Open Source Web Hosting

Trying to set up a BD-J development environment on my windows 7, 64-bit system, but when I run the ME Platform SDK 3.0, I get the subject error (java.net.ConnectException: Connection refused: connect)
So after googling around, stockoverflow to the rescue (see the top comment by Kevin).

After I saw that, I installed 32-bit Java from here, and then updated the paths in the 2 files named “java” (in installDir\bin and installDir\toolbar\bin) – I updated both; replaced the path with the path to the newly installed 32-bit JDK and then it worked.

I’ve run into this a couple times now. An old server is decommissioned, BB starts sending mad pages. So I’ll remove the host’s entry from BBHOME/etc/bb-hosts

But it’s still paging me non-stop with purple alerts! What the heck?

This software is ancient but luckily some folks out there still have good answers to this question! Turns out there are 3 other folders with files that need to be removed since they pertain to the host.

<I’ve removed a host in etc/bb-hosts and the screen is purple>

After you remove a host(s) in etc/bb-hosts, you must remove the
corresponding files in $BBLOGS, www/html and $BBHIST. BBLOGS and
BBHIST are defined in etc/bbinc.sh

cd www
rm html/thedeletedhost*
cd ../../bbvar
rm logs/thedeletedhost*
rm hist/thedeletedhost*

Solved! How to get past ‘tree connect failed: NT_STATUS_ACCESS_DENIED’?

I set up a new samba server and one of our windows users was unable to connect, even though 2 others were able to connect.

I was going crazy trying to figure this out, eventually the solution was to log in as that user once via ssh, then he was able to connect with samba!

the error message that was revealed by testing the samba login on the server with
# smbclient //HOSTNAME/FOLDER
Password:
Anonymous login successful
Domain=[HOMENET] OS=[Unix] Server=[Samba 3.0.2-7.FC1]
tree connect failed: NT_STATUS_BAD_NETWORK_NAME

While working on setting up some old scripts to process some forms, I was hitting the dreaded Internal Server Error. The apache log said:

suexec policy violation: see suexec log for more details, referer:
Premature end of script headers: submit_comments.pl, referer:


I read up on suexec at the apache site and found there are 20 rules that must be met in order for suexec to work. The script implementation must be violating (at least!) one of them, but how to find which one? I tried to “see the suexec log” as the error stated but anything I found that looked like a place for the suexec log was either missing, outdated, or empty 🙁

After looking around in frustration for this mythical suexec log, I looked at the man page for suexec (rtfm) and found there is a command to show the compilation options of suexec:

suexec -V (must be run as root!)

# suexec -V
-D AP_DOC_ROOT="/var/htdocs"
-D AP_GID_MIN=100
-D AP_HTTPD_USER="nobody"
-D AP_LOG_EXEC="/usr/local/apache2/logs/suexec_log"
-D AP_SAFE_PATH="/usr/local/bin:/usr/bin:/bin"
-D AP_UID_MIN=100
-D AP_USERDIR_SUFFIX="public"


Now I was able to find the real suexec log in this out-of-the-way location, and looking in there I quickly found the cause of my suffering:

[2011-09-25 18:11:59]: target uid/gid (816/816) mismatch with directory (408/408) or program (816/816)
[2011-09-25 18:16:21]: uid: (816/webguy) gid: (816/816) cmd: submit_comments.pl


After seeing this, I realized the problem is that the folder containing the script doesn’t match the ownership and group of the suexec user, so I looked in the virtual host config file to find the user & group specified with ‘SuexecUserGroup’. Then used chown to set the user and group of the script and the folder containing it to match the suexec user and group. After that the script worked without an error!

Tried twice and it hangs on LVM2 at 17%
I wonder if it’s a setting or a compatibility issue, or i just need to be more patient and let it sit more than an hour?

Well I tried a few more times. Letting it sit did not help. It seems that if it got stuck there for one second, it would be stuck there for one day. When it finally worked, it blew right past that portion of the install – near to the very end!

One thing that helped was choosing to format the disk with VMWare (the first option) rather than LVM2 (the second option). Doing it that way, it only got stuck one other time and I was able to successfully install dapper (6.06) with HDs of varying capacity from 10GB up to over 300GB.

Now the next problem is that the dapper update repos seem to be totally gone, nowhere to be found. Time to move on I guess…

Cool site for checking your password security, https://www.grc.com/haystack.htm

I have an Altima 2005 2.5, 4 cylinder – changing the plugs was actually really easy – except that i don’t have a rubber-lined socket to pull the plugs out (and neither did my local hw store)
So after needle nose pliers (too big), chopsticks (got some lift but not enough grab to get them out), contemplated chewing gum…. came up with the magic solution.
I used a thick paper clip and I’m sure a wire coat hanger would also work. using pliers, twist one end into a loop slightly bigger than the metal top of the spark plug. Shine a light down there, loop the wire over the top, tip it to the side and yank that sucker out. if its not grabbing, change the loop angle – mine ended up about 30 degrees to the vertical shaft of the rest of the paper clip.
Obviously you have to unscrew the plug first, with a 16mm deep socket on an extension arm. Good luck!

roku: Brightscript & XML
boxee app: python & rss
bluray app: java, BD-J

stuff I’d like to learn & do next:
haproxy on pfsense
mySQL -> noSQL migration?
Ruby

OK why is there even any debate? I just tried rsync’ing a directory with a couple hundred gigs to a USB drive, I had a few gigabytes that weren’t on the USB yet. After several hours there didn’t seem to be any progress. I set up rdist to copy to a different directory (/mnt/usb) on the localhost, and the copy is completed within 5 minutes!

Despite this, when I google around, I find comments going back a decade declaring rsync to be the superior copying tool. hmmm. hours = incomplete vs minutes = success; RDIST winning.

PS here is a great reference for using rdist

on Ubuntu, when configuring apache2 with-ssl, I received the subject error.
the fix: apt-get install libssl-dev