —————————————-
Nov 05 20:34:41: app| Msg_Post: Failed to initialize monitor device.

yea..its actually true, just reinstalling the vmware server solves the problem (no need to uninstall it first) and its pretty quick since you can skip the networking configuration (its been saved)

so in order to make it stop, i decided to try and help him with vmware server installation which he fucked up (can’t prove it, but all the evidence points to him)

Anyway, while trying to install it, one might encounter an error message like this

What is the location of the directory of C header files that match your running
kernel? [/lib/modules/2.6.23.1-foo/build/include]

Extracting the sources of the vmmon module.

Building the vmmon module.

Using 2.6.x kernel build system.
make: Entering directory `/tmp/vmware-config0/vmmon-only’
make -C /lib/modules/2.6.23.1-dewey/build/include/.. SUBDIRS=$PWD SRCROOT=$PWD/. modules
make[1]: Entering directory `/usr/src/linux-2.6.23.1′
CC [M] /tmp/vmware-config0/vmmon-only/linux/driver.o
/tmp/vmware-config0/vmmon-only/linux/driver.c: In function âLinuxDriver_Ioctlâ:
/tmp/vmware-config0/vmmon-only/linux/driver.c:1659: error: âstruct mm_structâ has no member named âdumpableâ
make[2]: *** [/tmp/vmware-config0/vmmon-only/linux/driver.o] Error 1
make[1]: *** [_module_/tmp/vmware-config0/vmmon-only] Error 2
make[1]: Leaving directory `/usr/src/linux-2.6.23.1′
make: *** [vmmon.ko] Error 2
make: Leaving directory `/tmp/vmware-config0/vmmon-only’
Unable to build the vmmon module.

For more information on how to troubleshoot module-related problems, please
visit our Web site at “http://www.vmware.com/download/modules/modules.html” and
“http://www.vmware.com/support/reference/linux/prebuilt_modules_linux.html”.

Execution aborted.

i’d like to point out a few things that you would most likely want to install to make the installation go smooth.

apt-get install libx11-6 libx11-dev libxtst6 libxtst-dev libxt6 libxt-dev wget linux-headers-`uname -r` build-essential make gcc g++

i might be missing some, or some of them might not be needed, but I’m too careless to check it out, so feel free to bug me if you see something wrong.

IF it whines about the same error again, the cause of this might be that your running a newer kernel than the installer requires, if so, there’s a patch out which solves the problem. beware that when your reading this, the direct link to the patch i post, might already be outdated, so be sure to check if there’s a newer version out. currently when writing this, version 114 is out.

So lets do it.

foo:/usr/src# wget http://knihovny.cvut.cz/ftp/pub/vmware/vmware-any-any-update114.tar.gz

foo:/usr/src# tar zxfv vmware-any-any-update114.tar.gz

foo:/usr/src# cd vmware-any-any-update114

foo:/usr/src/vmware-any-any-update114# ./runme.pl

should spit out something smiliar to this

Updating /usr/bin/vmware-config.pl … already patched
Updating /usr/bin/vmware … No patch needed/available
Updating /usr/bin/vmnet-bridge … No patch needed/available
Updating /usr/lib/vmware/bin/vmware-vmx … No patch needed/available
Updating /usr/lib/vmware/bin-debug/vmware-vmx … No patch needed/available
VMware modules in “/usr/lib/vmware/modules/source” has been updated.

and from there, the installation should succeed

 At least the client has improved over the years since i last used it, which must have been 7 years ago or something. anyway, my memory isn’t what it used to b…hmm, nevermind. my memory more or less always fails to remember anything at all, so of course i ran into problems again. here’s some output from /var/log/syslog after trying to start pppoe.

 (i actually did compile a new kernel with support for PPP, but i forgot something…)

Oct 26 23:39:16 foo pppd[3069]: pppd 2.4.4 started by root, uid 0
Oct 26 23:39:16 foo pppd[3069]: Serial connection established.
Oct 26 23:39:16 foo pppd[3069]: Couldn’t set tty to PPP discipline: Invalid argument
Oct 26 23:39:16 foo pppoe[3071]: PADS: Service-Name: ”
Oct 26 23:39:16 foo pppoe[3071]: PPP session is 41216 (0xa100)
Oct 26 23:39:16 foo pppoe[3071]: read (asyncReadFromPPP): Session 41216: Input/output error
Oct 26 23:39:16 foo pppoe[3071]: Sent PADT
Oct 26 23:39:46 foo pppd[3069]: Serial connection established.

In short,i forgot to compile support for “PPP support for async serial ports”