Category: Technology
Matraex Upgraded Mail Client From Squirrelmail to Roundcube
Matraex Upgraded Mail Client From Squirrelmail to Roundcube
Matraex has officially upgraded our web based mail client from Squirrelmail to Roundcube.
Roundcube is a modern mail client utilizing newer technologies for faster and more feature rich mail interaction. Roundcube runs on our Linux webservers, utilizing Apache, PHP and MySQL. The software connects to the mail server using the IMAP protocol.
All address book contacts and preferences were imported to Roundcube from Squirellmail at the time of the transition.
As well as updating and implementing their own technologies, Matraex provides server administration, open source production implementation and software customizations to business as a service.
Users with questions about the new mail service or Matraex Consulting Services should contact:
Michael Blood
Matraex, Inc
208.344.1115
www.matraex.com
Fixed ActiveX error: Automation Error: the object invoked has disconnected from its clients
Fixed ActiveX error: Automation Error: the object invoked has disconnected from its clients
We have a custom ActiveX Applet we built which was randomly throwing this error, it was happening on a new machine only and it was rather random. Automation Error: the object invoked has disconnected from its clients
Sometimes the error did not happen at all. The new machine had a fresh install of XP on it. Lots of other new XP machines were not running into the problem.
We researched and found that the problem always happened during some sort of usage with the “MSINET.OCX” activex object. This led us to an article on Microsoft’s support site with this article.
http://support.microsoft.com/kb/254908
Turns out the new machine was dual core which somehow caused problems with threading of the object, the problem was fixed on a later version of Visual Studio so if I redeployed the application with the latest MSINET.OCX object it should have worked. As a quick fix though I downloaded a later version MSINET.OCX to the machine and reregistered it so that the system would use that object.
regsvr32 /u msinet.ocx
regsvr32 msinet.ocx
Voila, no more random error.
Network Boot Server with Linux Install, Debian Etch and Lenny, CentOS and KNOPPIX
Network Boot Server with Linux Install, Debian Etch and Lenny, CentOS and KNOPPIX
I just LOVE my dedicated PXE boot server at the office with several flavors of linux install on it.
I can bring a new server online with a base install in as few as five minutes with Debian or CentOS
I can debug workstations and servers with a quickbooting KNOPPIX install.
I even have some kernel installations customized to install network drivers for the Dell 2650 so that the installs I do for those are quick and simple. (basically the broadcom network drivers and the openssh-server packages are preseeded to be installed with the default package)
Here are the contents my pxelinux.cfg/default file:
DISPLAY boot.txt
#DEFAULT etch_i386_install
LABEL etch_i386_install
kernel debian/etch/i386/linux
append vga=normal initrd=debian/etch/i386/initrd.gz —
LABEL etch_i386_expert
kernel debian/etch/i386/linux
append priority=low vga=normal initrd=debian/etch/i386/initrd.gz —
LABEL etch_i386_rescue
kernel debian/etch/i386/linux
append vga=normal initrd=debian/etch/i386/initrd.gz rescue/enable=true —
LABEL knoppix
kernel knoppix/vmlinuz
append secure myconfig=scan nfsdir=192.168.0.1:/srv/diskless/knoppix nodhcp lang=us ramdisk_size=100000 init=/etc/init apm=p
ower-off nomce vga=791 initrd=knoppix/miniroot.gz quiet BOOT_IMAGE=knoppix
LABEL centos5_install
kernel centos/5/vmlinuz
append ks=nfs:192.168.0.1:/srv/diskless/centos/5/ks_prompt.cfg initrd=centos/5/initrd.img ramdisk_size=100000 ksdevice=eth0
ip=dhcp url –url http://mirror.centos.org/centos/5/os/i386/CentOS/
LABEL centos5_raid_install_noprompt
kernel centos/5/vmlinuz
append ks=nfs:192.168.0.1:/srv/diskless/centos/5/ks_raid.cfg initrd=centos/5/initrd.img ramdisk_size=100000 ksdevice=eth0 ip
=dhcp url –url http://mirror.centos.org/centos/5/os/i386/CentOS/
LABEL centos5_hda_install_noprompt
kernel centos/5/vmlinuz
append ks=nfs:192.168.0.1:/srv/diskless/centos/5/ks_hda.cfg initrd=centos/5/initrd.img ramdisk_size=100000 ksdevice=eth0 ip=
dhcp url –url http://mirror.centos.org/centos/5/os/i386/CentOS/
LABEL centos5_install_noprompt
kernel centos/5/vmlinuz
append ks=nfs:192.168.0.1:/srv/diskless/centos/5/ks.cfg initrd=centos/5/initrd.img ramdisk_size=100000 ksdevice=eth0 ip=dhcp
url –url http://mirror.centos.org/centos/5/os/i386/CentOS/[dfads params=’groups=221&limit=1′]
LABEL lenny_i386_install
kernel debian/lenny/i386/linux
append vga=normal initrd=debian/lenny/i386/initrd.gz —LABEL lenny_amd64_install
kernel debian/lenny/amd64/linux
append vga=normal initrd=debian/lenny/amd64/initrd.gz —LABEL etch_amd64_install
kernel debian/etch/amd64/linux
append vga=normal initrd=debian/etch/amd64/initrd.gz —LABEL etch_amd64_linux
kernel debian/etch/amd64/linux
append vga=normal initrd=debian/etch/amd64/initrd.gz —LABEL etch_amd64_expert
kernel debian/etch/amd64/linux
append priority=low vga=normal initrd=debian/etch/amd64/initrd.gz —LABEL etch_amd64_rescue
kernel debian/etch/amd64/linux
append vga=normal initrd=debian/etch/amd64/initrd.gz rescue/enable=true —LABEL etch_amd64_auto
kernel debian/etch/amd64/linux
append auto=true priority=critical vga=normal initrd=debian/etch/amd64/initrd.gz —PROMPT 1
Here are the contents of my boot.txt file (so that I know what to type at the command line when booting)
– Boot Menu –
=============etch_i386_install  –  Debian Stable
etch_i386_expert   –  Debian Stable (Shows install menu every step)
etch_i386_rescue   –  Debian Stable Rescue
lenny_i386_install — has Broadcom net card customization
lenny_amd64_install — has Broadcom net card customization
etch_amd64_install
etch_amd64_linux
etch_amd64_expert
etch_amd64_rescue
etch_amd64_auto
centos5_install –Â CentOS 5 (Will prompt for disks)
centos5_install_noprompt –Â CentOS 5 (Will auto install without prompts)
centos5_hda_install_noprompt –Â CentOS 5 (Will auto install without prompts)
centos5_raid_install_noprompt –Â CentOS 5 (Will auto install on raid 1 without prompts)
knoppix
Hope someone out there can find some use from this.
We of course can help people having trouble with their own TFTP and PXE Boot Server .
Installed PERC management software afaapps and created simple mirror
Installed PERC management software afaapps and created simple mirror
I just installed Debian Lenny on a Dell 2650 with an OLD PERC 3 RAID controller.
I then installed the afaapps package from Dell’s website (http://support.us.dell.com/support/downloads/download.aspx?c=us&l=en&s=gen&releaseid=R85529&formatcnt=1&libid=0&fileid=112003)
Use this link or just search for ‘afaapps’ under the Drivers and Downloads section of the Dell support site.
After extracting the rpm from the downloaded file I ran alien against the file to turn it into a debian file
#apt-get install alien
#alien -d –scripts afaapps-2.8-0.i386.rpm
Now just install the created debian package
#dpkg -i afaapps_2.8-1_i386.deb
Now that you have installed the afacli you can run it at the command line prompt which will open the PERC command line “FASTCMD>”
Then you’ll open / connect to the RAID controller using “open afa0”
#afacli
FASTCMD> open afa0
Executing: open “afa0”
A simple ‘disk list’ command to find out what your disk situation looks like
AFA0> disk list
Executing: disk list
B:ID:L Device Type Blocks Bytes/Block Usage Shared
—— ————– ——— ———– —————- ——
0:00:0 Disk 35566478Â 512 Initialized NO
0:01:0 Disk 287132440 512 Initialized NO
0:02:0 Disk 287132440 512 Initialized NO
you may have to initialize your disks by typeing ‘disk initialize 1’ and ‘disk initialize 2’ to make sure that the container can access them, you can see in my example above that my two disks are already initialized.
Now I will create a volume on disk 1 and mirror that disk to disk 2
AFA0> container create volume 1
AFA0> container create mirror 1 2
At the bottom of your screen you should see the status of the mirroring Job, something like.
Stat:OK!, Task:100, Func:MSC Ctr:1, State:RUN 16.2%
Once the job completes you can partition and format the disk. Check the label on the disk by running:
AFA0> container list
Executing: container list
Num Total Oth Chunk Scsi Partition
Label Type Size Ctr Size Usage B:ID:L Offset:Size
—– —— —— — —— ——- —— ————-
0 xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx1 Mirror 136GB Valid 0:01:0 64.0KB: 136GB
/dev/sdb 0:04:0 64.0KB: 136GB
From this I can see that I will need to partition and format disk “/dev/sdb”
Have fun! And if I can help you on it let me know.
Print jobs on Remote Desktop wont delete from print queue
Print jobs on Remote Desktop wont delete from print queue
A remote user was logging into their server to do some work from afar.
When they went to print their work some sort of connection problem occurred and when they reconnected they had two local printers in their “Printers and Faxes” window with different sessions, one with a print job (from the disconnected session) and one with no print jobs (from the current session).
When attempting to print again the job would not print and they would disconnect and try again only to find that they now have 3 printers (2 from disconnected session). This went on for a while they tried to delete the print jobs from the queue but the job never deleted, it just stayed in the status of “Deleting” or “Deleting – Paused”. Eventually they ended up with 6 of these printers each with one print job (I am sure they could have put more jobs in a print queue if they really wanted to).
This is the point at which I came in to clear up the issue.
I knew we had to clear out the print queue or that printer would never work.
Here are the steps we had to go through in order to fix it,
- Stop the “Print Spooler” Service. You can do this two ways
- From the Command Prompt (Start -> Run -> Type ‘cmd’ -> Click ‘OK’)
This opens a Command Line Interface, type “net stop spooler” - Open the Services Console (Start -> Control Panel -> Administrative Tools -> Services)
Select “Print Spooler” from the list of Services by double clicking and then click “Stop”
- From the Command Prompt (Start -> Run -> Type ‘cmd’ -> Click ‘OK’)
- Delete all print jobs. You can also do this two ways
- In the open Command Prompt window type del c:windowssystem32spoolprinters*
Type ‘y’ if it asks you if you are sure - In your Windows Explorer window goto C:windowssystem32spoolprinters
Delete all files
- In the open Command Prompt window type del c:windowssystem32spoolprinters*
- Restart the print spooler. Either:
- In the open Command Prompt window type “net start spooler”
- Click the “Start” button on the open “Print Spooler” Service Properties window
There you have it. For those of you that know how you can create a batch file which will quickly do all of this
net stop spooler
sleep for 5
del /Q /F c:windowssystem32spoolprinters*
net start spooler
The “/Q” switch tells the “del” command to delete the files without prompting if it is “ok to delete on global wildcard”
The “/F” switch tells the “del” command to delete the files even if they are marked read only
To get this batch file to work you will need to get the “sleep” program to pause for a few seconds after the service stops so that the system can release the lock it has on the files. If you do not have the sleep program you may find that running the batch file the first time errors out saying it can not access the file. But run it a second time and it works.
Cygwin Not Connecting via SSH
Cygwin Not Connecting via SSH
I have a scheduled task that is setup to run a command with Cygwin each night to backup files to a Linux machine. I would find Bash and SSH processes left running but the backup didn’t start. This is because I needed permissions to the /cygwin/etc/group and /cygwin/etc/passwd files setup correctly for the user account that’s running the backup.
This is what the task is set to run “c:cygwinbinbash.exe -l -c /home/backup/rbackup_redirect”
To troubleshoot I would schedule the backup for a few minutes ahead. It would fail running under the backup user account. I tried it on an admin account and the backup account, which isn’t an admin, by manually starting it in Cygwin. The admin account would work. The backup account wouldn’t. I then watched the SSH traffic on the Linux server as I manually started the backup. I saw that it would attempt to connect but fail when started from the backup account. It would hang for some time not moving on to the next step then stop trying. On the windows system Bash and SSH would still be running but on the Linux system SSH would stop running.
When I start Cygwin under the backup user account it would give a message saying that I may not be in the right group and I may need to rebuild /cygwin/etc/group and /cygwin/etc/passwd but it would go on to bash and let me run commands like normal. I tried rebuilding them. That didn’t work. I checked the file permissions on those files and found the issue. Just adding the backup user account didn’t work though. I had to actually delete the files. Make new ones that were blank. Then add the backup user account to have full rights. Then I rebuilt the files & SSH connections work fine.